Aiming at the problems that are poor generalization performance, low retrieval accuracy and large time consumption of existing content-based image retrieval system, the hierarchical image retrieval method based on multi feature fusion is proposed in this paper. The retrieval accuracy rates on Corel5K, UKbeach and Holidays are 68.23(Top 1), 3.73(N-S) and 88.20(mAp), respectively. The experimental results show that the method proposed in this paper can effectively improve the deficiency of single feature retrieval and save time significantly in the premise of a small amount of loss of accuracy.

Pedestrian multi-target tracking technology plays an important role in artificial intelligence, driverless, virtual reality and other fields. The pedestrian multi-target tracking algorithm DeepSORT based on detection is widely used in industry. It mainly tracks multiple pedestrian targets continuously and keeps their ID unchanged. In order to improve the applicability and tracking accuracy of DeepSORT algorithm, this paper improved the IOU distance measurement in the matching process. At the same time, ResNet50 is used as the feature extraction backbone network, and combined with FPN (Feature Pyramid Network), the appearance features of multi-layer pedestrians are fused to improve the tracking accuracy of DeepSORT algorithm. The proposed algorithm is verified on the public data set MOT-16 and it’s tracking accuracy is enhanced to 4.1%.

The study put forward a data fusion approach for urban remote sensing that combines SAR (Synthetic Aperture Radar) and optical satellite data. By integrating datasets from different sensors and spatial-temporal scales, the technique aims to extract more accurate information. The fusion approach utilizes two methods: feature-based fusion, where relevant features are extracted and fused, and simple layer stacking (SLS), where the original datasets are directly stacked as multiple layers. This study extracted features using SAR textures (using Sentinel-1) and modified indices (using Landsat-8), and then classified these features using an XGBoost algorithm implemented in Python and Google Earth Engine. Researchers examined five cities, each representing a distinct climatic zone and urban dynamic: Cape Town, Guangzhou, Los Angeles, Mumbai, and Osaka. An accuracy assessment was conducted using random validation points, achieving an overall accuracy of 89.5% using the proposed MSFI method. A comparison was also performed with three well-known global products. The proposed approach, outperformed all three global products achived 89% accuracy while ESA (84%), ESRI (81%) and Dynamic World (82%). Additionally, Land surface temperature analysis was accomplished to investigate the relationship between extracted UIS and Land Surface Temperature (LST) across selected cities to show the practical use of proposed MSFI method. Los Angeles, a warm temperate city, showed the highest LST among all five cities. The datasets, along with the GEE and Python codes, are available at https://github.com/mnasarahmad/sls.

Laser-targeted weeding methods further enhance the sustainable development of green agriculture, with one key technology being the improvement of weed localization accuracy. Here, we propose an improved YOLOv8 instance segmentation based on bidirectional feature fusion and deformable convolution (BFFDC-YOLOv8-seg) to address the challenges of insufficient weed localization accuracy in complex environments with resource-limited laser weeding devices. Initially, by training on extensive datasets of plant images, the most appropriate model scale and training weights are determined, facilitating the development of a lightweight network. Subsequently, the introduction of the Bidirectional Feature Pyramid Network (BiFPN) during feature fusion effectively prevents the omission of weeds. Lastly, the use of Dynamic Snake Convolution (DSConv) to replace some convolutional kernels enhances flexibility, benefiting the segmentation of weeds with elongated stems and irregular edges. Experimental results indicate that the BFFDC-YOLOv8-seg model achieves a 4.9% increase in precision, an 8.1% increase in recall rate, and a 2.8% increase in mAP50 value to 98.8% on a vegetable weed dataset compared to the original model. It also shows improved mAP50 over other typical segmentation models such as Mask R-CNN, YOLOv5-seg, and YOLOv7-seg by 10.8%, 13.4%, and 1.8%, respectively. Furthermore, the model achieves a detection speed of 24.8 FPS on the Jetson Orin nano standalone device, with a model size of 6.8MB that balances between size and accuracy. The model meets the requirements for real-time precise weed segmentation, suitable for complex vegetable field environments and resource-limited laser weeding devices.

High-power laser facilities necessitate predicting incremental damage to final optics to identify evolving damage trends. In this study, we propose a surface damage detection method utilizing image segmentation employing ResNet-18, and a damage area estimation network employing U-Net++. Paired sets of online and offline images of optics obtained from a large laser facility is used to train the network. The trends of varying damage could be identified by incorporating additional experimental parameters. A key advantage of the proposed method is that the network can be trained end-to-end on small samples, eliminating the need for manual labeling or feature extraction. The software developed based on these models can facilitate the daily inspection and maintenance of optics in large laser facilities.

In response to reducing the energy cost of unmanned surface vehicles (USVs) while overcoming the low accuracy problem in surface target detection, a lightweight detection algorithm with multi-scale feature fusion is proposed. Based on the popular one-stage lightweight Yolov7-tiny target detection model, a lightweight extraction module is designed first by introducing the multiscale residual module to reduce the number of parameters and computational complexity while improving accuracy. The Mish and SiLU activation functions are used to enhance network feature extraction. Second, the path aggregation network employs coordinate convolution to strengthen spatial information perception. Finally, the dynamic head, which is based on the at-tention mechanism, improves the representation ability of object detection heads without any computational overhead. According to the experimental findings, the proposed model has 22.1% fewer parameters than the original model, 15% fewer GFLOPs, a 6.2% improvement in [email protected], a 4.3% rise in [email protected]:0.95, and it satisfies the real-time criteria. According to the research, the suggested lightweight water surface detection approach includes a lighter model, a simpler computational architecture, more accuracy, and a wide range of generalizability. It performs bet-ter in a variety of difficult water surface circumstances.

Gastric cancer is the fifth most common and fourth deadliest cancer worldwide, with a bleak 5-year survival rate of about 20%. Despite significant research into its pathobiology, prognostic predictability remains insufficient due to pathologists’ heavy workloads and the potential for diagnostic errors. Consequently, there is a pressing need for automated and precise histopathological diagnostic tools. This study leverages Machine Learning and Deep Learning techniques to classify histopathological images into healthy and cancerous categories. By utilizing both handcrafted and deep features and shallow learning classifiers on the GasHisSDB dataset, we conduct a comparative analysis to identify the most effective combinations of features and classifiers for differentiating normal from abnormal histopathological images without employing fine-tuning strategies. Our methodology achieves an accuracy of 95% with the SVM classifier, underscoring the effectiveness of feature fusion strategies. Additionally, cross-magnification experiments produced promising results with accuracies close to 80% and 90% when testing the models on unseen testing images with different resolutions.

Hyperspectral images (HSI) contain abundant spectral information. Efficient extraction and utilization of this information for image classification remain prominent research topics. Previously, hyperspectral classification techniques primarily relied on statistical attributes and mathematical models of spectral data. Deep learning classification techniques have recently been extensively utilized for hyperspectral data classification, yielding promising outcomes. This study proposes a deep learning approach that uses polarization feature maps for classification. Initially, the polar coordinate transformation method is employed to convert the spectral information of all pixels in the image into spectral feature maps. Subsequently, the proposed Deep Context Feature Fusion Network (DCFF-NET) is utilized to classify these feature maps. The model is validated using three open-source hyperspectral datasets: Indian Pines, Pavia University, and Salinas. The experimental results indicated that DCFF-NET achieves excellent classification performance. Experimental results on three public HSI datasets demonstrate that the proposed method accurately recognizes different objects with an overall accuracy (OA) of 86.68%, 94.73%, and 95.14% based on the pixel method, and 98.15%, 99.86%, and 99.98% based on the pixel-patch method.

The rational design of magnetic carbon composites, encompassing both composition and microstructure, holds significant potential for achieving exceptional electromagnetic wave absorbing materials (EAMs). In this study, FeCo@CM composites was efficiently fabricated through an advanced microwave plasma assisted reduction chemical vapor deposition (MPARCVD) technique, offering high efficiency, low cost, and energy-saving benefits. By depositing graphitized carbon microspheres, the dielectric properties were significantly enhanced, resulting in improved electromagnetic wave absorption performances through optimized impedance matching and a synergistic effect with magnetic loss. A systematic investigation revealed that the laminar-stacked structure of FeCo exhibited superior properties compared to its spherical counterpart, supplying a higher number of exposed edges and enhanced catalytic activity, which facilitated the deposition of uniform and low-defect graphitized carbon microspheres. Consequently, the dielectric loss performance of the FeCo@CM composites was dramatically improved due to increased electrical conductivity and the formation of abundant heterogeneous interfaces. At a 40 wt% filling amount and a frequency of 7.84 GHz, the FeCo@CM composites achieved a minimum reflection loss value of -58.2 dB with an effective absorption bandwidth (fE) of 5.13 GHz. This study presents an effective strategy for developing high-performance EAMs.

Polarimetric Synthetic Aperture Radar (PolSAR) images encompass valuable information that can facilitate extensive land cover interpretation and generate diverse output products. Extracting meaningful features from PolSAR data poses challenges distinct from those encountered in optical imagery. Deep Learning (DL) methods offer effective solutions for overcoming these challenges in PolSAR feature extraction. Convolutional Neural Networks (CNNs) play a crucial role in capturing PolSAR image characteristics by leveraging kernel capabilities to consider local information and the complex-valued nature of PolSAR data. In this study, a novel three-branch fusion of complex-valued CNN named Shallow to Deep Feature Fusion Network (SDF2Net) is proposed for PolSAR image classification. To validate the performance of the proposed method, classification results are compared against multiple state-of-the-art approaches using the Airborne Synthetic Aperture Radar (AIRSAR) datasets of Flevoland, San Francisco, and ESAR Oberpfaffenhofen dataset. The results indicate that the proposed approach demonstrates improvements in OA, with 1.3% and 0.8% enhancement for the AIRSAR datasets and 0.5% improvement for the ESAR dataset. The analyses conducted on the Flevoland data underscore the effectiveness of the SDF2Net model, revealing a promising OA of 96.01% even with only 1% sampling ratio.

The multi-stage, multi-level assembly job shop scheduling problem (MsMlAJSP) is commonly encountered in the manufacturing of complex customized products. Ensuring production efficiency while effectively improving energy utilization is a key focus in the industry. For the energy-efficient MsMlAJSP (EEMsMlAJSP), an improved imperialist competitive algorithm based on Q-learning (IICA-QL) is proposed to minimize the maximum completion time and total energy consumption. In IICA-QL, a decoding strategy with energy-efficient triggers based on problem characteristics is designed to ensure solution quality while effectively enhancing search efficiency. Additionally, an assimilation operation with operator parameter self-adaptation based on Q-learning is devised to overcome the challenge of balancing exploration and exploitation with fixed parameters, thus the convergence and diversity of the algorithmic search is enhanced. Finally, the effectiveness of the energy-efficient strategy decoding trigger mechanism and the operator parameter self-adaptation operation based on Q-learning is demonstrated through experimental results, and the effectiveness of IICA-QL for solving EEMsMlAJSP is verified by comparing with other algorithms.

This paper proposes a new approach for multi-focus images fusion based on Region Mosaicing on Contrast Pyramids (REMCP). A density-based region growing method is developed to construct a focused region mask for multi-focus images. The segmented focused region mask is decomposed into a mask pyramid, which is then used for supervised region mosaicking on a contrast pyramid. In this way, the focus measurement and the continuity of focused regions are incorporated and the pixel level pyramid fusion is improved at the region level. Objective and subjective experiments show that the proposed REMCP is more robust to noise than compared algorithms and can fully preserves the focus information of the multi-focus images meanwhile reducing distortions of the fused images.

Hyperspectral anomaly detection is to recognize anomalies from complex scene in an unsupervised way. Currently, many spectral-spatial detection methods have been proposed with a cascaded manner. However, they often neglect complementary characteristics between spectral and spatial dimensions, which easily leads to yield high false alarm rate. To alleviate this issue, a spectral-spatial information fusion (SSIF) method is designed for hyperspectral anomaly detection. First, an isolation forest is exploited to obtain spectral anomaly map, in which the object-level feature is constructed with entropy rate segmentation algorithm. Then, a local spatial saliency detection scheme is proposed to produce spatial anomaly result. Finally, the spectral and spatial anomaly scores are integrated together followed by a domain transform recursive filtering to generate the final detection result. Experiments on five hyperspectral datasets prove that the proposed SSIF produces superior detection result over other state-of-the-art detection techniques.

Proton Exchange Membrane Fuel Cell (PEMFC) is a fast-developing battery technology, and the key to its reliability and lifespan improvement lies in the accurate assessment of durability. However, the degradation mechanism of PEMFC is hard to determine and its internal parameters are highly coupled. Thus, the development of the more accurate life prediction model that meets the actual scenarios is urgent to investigate. To solve this problem, a multi-feature fusion life prediction method based on Temporal Convolutional Network-Gated Recurrent Unit (TCN-GRU) is proposed. A TCN algorithm is used as the prediction base model, and two GRU modules are included to the model to strengthen the model expression ability and improve its predictive accuracy. Two widely recognized datasets and two operating conditions are utilized for model training and prediction, respectively. Comparisons are made with single-feature parameter models in terms of Root Mean Square Error (RMSE) and Determination Coefficient (R2). The results show that the prediction accuracy of the TCN-GRU multi-feature fusion model is higher than that of the single-feature models in terms of stability and anti-interference under both operating conditions. Furthermore, with the increase of input feature parameters, the TCN-GRU model is closer to the real value, which proves once again that the proposed model can meet the accuracy requirements of the life prediction of PEMFC.

Object detection algorithms for open water aerial images present challenges such as small object size, unsatisfactory detection accuracy, numerous network parameters, and enormous computational demands. Current detection algorithms struggle to meet the accuracy and speed requirements while being deployable on small mobile devices. This paper proposes DFLM-YOLO, a lightweight small-object detection network based on the YOLOv8 algorithm with multiscale feature fusion. Firstly, to solve the class imbalance problem of the SeaDroneSee dataset, we propose a data augmentation algorithm called Small Object Multiplication (SOM). SOM enhance dataset balance by increasing the number of objects in specific categories, thereby improving model accuracy and generalization capabilities. Secondly, we optimize the backbone network structure by implementing Depthwise Separable Convolution (DSConv) and the newly designed FC-C2f, which reduces the model's parameters and inference time. Finally, we design the Lightweight Multiscale Feature Fusion Network (LMFN) to address the challenges of multiscale variations by gradually fusing the four feature layers extracted from the backbone network in three stages. In addition, LMFN incorporates the Dilated Re-param Block structure to increase the effective receptive field and improve the model's classification ability and detection accuracy. Experimental results on the SeaDroneSea dataset indicate that DFLM-YOLO improves mAP by 12.4% compared to the original YOLOv8s, while reducing parameters by 67.2%. This achievement provides a new solution for UAVs to conduct object detection missions in open water efficiently.

Ship detection from synthetic aperture radar (SAR) imagery is crucial for various fields in real-world applications. Numerous deep learning-based detectors have been investigated for SAR ship detection, which require a substantial amount of labeled data for training. However, SAR data annotation is time-consuming and demands specialized expertise, resulting in that deep learning-based SAR ship detectors struggle due to a lack of annotations. With limited labeled data, semi-supervised learning is a popular approach for boosting detection performance by excavating valuable information from unlabeled data. In this paper, a semi-supervised SAR ship detection network is proposed, termed as Multi-Teacher Dempster-Shafer Evidence Fusion Net-work (MTDSEFN). The MTDSEFN is an enhanced framework based on the basic teacher-student skeleton frame, comprising two branches: the Teacher Group (TG) and the Agency Teacher (AT). The TG utilizes multiple teachers to generate pseudo-labels for different augmentation versions of unlabeled samples, which are then refined to obtain high-quality pseudo-labels by using Dempster-Shafer (D-S) fusion. The AT not only serves to deliver weights of its own teacher to the TG at the end of each epoch, but also updates its own weights after each iteration, enabling the model to effectively learn rich information from unlabeled data. The combination of TG and AT guarantees both reliable pseudo-label generation and learning comprehensive diversity information from numerous unlabeled samples. Extensive experiments were performed on two public SAR ship datasets, and the results demonstrated the effectiveness and superiority of the proposed approach.

In this work, a systematic study was conducted on the fabrication of multi-material components obtained employing Laser-Powder Bed Fusion (L-PBF) technology. The idea of making multi-material components is a winning capability of additive technologies because it allows the fabrication of Functionally Graded Materials (FGMs) with the customization of parts according to different required properties. The transition from one material to another was achieved gradually and continuously within the same layer using ad-hoc equipment designed for L-PBF systems with a powder spreading technique based on coaters or rollers. The influence of the relative position of the different materials within the powder chamber and the geometry of the developed equipment on the metallurgical and mechanical properties of the manufactured samples was evaluated. The performed tests involved the use of two materials, a nickel-based superalloy, and a stainless steel, having different chemical, physical and mechanical properties to obtain gradual properties variation in the manufactured samples. Based on the results of post-process characterization obtained through metallographic, chemical, and mechanical analysis, the relative positions of the materials and the geometry of the developed equipment have a limited effect on the sample’s manufactured properties. The characteristics of the FGM zone depend on the nature of the employed powders, and its extent coincides with that defined during the design of the divider.

The joint extraction of entities and relations is a critical task in information extraction, and its performance directly affects the performance of downstream tasks. However, existing joint extraction models based on deep learning exhibit weak processing capabilities for the phenomenon of multiple pronunciations of one character and multiple characters of one pronunciation when processing Chinese texts, resulting in performance loss. To address these issues, this paper introduces part-of-speech (POS) and pinyin features to aid the model in learning semantic features that are more contextually appropriate. A Chinese Joint Entity and Relation Extraction Model based on Progressive Contrastive Learning and Heterogeneous Feature Fusion is proposed (CJE-PCHF). During model training, an interactive fusion network based on progressive contrastive learning is employed to learn the dependencies between pinyin, POS, and semantic features. This guides the model in heterogeneous feature fusion, capturing higher-order semantic associations between heterogeneous features. On the commonly used DuIE evaluation dataset for joint extraction, this model achieved a significant improvement, with the F1 score increasing by 5.4% compared to the benchmark model CasRel.

Face verification for unrestricted faces in the wild is a challenging task. This paper proposes a method based on two deep convolutional neural networks(CNN) for face verification. In this work, we explore to use identification signal to supervise one CNN and the combination of semi-verification and identification to train the other one. In order to estimate semi-verification loss at a low computation cost, a circle, which is composed of all faces, is used for selecting face pairs from pairwise samples. In the process of face normalization, we propose to use different landmarks of faces to solve the problems caused by poses. And the final face representation is formed by the concatenating feature of each deep CNN after PCA reduction. What's more, each feature is a combination of multi-scale representations through making use of auxiliary classifiers. For the final verification, we only adopt the face representation of one region and one resolution of a face jointing Joint Bayesian classifier. Experiments show that our method can extract effective face representation with a small training dataset and our algorithm achieves 99.71% verification accuracy on LFW dataset.

The study aimed to analyze thoracic and lumbar erector spinae muscle activations during 3D elongation exercises in individuals with adolescent idiopathic scoliosis (AIS). Participants included 24 adolescents, meeting specific criteria such as a Cobb angle between 10° and 20°, and not having undergone surgical or brace treatments. EMG data were collected to evaluate muscle activations. The study found no significant interaction between exercise type and the side of the curve, but a significant main effect of exercise type on muscle activation, irrespective of Cobb angle. Different exercise types significantly affected muscle activity, with the highest activations in the TheraBand exercise, followed by manual and without equipment conditions. This suggests that exercise type significantly impacts muscle engagement in AIS patients, providing valuable insights for targeted exercise program design. The study, however, focused on short-term muscle activation, not long-term Cobb angle progression, pointing to the need for further research in this area.

AbstractBackground: In Ethiopia, 8.6% of maternal deaths are attributed to abortion. Despite the legalization of abortion and improved accessibility of safe abortion services, almost six in 10 abortions in Ethiopia are performed outside of the health facility. Abortion considered as an indicator for various aspects of reproductive health and healthcare access within a particular region or country. Despite the significance of the problem, there is a limited understanding of the magnitude of abortion and the contributing factors in the study area. Therefore, this study aimed to assess the prevalence of abortion and determinant factors among reproductive-age women in the Sidama regional state of Ethiopia.Methods: A community-based cross-sectional study was conducted from October to November 11, 2022, among 1,130 women of reproductive age. We utilized a multi-stage sampling method to select eligible women. A structured and pretested questionnaire was utilized to collect the data using Open Data Kit (ODK) mobile application. Due to the hierarchical nature of the data, a multilevel logistic regression model was applied. Variables with p-values < 0.25 on bi-variable analysis were included in a multivariable regression model. In the multivariable regression model, variables with a p-value < 0.05 were declared as statistically significant factors of abortion. Results: Nearly one-fifth (19.5%) of the women had a history of abortion. From the individual level determinates; women with formal education (AOR=0.44; 95% CI: 0.24-0.79), and planned pregnancy (AOR=0.41; 95% CI: 0.28-0.61) had lower odds of abortion. While from the community related factors; urban residence increased the odds of abortion (AOR=2.25; 95% CI: 1.20-4.20) as compared to the rural residence.Conclusion: the prevalence of abortion among reproductive age women of Northern Zone of Sidama Region was high. Education and a planned pregnancy serve as protective factors for abortion, while the risk of abortion increased among women reside in urban area. Therefore, to lower the abortion rate, it is better to implement a comprehensive approach, focusing on education, family planning, and urban-specific intervention.

This paper applies a new artificial intelligence architecture, the Temporal Fusion Transformer (TFT), for the joint GDP forecasting of 25 OECD countries at different time horizons. This new attention-based architecture offers significant advantages over other deep learning methods. First, results are interpretable since the impact of each explanatory variable on each forecast can be calculated. Second, it allows to visualize persistent temporal patterns and to identify significant events and different regimes. Third, it provides quantile regressions and permits to train the model on multiple time series from different distributions. Results suggest that TFTs outperform regression models, especially in periods of turbulence such as the COVID-19 shock. Interesting economic interpretations are obtained depending on whether the country is domestic demand-led or export-led growth. In essence, TFT is revealed as a new tool that artificial intelligence provides to economists and policy makers, with enormous prospects for the future.

The arsRBC operon encodes a 3-protein arsenic resistance system. ArsR regulates the transcription of the operon, while ArsB and ArsC are involved in exporting trivalent arsenic and reducing pentavalent arsenic respectively. Previous research in Agrobacterium tumefaciens 5A demonstrated ArsR has regulatory control over a wide range of metal related proteins and metabolic pathways. We hypothesized that ArsR has broad regulatory control in other Gram-negative bacteria and set out to test this. Here, we use differential proteomics to investigate changes caused by the presence of the arsR gene in human microbiome-relevant Escherichia coli during arsenite (AsIII) exposure. We show that ArsR has broad-ranging impacts such as the expression of TCA cycle enzymes during AsIII stress. Additionally, we found that the Isc [Fe-S] cluster and molybdenum cofactor assembly proteins are upregulated regardless of the presence of ArsR under these same conditions. An important finding from this differential proteomics analysis was the identification of response mechanisms that were strain, ArsR, and arsenic specific, providing new clarity to this complex regulon. Given the widespread occurrence of the arsRBC operon, these findings should have broad applicability across microbial genera, including sensitive environments such as the human gastrointestinal tract.

Skin cancer is an exquisite disease globally nowadays. Because of the poor contrast and apparent resemblance between skin and lesions, automatic identification of skin cancer is complicated. The rate of human death can be massively reduced if melanoma skin cancer can be detected quickly using dermoscopy images. In this research, an anisotropic diffusion filtering method is used on dermoscopy images to remove multiplicative speckle noise and the fast-bounding box (FBB) method is applied to segment the skin cancer region. Furthermore, the paper consists of two feature extractor parts. One of the two features extractor parts is the hybrid feature extractor (HFE) part and another is the convolutional neural network VGG19 based CNN feature extractor part. The HFE portion combines three feature extraction approaches into a single fused feature vector: Histogram-Oriented Gradient (HOG), Local Binary Pattern (LBP), and Speed Up Robust Feature (SURF). The CNN method also is used to extract additional features from test and training datasets. This two-feature vector is fused to design the classification model. This classifier performs the classification of dermoscopy images whether it is melanoma or non-melanoma skin cancer. The proposed methodology is performed on two ordinary datasets and achieved the accuracy 99.85%, sensitivity 91.65%, and specificity 95.70%, which makes it more successful than previous machine learning algorithms.

Convolutional neural networks (CNNs) face challenges in capturing long-distance text correlations, and Bidirectional Long Short-Term Memory (BiLSTM) networks exhibit limited feature extraction capabilities for text classification of work order. To address the abovementioned problems, this work utilizes an ensemble learning approach to integrate model elements efficiently. This study presents a method for classifying work order texts using a hybrid neural network model called BERT-BiLSTM-CNN. First, use Bert for preprocessing to obtain text vector representations. Then, capture context and process sequence information through BiLSTM. Next, capture local features in the text through CNN. Finally, obtain classification results through Softmax. Through comparative analysis, the method of fusing these three models is superior to other hybrid neural network model architectures in multiple classification tasks. It has a significant effect on work order classification.

Abstract: Image aesthetic assessment (IAA) with neural attention has made significant progress due to its effectiveness in object recognition. Current studies have shown that the features learned by convolutional neural networks (CNN) at different learning stages indicate meaningful information. The shallow feature contains the low-level information of images and the deep feature perceives the image semantics and themes. Inspired by this, we propose a visual enhancement network with feature fusion (FF-VEN). It consists of two sub-modules, the visual enhancement module (VE module) and the shallow and deep feature fusion module (SDFF module). The former uses an adaptive filter in the spatial domain to simulate human eyes according to the region of interest (ROI) extracted by neural feedback. The latter not only takes out the shallow feature and the deep feature by transverse connection, but also uses a feature fusion unit (FFU) to fuse the pooled features together with the aim of information contribution maximization. Experiments on standard AVA dataset and Photo.net dataset show the effectiveness of FF-VEN.

Sensitivity to weather conditions, and specially to clouds, is a severe limiting factor to the use of optical remote sensing for Earth monitoring applications. A possible alternative, is to resort to weather-insensitive synthetic aperture radar (SAR) images. However, in many real-world applications, critical decisions are made based on some informative spectral features, such as water, vegetation or soil indices, which cannot be extracted from SAR images. In the absence of optical sources, these data must be estimated. The current practice is to perform linear interpolation between data available at temporally close time instants. In this work, we propose to estimate missing spectral features through data fusion and deep-learning. Several sources of information are taken into account - optical sequences, SAR sequences, DEM - so as to exploit both temporal and cross-sensor dependencies. Based on these data, and a tiny cloud-free fraction of the target image, a compact convolutional neural network (CNN) is trained to perform the desired estimation. To validate the proposed approach, we focus on the estimation of the normalized difference vegetation index (NDVI), using coupled Sentinel-1 and Sentinel-2 time-series acquired over an agricultural region of Burkina Faso from May to November 2016. Several fusion schemes are considered, causal and non-causal, single-sensor or joint-sensor, corresponding to different operating conditions. Experimental results are very promising, showing a significant gain over baselines methods according to all performance indicators.

Extracting entities from large volumes of chicken epidemic texts is crucial for knowledge sharing, integration, and application. However, Named Entity Recognition (NER) encounters significant challenges in this domain, particularly due to the prevalence of nested entities and domain-specific named entities, coupled with a scarcity of labeled data. To address these challenges, we compiled a corpus from 50 books on chicken diseases, covering 28 different disease types. Utilizing this corpus, we developed a nested NER model, MFGFF-BiLSTM-EGP. This model integrates the Multiple Fine-Grained Feature Fusion (MFGFF) module with a BiLSTM neural network and employs an Efficient Global Pointer (EGP) for predicts the entity location encoding. In MFGFF module we designed three encoders, character encoder, word encoder and sentence encoder, this design effectively captures fine-grained features and improves the recognition accuracy of nested entities. Experimental results show that the model performs robustly, with F1 scores of 91.98%, 73.32%, and 82.54% on the CDNER, CMeEE V2, and CLUENER datasets, respectively, outperforming other commonly used NER models. Specifically, on the CDNER dataset, the model achieved an F1 score of 79.68% for nested entity recognition. This research not only advances the development of knowledge graph and intelligent question-answering system for chicken diseases but also provides a viable solution for extracting disease information that can be applied to other livestock species.

In a hyperconnected society, computer networks play a pivotal role in all activities that permeate our daily lives. In the context of smart cities, theses networks play a pivotal role in enhancing urban efficiency and sustainability. The continuous flow of packets in these networks poses security challenges, from personal data leakage to service unavailability. Anomaly-based methods are commonly used to detect network attacks. In this context, Software-Defined Networking (SDN) emerges as a solution for anomaly detection in network traffic, facilitating comprehensive, real-time analyses and dynamic adaptation to changes. This study aims to present a systematic review of feature selection techniques and evaluate the effectiveness of attribute selection with a multi-class classifier in network anomaly detection. The objective of this paper is to inspire future research and identify trade-offs among these techniques for detecting Denial of Service (DoS) attacks. Based on the findings, the analysis of the entire dataset yields superior results in terms of precision (99%), but using the output of OneR, although resulting in a slight loss of precision compared to the complete dataset, presents the highest precision among other studied techniques, indicating a trade-off between precision and processing time efficiency.

Introduction: One of the key strategies to achieve sustainable development goal by reduce preventable maternal deaths below 70 per 100,000 is increasing awareness about major obstetrics danger sign. Mother's awareness of obstetric danger signs in developing country is relatively low regardless of attempts trough many local and national program. It is crucial to increase women’s knowledge of obstetric danger signs through education and awareness-raising efforts. Despite its importance to reduce maternal deaths, Studies are limited only to the Northern Ethiopia. Objective: This study aimed to assess prevalence of knowledge of obstetric danger signs and associated factors among women who delivered in the last 12 months in Sidama region, Ethiopia. Methods: We conducted a community-based cross-sectional study from October to November, 2022. A multi-stage sampling procedure was utilized to select study participants. A structured and pretested questionnaire was utilized to collect data. A multilevel mixed-effects negative binomial regression model was used to control effects of clusters and confounders.Results: Out of 1,140 study participants, only 1,130 study participants took part in the study making a response rate of 99.12%. The overall prevalence of knowledge of obstetrics danger sign was estimated as 22.3% (95%CI: 18.7, 25.9). Government employed women (AIRR = 1.37; 95% CI: 1.2, 1.6), women who had access to mass media (AIRR = 1.16; 95% CI: 1.08, 1.25), women who had received model family training (AIRR = 1.15; 95% CI: 1.1, 1.25), autonomous women (AIRR = 1.15; 95% CI: 1.04, 1.25), women who had faced health problems during pregnancy (APR = 1.21; 95% CI: 1.11, 1.32) and urban women (APR = 1.22; 95% CI: 1.09, 1.62) were factors associated positively with knowledge of obstetrics danger sign. Conclusion: The prevalence of knowledge of obstetrics danger sign is found to be lower compared with the national average. Residence, mass-media exposure, model family planning, problem faced during pregnancy and women autonomy are the main predictors of knowledge of obstetrics danger sign. Public health policy makers should enhance health literacy at for focused on rural community. Awareness creation should be improved through mass media. Health providers should educate women during maternity health services to empower women for decision making process.

Target detection technology plays a crucial role in fishery ecological monitoring, fishery diversity research, and intelligent aquaculture. Deep learning, with its distinct advantages, provides significant convenience to the fishery industry. However, it still faces various challenges in practical applications, such as significant differences in image species and image blurring. To address these issues, this study proposes a multi-scale, multi-level, and multi-stage cross-domain feature fusion model. In order to train the model more effectively, a new dataset called Fish52 (Multi-scene Fish dataset) was constructed, on which the model achieved an mAP of 82.57%. Furthermore, we compared prevalent one-stage and two-stage detection methods on the Lahatan (single-scene fish dataset) and Fish30 dataset and tested them on the F4k and FishNet dataset. The mAP of our proposed model on the Fish30, Lahatan, F4k, and FishNet datasets reaches 91.72%, 98.7%, 88.6%,and 81.5% respectively, outperforming existing mainstream models. Comprehensive empirical analysis indicates that our model possesses high generalization ability and has reached advanced performance levels. In this study, the depth of the model backbone is deepened, a novel neck structure is proposed, and anew module is embedded therein. To enhance the fusion ability of the model, a new attention mechanism module is introduced. In addition, in the adaptive decoupling detection header module, introducing classes with independent parameter sand regression adapters reduces interaction between different tasks. The proposed model can better monitor fishery resources and enhance aquaculture efficiency. It not only provides an effective approach for fish detection but also has certain reference significance for the identification of similar target sin other environments, and offers assistance for the construction of smart fisheries and digital fisheries.

Heart disease is a global health concern of paramount importance, causing a significant number of fatalities and disabilities. Precise and timely diagnosis of heart disease is pivotal in pre-venting adverse outcomes and improving patient well-being, thereby creating a growing demand for intelligent approaches to predict heart disease effectively. This paper introduces an Ensemble Heuristic-Metaheuristic Feature Fusion Learning (EHMFFL) algorithm for heart disease diagnosis. Within the EHMFFL algorithm, a diverse ensemble learning model is crafted, featuring different feature subsets for each heterogeneous base learner, including support vector machine, K-nearest neighbors, logistic regression, random forest, naive bayes, decision tree, and XGBoost. The primary objective is to identify the most pertinent features for each base learner, leveraging a combined heuristic-metaheuristic approach that integrates the heuristic knowledge of Pearson correlation coefficient with the metaheuristic-driven grey wolf optimizer. The second objective is to aggregate the decision outcomes of the various base learners through ensemble learning, aimed at constructing a robust prediction model. The performance of the EHMFFL algorithm is rigorously assessed using the Cleveland and Statlog datasets yielding remarkable results with an accuracy of 91.8% and 88.9%, respectively, surpassing state-of-the-art machine learning, ensemble learning, and feature selection techniques in heart disease diagnosis. These findings underscore the potential of the EHMFFL algorithm in enhancing diagnostic accuracy for heart disease and providing valuable support to clinicians in making more informed decisions regarding patient care.

The normal growth and development of skeletal muscle is essential for the health of the body. The regulation of skeletal muscle by intestinal microorganisms and their metabolites has been continuously revealed. Acetate is the predominant short-chain fatty acid synthesized by gut microbiota through the fermentation of dietary fiber, however, the underlying molecular mechanisms governing the interaction between acetate and skeletal muscle during rapid growth stage remain to be further elucidated; Here, specific pathogen-free (SPF) mice, germ-free (GF) mice, and germ-free mice supplemented with sodium acetate (GS) were used to evaluate the effects of acetate on skeletal muscle growth and development of young mice with gut microbiota deficiency; We found the concentration of serum acetate, body mass gain, succinate dehydrogenase activity and expression of myogenesis maker gene of skeletal muscle in the GS group was higher than GF group following sodium acetate supplement. Furthermore, the transcriptome analysis revealed that acetate activated biological processes regulating skeletal muscle growth and development in GF group that are inhibited due to gut microbiota deficiency. In vitro experiment, we proved that acetate up-regulated Gm16062 to promote skeletal muscle cell differentiation; Overall, our findings proved that acetate promote skeletal muscle growth and development in young mice via increase Gm16062 expression.

Epilepsy is a prevalent neurological disorder with considerable risks, including physical impairment and irreversible brain damage from seizures. Given these challenges, the urgency for prompt and accurate seizure detection cannot be overstated. Traditionally, experts have relied on manual EEG signal analyses for seizure detection, which is labor-intensive and prone to hu-man error. Recognizing this limitation, the rise of deep learning methods has been heralded as a promising avenue, offering more refined diagnostic precision. On the other hand, the prevailing challenge in many models is their constrained emphasis on specific domains, potentially diminishing their robustness and precision in complex real-world environments. This paper presents a novel model that seamlessly integrates the salient features from the time-frequency domain alongside pivotal statistical attributes derived from EEG signals. This fusion process involves the integration of essential statistics, including the mean, median, and variance, combined with the rich data from compressed time-frequency (CWT) images processed using auto-encoders. This multidimensional feature set provides a robust foundation for subsequent analytic steps. A long short-term memory (LSTM) network, meticulously optimized for the renowned Bonn Epilepsy dataset, was used to enhance the capability of the proposed model. Preliminary evaluations underscore the prowess of the proposed model: remarkable 100% accuracy in most of the binary classifications, impressive performance exceeding 95% for three-class and four-class challenges, and a commendable rate exceeding 93.5% for the five-class classification.

The rampant misuse of drones poses a serious threat to national security and human life. Currently, the CNN method has been widely used in drone detection. However, there is a challenge that traditional CNN cannot cope with, which is small drone targets often have reduced amplitude or even lost features in infrared images. This paper proposes a Progressive Feature Fusion Network (PFFNet) that gradually increases the response amplitude of the target in the deep network. The Feature Selection Model (FSM) is designed to improve the utilisation of the output coding graph and enhance the feature representation of the target in the network. A lightweight segmentation head is also designed to achieve progressive feature fusion with multi-layer outputs. Experimental results show that the proposed algorithm can achieve low duration and high accuracy in drone target detection. On the public dataset, the IoU is improved by 2.53% and the detection time is reduced by 81.03%.

Like traditional single label learning, multi-label learning is also faced with the problem of dimensional disaster.Feature selection is an effective technique for dimensionality reduction and learning efficiency improvement of high-dimensional data. In this paper, Logistic regression, manifold learning and sparse regularization were combined to construct a joint framework for multi-label feature selection (LMFS). Firstly, the sparsity of the eigenweight matrix is constrained by the $L_{2,1}$-norm. Secondly, the feature manifold and label manifold can constrain the feature weight matrix to make it fit the data information and label information better. An iterative updating algorithm is designed and the convergence of the algorithm is proved.Finally, the LMFS algorithm is compared with DRMFS, SCLS and other algorithms on eight classical multi-label data sets. The experimental results show the effectiveness of LMFS algorithm.

Cloud detection in remote sensing images is a crucial preprocessing step that efficiently identifies and extracts cloud-covered areas within the images, ensuring the precision and reliability of subsequent analyses and applications. Given the diversity of clouds and the intricacies of the surface, distinguishing the boundaries between thin clouds and the underlying surface is a major challenge in cloud detection. To address these challenges, an advanced cloud detection method, CloudformerV3, is presented in this paper. The proposed method employs a multi-scale adapter to incorporate dark and bright channel prior information into the model's backbone, enhancing the model's ability to capture prior information and multi-scale details from remote sensing images. Additionally, multi-level large window attention is utilized, enabling high-resolution feature maps and low-resolution feature maps to mutually focus and subsequently merge during the resolution recovery phase. This facilitates the establishment of connections between different levels of feature maps and offers comprehensive contextual information for the model's decoder. Experimental results on the GF1_WHU dataset demonstrate that the method introduced in this paper exhibits superior detection accuracy when compared to state-of-the-art cloud detection models. Furthermore, enhanced detection performance is achieved along cloud edges and with respect to thin clouds, showcasing the efficacy of the proposed method.

This study describes a novel method for grading pathological sections of gliomas. Our own integrated hyperspectral imaging system was employed to characterize 270 bands of cancerous tissue samples from microarray slides of gliomas. These samples were then classified according to the guidelines developed by the World Health Organization, which define the subtypes and grades of diffuse gliomas. We explored a hyperspectral feature extraction model called SMLME-Resnet using microscopic hyperspectral images of brain gliomas of different malignancy grades. The model combines the channel attention mechanism and multi-scale image features to automatically learn the pathological organization of gliomas and obtain hierarchical feature representations, effectively removing the interference of redundant information. It also completes multi-modal, multi-scale spatial and spectral feature extraction to improve the automatic classification of glioma subtypes. The proposed classification method demonstrated high average classification accuracy (>97.3\%) and Kappa coefficient (0.954), indicating its effectiveness in improving the automatic classification of hyperspectral gliomas. The method is readily applicable in a wide range of clinical settings, offering valuable assistance in alleviating the workload of clinical pathologists. Furthermore, the study contributes to the development of more personalized and refined treatment plans, as well as subsequent follow-up and adjustment of treatment, by providing physicians with insights into the underlying pathological organization of gliomas.

Use of a multi-sensor approach can provide citizens a holistic insight in the air quality in their immediate surroundings and assessment of personal exposure to urban stressors. Our work, as part of the ICARUS H2020 project, which included over 600 participants from 7 European cities, discusses data fusion and harmonization on a diverse set of multi-sensor data streams to provide a comprehensive and understandable report for participants, and offers possible solutions and improvements. Harmonizing the data streams identified issues with the used devices and protocols, such as non-uniform timestamps, data gaps, difficult data retrieval from commercial devices, and coarse activity data logging. Our process of data fusion and harmonization allowed us to automate the process of generating visualizations and reports and consequently provide each participant with a detailed individualized report. Results showed that a key solution was to streamline the code and speed up the process, which necessitated certain compromises in visualizing the data. A thought-out process of data fusion and harmonization on a diverse set of multi-sensor data streams considerably improved the quality and quantity of data that a research participant receives. Though automatization accelerated the production of the reports considerably, manual structured double checks are strongly recommended.

Saudi Arabian cities have experienced a rapid expansion of their urban areas - particularly their city centers - over the last four decades. This has resulted to an increase in vehicular usage, which has adversely affected the daily walking experience of the residents. Walkability has several positive effects on people's health as well as the environment in cities. In addition to being a means of transportation, it also helps people create a sense of place, enhancing their image ability to be easily understood as a legible urban structure, and deepen their emotional bond with the city. Using the medium-sized Saudi Arabian city of Onaizah as a case study, this study looks into the feasibility of creating urban walking corridors to encourage users to walk more. According to Saudi Arabia's Vision 2030, which emphasizes the value of sustainable urban development and raising quality of life for residents, the study addresses walkability as a generator to maintain and improve the urban scene of the city center. Geographic Information Systems (GIS) was used to analyze the data and generate the urban corridors in the city center. The results showed the possibility of improving walking in Onaizah through three urban corridors to the city center, measuring 1,401 meters, 1,283 meters, and 1,507 meters, respectively. These corridors represent urban, commercial and heritage corridors, giving an opportunity for planners and decision makers to challenge prioritize pedestrian connectivity and enhance the physical environment, leading to sustainable urban development. This research paper used various criteria analysis methods to provide a comprehensive assessment of various factors to conclude these urban corridors; for example, assessments of land use, transportation and environmental considerations. The study aligns with Saudi Arabia's vision 2030 by promoting walking and enhancing the city's overall walkability while offering Onaizah society by creating and suggesting urban corridors to the city center a sustainable and livable urban environment.

Water area segmentation in remote sensing is of great importance for flood monitoring. To overcome some challenges in this task, we construct the Water Index and Polarization Information (WIPI) multi-modality dataset and propose a multi-Modality Fusion and Gated multi-Filter U-Net (MFGF-UNet) convolutional neural network. The WIPI dataset enhances the water information and reduces the data dimension. In particular, the Cloud-Free label provided in the database effectively mitigates the issue of scarcity of labeled samples Since a single form or uniform kernel size cannot handle the variety in size and shape of water bodies, we proposed the gated multi-filter inception (GMF-Inception) module in our MFGF-UNet. Moreover, we utilize the attention mechanism by introducing the Gated Channel Transform (GCT) skip connection and integrating GCT into GMF- Inception to further improve model performance. Extensive experiments on three benchmarks, including WIPI, Chengdu and GF2020 datasets demonstrate that our method achieves favorable performance with lower complexity and better robustness against six competing approaches. For example, on WIPI, Chengdu and GF2020 datasets, the proposed MFGF-UNet model achieves F1 of 0.9191, 0.7410, and 0.8421, respectively, which average F1 on three datasets is 0.0045 higher than the U-Net model and reduces 62% average GFLOPs. New WIPI dataset, code and trained models have been released on GitHub

Multi-source image registration often suffered from great radiation and geometric differences. Specifically, gray scale and texture from similar landforms in different source images often show significantly different visual features. And these differences disturb the corresponding point extraction in the following image registration process. Considering that edges between heterogeneous images can provide homogeneous information and more consistent features can be extracted based on image edges, an edge consistency radiation-change insensitive feature transform (EC-RIFT) method is proposed in this paper. Firstly, the noise and texture interference are reduced by preprocessing according to the image characteristics. Secondly, image edges are extracted based on phase congruency, and an orthogonal Log-Gabor filter is performed to replace the global algorithm. Finally, the descriptors are built with logarithmic partition of the feature point neighborhood, which improves the robustness of descriptors. Comparative experiments on datasets containing multi-source remote sensing image pairs show that the proposed EC-RIFT method outperforms other registration methods in terms of precision and effectiveness.

The development of one technology can be portrayed by common methods like Gartner’s hype cycle or S-curve, however, there is no method to characterize the feature distributions of multiple technologies within a specific domain. This study proposes a big data-based method in terms of four proposed features, namely versatility, significance, commerciality, and disruptiveness, to characterize the technologies within a given domain. The features of technologies are quantitively portrayed using the representative keywords and volumes of returned search results from Google and Google Scholar in two-dimensional analytical spaces of technique and application. We demonstrate the applicability of this method using 452 technologies in the domain of intelligent robotics. The results of our assessment indicate that the versatility values are normally distributed, while the values of significance, commerciality, and disruptiveness follow power-law distributions, in which few technologies possess higher feature values. We also show that significant technologies are more likely to be commercialized or causing potential disruption, as such technologies have higher scores in these features. Further, we validly prove the robustness of our approach via comparing historical trends with literature and characterizing technologies in reduced analytical spaces. Our method can be widely applied in analyzing feature distributions of technologies in different domains, and it can potentially be exploited in decisions like investment, trade, and science policy.

A portable laser photoacoustic sensor based on a Field-Programmable Gate Array (FPGA) is reported for methane detection. A tunable DFB diode laser in the 1654 nm wavelength range is used as an excitation source. The photoacoustic signal processing was implemented by a FPGA device. A small resonant photoacoustic cell is designed. The minimum detection limit (1σ) of 10 ppm for methane (CH₄) is demonstrated.

Due to an increase in the number of image achieves, Content-Based Image Retrieval (CBIR) has gained attention for research community of computer vision. The image visual contents are represented in a feature space in the form of numerical values that is considered as a feature vector of image. Images belonging to different classes may contain the common visuals and shapes that can result in the closeness of computed feature space of two different images belonging to separate classes. Due to this reason, feature extraction and image representation is selected with appropriate features as it directly affects the performance of image retrieval system. The commonly used visual features are image spatial layout, color, texture and shape. Image feature space is combined to achieve the discriminating ability that is not possible to achieve when the features are used separately. Due to this reason, in this paper, we aim to explore the low-level feature combination that are based on color and shape features. We selected color moments and color histogram to represent color while shape is represented by using invariant moments. We selected this combination, as these features are reported intuitive, compact and robust for image representation. We evaluated the performance of our proposed research by using the Corel, Coil and Ground Truth (GT) image datasets. We evaluated the proposed low-level feature fusion by calculating the precision, recall and time required for feature extraction. The precision, recall and feature extraction values obtained from the proposed low-level feature fusion outperforms the existing research of CBIR.

With the rapid development of Intelligent Transportation Systems (ITS), vehicle type classification, as a key link in Automatic Toll Collection systems (ATC), is of great significance in improving traffic efficiency and reducing economic losses. This study proposes an intelligent vehicle classification system based on deep learning and multi-sensor data fusion to address the accuracy issues existing in vehicle classification methods based on optical sensors (OS) and human observers. The system significantly improves the accuracy and robustness of vehicle classification by combining deep Convolutional Neural Networks (CNN), LiDAR sensors, and machine learning algorithms. We first constructed a large-scale annotated dataset containing multiple vehicle types and complex traffic scenes to improve our model's capability to identify different vehicle characteristics. Next, CNN models based on different architectures were designed to extract global and local features of the vehicle, respectively. In addition, the LiDAR sensor was used to achieve the spatial structure architectures of the vehicle and combined with the output of the CNN model to improve classification performance under occlusion and complex scenes.

European strategic policy directions towards a sustainable blue economy have strengthened interest in maritime investments, thus increasing sectoral competition for marine space. Emerging repercussions out of such a rising interest need to be handled by Marine Spatial Planning (MSP) as a means for: properly allocating marine space to diverse uses; managing conflicts and promoting synergies among them; and pursuing a multi-use perspective of this space. A critical stage of each MSP exercise is the exploration of land-based spatial/developmental directions and their current or potential repercussions in the marine environment in order for constraints but also perspectives in policy choices/maritime uses in the MSP context to be featured; and attainment of a successful coherence or coordination between land and marine spatial/developmental policies to be ensured, by analyzing Land-Sea-Interaction (LSI) and embedding it in the MSP. The latter is the focus of this paper aiming to illuminate critical issues emerging from land-based (terrestrial) policy frameworks for feeding MSP endeavors; and is accomplished in the Northern Aegean Sea, Greece. Towards this end, a multilevel approach is adopted, delving into diverse policy contexts at the local, regional and national/European level and aiming to attain more informed MSP decisions in the specific study region.

The article is devoted to solving the problem of searching for universal explainable features that can remain explainable for a wide class of objects and phenomena and become an integral part of Explainable AI (XAI). The study is implemented on the example of an applied problem of early diagnostics of plant stress, using Thermal IR (TIR) and HSI, presented by 8 vegetation indices/channels. Each such index was presented by 5 statistical values. A Single-Layer-Perceptron classifier was used as the main instrument. TIR turned out to be the best of the indices in terms of efficiency in the field and sufficient to detect all 7 key days with 100% accuracy. Our study shows also that there are a number of indices, inluding NDVI, and usual color channels Red, Green, Blue, which are close to TIR possibilities in early plant stress detection with 100% accurasy or near, and can be used for wide class of plants and in different conditions their treatment. The stability of the stress classification in our study was maintained when the training set was reduced up to 10% of the dataset volume. The entropy-like feature of (max-min) for any indices/channels have determined as a leadersheep universal high-level explainable feature for the plant stress detection, which used in interaction with some of other statistical features.

The rapid urbanization in China has significantly contributed to the vast expansion of urban built-up areas. Precisely extracting and monitoring these areas is crucial for understanding and optimizing the developmental process and spatial attributes of smart, compact cities. However, most existing studies tend to focus narrowly on a single city or global scale with a single dimension, often ignoring mesoscale analysis across multiple urban agglomerations. In contrast, our study employs GIS and image processing techniques to integrate multi-source data for the identification of built-up areas. We specifically compare and analyze two representative urban agglomerations in China: the Yangtze River Delta (YRD) in the east and the Chengdu-Chongqing (CC) region in the west. We use different methods to extract built-up areas from socio-economic factors, natural surfaces, and traffic network dimensions. Additionally, we utilize a high-precision built-up area dataset of China as a reference for verification and comparison. Our findings reveal several significant insights: (1) The multi-source data fusion approach effectively enhances the extraction of built-up areas within urban agglomerations, achieving higher accuracy than previously employed methods. (2) Our research methodology performs particularly well in the CC urban agglomeration. The average precision rate in CC is 96.03%, while the average precision rate in YRD is lower at 80.33%. This study provides an objective and accurate assessment of distribution characteristics and internal spatial structure of built-up areas within urban agglomerations. This method offers a new perspective for identifying and monitoring built-up areas in Chinese urban agglomerations.

This research centres on advancing the utilisation of a novel Deep Ensemble model called RicEns-Net for predicting crop yields through sophisticated multi-modal data engineering. Initially, data for the study were acquired from Ernst \& Young (EY), a prominent global technology and management consulting firm, as part of their annual data science challenge (EY Open Science Challenge 2023). The predictive models leverage multi-modal data by integrating radar and optical-type remote sensing data alongside meteorological data to create supplementary features essential for predicting crop yield. The primary data features are derived from spectral indices extracted from Sentinel 2 satellites, while more advanced features are generated through combinations and interactions of these fundamental features. Furthermore, data from Sentinel 1 and 3, along with meteorological measurements such as surface temperature and rainfall, account for the intricate natural factors influencing crop yield, creating a comprehensive set of data features based on existing research. To tackle the complexity introduced by numerous data features, extensive data engineering was conducted to identify the most predictive features among 100+ and mitigate the "curse of dimensionality," resulting in 15 features from 5 different modalities. The proposed RicEns-Net comprises multiple machine learning algorithms in a deep ensemble approach to leverage the best features of each of these techniques, achieving optimal performance. The proposed model achieves an MAE value of 341 kg/Ha for prediction, surpassing the performance of the EY Open Science Challenge 2023 winner models and state-of-the-art deep/machine learning approaches.

Purposes: To provide abdominal contrast-enhanced MR image synthesis, we developed an image gradient regularized multi-modal multi-discrimination sparse-attention fusion generative adversarial network (GRMM-GAN) to avoid repeated contrast injections to patients and facilitate adaptive monitoring. Methods: With IRB approval, 165 abdominal MR studies from 61 liver cancer patients were retrospectively solicited from our institutional database. Each study included T2, T1 pre-contrast (T1pre), and T1 contrast-enhanced (T1ce) images. The GRMM-GAN synthesis pipeline consists of a sparse attention fusion network, an image gradient regularizer (GR), and a generative adversarial network with multi-discrimination. The studies were randomly divided into 115 for training, 20 for validation, and 30 for testing. The two pre-contrast MR modalities, T2 and T1pre images, were adopted as inputs in the training phase. The T1ce image at the portal venous phase was used as an output. The synthesized T1ce images were compared with the ground truth T1ce images. The evaluation metrics include peak signal-to-noise ratio (PSNR), structural-similarity-index (SSIM), and mean-squared-error (MSE). A Turing test and experts’ contours evaluated the image synthesis quality. Results: The proposed GRMM-GAN model achieved a PSNR of 28.56, an SSIM of 0.869, and an MSE of 83.27. The proposed model showed statistically significant improvements in all metrics tested with p-values <0.05 over the state-of-the-art model comparisons. The average Turing test score was 52.33%, close to random guessing, supporting the model's effectiveness for clinical application. In the tumor-specific region analysis, the average tumor contrast-to-noise ratio (CNR) of the synthesized MR images was not statistically significant from the real MR images. The average DICE from real vs. synthetic images was 0.90, compared to the inter-operator DICE of 0.91. Conclusion: We demonstrated the function of a novel multi-modal MR image synthesis neural network, GRMM-GAN, for T1ce MR synthesis based on pre-contrast T1 and T2 MR images. GRMM-GAN shows promise for avoiding repeated contrast injections during radiation therapy treatment.

This paper describes a proposed fractional filter-based multi-scale underwater and hazy image enhancement algorithm. The proposed system combines a modified global contrast operator with fractional order-based multi-scale filters used to generate several images, which are fused based on entropy and standard deviation. The multi-scale-global enhancement technique enables fully adaptive and controlled colour correction and contrast enhancement without over exposure of highlights when processing hazy and underwater images. This in addition to illumination/reflectance estimation coupled with global and local contrast enhancement. The proposed algorithm is also compared with the most recent available state-of-the-art multi-scale fusion de-hazing algorithm. Experimental comparisons indicate that the proposed approach yields better edge and contrast enhancement results without halo effect, colour degradation and is faster and more adaptive than all other algorithms from the literature.

Large-scale crowd phenomena are complex to model because the behaviour of pedestrians needs to be described at both strategic, tactical, and operational levels and is impacted by the density of the crowd. Microscopic models manage to mimic the dynamics at low densities, whereas mesoscopic models achieve better performances in dense situations. This paper proposes and evaluates a novel agent-based architecture to enable agents to dynamically change their operational model based on local density. The ability to combine microscopic and mesoscopic models for multi-scale simulation is studied through a use case of pedestrians at the Festival of Lights, Lyon, France. Pedestrian outflow data is extracted from video recordings of exiting crowds at the festival. The hybrid model is calibrated and validated using a genetic algorithm that optimizes the match between simulated and observed outflow data. Additionally, a local sensitivity analysis is then conducted to identify the most sensitive parameters in the model. Finally, the performance of the hybrid model is compared to different models in terms of density map and computation time. The results demonstrate that the hybrid model has the capacity to effectively simulate pedestrians across varied density scenarios while optimizing computational performance compared to other models.

Evolutionary convergence provides natural opportunities to investigate how, when, and why novel traits evolve. Many convergent traits are complex, highlighting the importance of explicitly considering convergence at different levels of biological organization, or ‘multi‐level convergent evolution’. To investigate multi‐level convergent evolution, we propose a holistic and hierarchical framework that emphasizes breaking down traits into several functional modules. We begin by identifying long‐standing questions on the origins of complexity and the diverse evolutionary processes underlying phenotypic convergence to discuss how they can be addressed by examining convergent systems. We argue that bioluminescence, a complex trait that evolved dozens of times through either novel mechanisms or conserved toolkits, is particularly well suited for these studies. We present an updated estimate of at least 94 independent origins of bioluminescence across the tree of life, which we calculated by reviewing and summarizing all estimates of independent origins. Then, we use our framework to review the biology, chemistry, and evolution of bioluminescence, and for each biological level identify questions that arise from our systematic review. We focus on luminous organisms that use the shared luciferin substrates coelenterazine or vargulin to produce light because these organisms convergently evolved bioluminescent proteins that use the same luciferins to produce bioluminescence. Evolutionary convergence does not necessarily extend across biological levels, as exemplified by cases of conservation and disparity in biological functions, organs, cells, and molecules associated with bioluminescence systems. Investigating differences across bioluminescent organisms will address fundamental questions on predictability and contingency in convergent evolution. Lastly, we highlight unexplored areas of bioluminescence research and advances in sequencing and chemical techniques useful for developing bioluminescence as a model system for studying multi‐level convergent evolution.

Blind deconvolution (BD) is one of the effective methods that help pre-process vibration signals and assist in bearing fault diagnosis. Currently, most BD methods design an optimization criterion and use frequency or time domain information independently to optimize a deconvolution filter. It recovers weak periodic impulses related to incipient faults. However, the random noise interference may cause the optimizer to overfit. The time-domain-based BD methods tend to extract fault-unrelated single peak impulse, and the frequency-domain-based BD methods tend to retain the maximum energy frequency component, which will lose the fault-related harmonics frequency components. To solve the above issue, we propose a hybrid criterion that combines the kurtosis for time domain optimization and the $G-l_1/l_2$ norm for the frequency domain. These two criteria are monotonically increasing and decreasing, so they mutually constrain to avoid overfitting. After that, we design a multi-task one-dimensional convolutional neural network with time and frequency branches to achieve an optimal solution for this hybrid criterion. The multi-task neural network realizes the simultaneous optimization of two domains. Experimental results show that our proposed method outperforms other state-of-the-art methods.

The paper proposes TOPSIS method for solving multi-objective multi-level programming problem (MO-MLPP) with fuzzy parameters via fuzzy goal programming (FGP). At first, - cut method is used to transform the fuzzily described MO-MLPP into deterministic MO-MLPP. Then, for specific , we construct the membership functions of distance functions from positive ideal solution (PIS) and negative ideal solution (NIS) of all level decision makers (DMs). Thereafter, FGP based multi-objective decision model is established for each level DM for obtaining individual optimal solution. A possible relaxation on decisions for all DMs is taken into account for satisfactory solution. Subsequently, two FGP models are developed and compromise optimal solutions are found by minimizing the sum of negative deviational variables. To recognize the better compromise optimal solution, the concept of distance functions is utilized. Finally, a novel algorithm for MO-MLPP involving fuzzy parameters is provided and an illustrative example is solved to verify the proposed procedure.

Forest fire identification is important for forest resource protection. Effective monitoring of forest fires requires the deployment of multiple monitors with different viewpoints, while most traditional recognition models can only recognize images from a single source. By ignoring the information from images with different viewpoints, these models produce high rates of missed and false alarms. In this paper, we propose a graph neural network model based on the similarity of dynamic features of multi-view images to improve the accuracy of forest fire recognition. The input features of the nodes on the graph are converted into relational features of different gallery pairs by establishing pairs (nodes) representing different viewpoint images and gallery images. The new feature library relationship is used to update the image gallery with dynamic features in order to achieve the estimation of similarity between images and improve the image recognition rate of the model. In addition, to reduce the complexity of image pre-processing process and extract key features in images effectively, this paper also proposes a dynamic feature extraction method for fire regions based on image segment ability. By setting the threshold value of HSV color space, the fire region is segmented from the image, and the dynamic features of successive frames of the fire region are extracted. The experimental results show that, compared with the baseline method Resnet, this paper's method is more effective in identifying forest fires, and its recognition accuracy is improved by 2%. And the scheme of this paper can adapt to different forest fire scenes, with better generalization ability and anti-interference ability.

Non-intrusive load monitoring (NILM) is a core technology for demand response (DR) and energy conservation services. Traditional NILM methods are rarely combined with practical applications, and most studies aim to decompose the whole loads in a household, which leads to low identification accuracy. In this paper, an NILM approach based on multi-feature integrated classification (MFIC) is explored, which combines some non-electrical features such as ON/OFF duration, usage frequency of appliances, and usage period to improve load differentiability. The implementation of MFIC algorithm is consistent with traditional event-based method. The uniqueness of our algorithm is that it designs an event detector based on steady-state segmentation and a linear discriminant classifier group based on multi-feature global similarity. Simulation results using an open-access dataset demonstrate the effectiveness and high accuracy of MFIC algorithm, with the state-of-the-art NILM methods as benchmarks.

A stable driver state is essential in the process of manually transition control, which inevitably occurs in Level 3 automated driving situations. To this end, this paper proposed a CNN algorithm-based driver state monitoring system that uses multi-sensor data such as driver's face image, biometric information, and vehicle behavior information as input. This system calculates the probability of drowsiness for each of the four time periods using a convolutional neural network (CNN) based on ToF camera-based eye blinking, ECG information (pulse rate) embedded in the steering wheel, and vehicle information (steering angle data). In order to build a reliable and high-quality training dataset (Ground Truth) for the CNN algorithm, a baseline was established by matching the driver's face image with the electrocardiogram (ECG) and electroencephalogram (EEG) changes in the drowsy and normal states. In a simulation test of the proposed CNN algorithm using more than 20,000 driver image data acquired using a driving simulator, the TNR was 94.8% and the accuracy was 94.2%. Our proposed method is expected to minimize human errors that may occur when switching control by monitoring inappropriate driver state (drowsiness) in real time.

Aiming at the problems of inaccurate segmentation of long object and information loss of small object in real-time semantic segmentation algorithm, this paper proposes a lightweight multi-branch real-time semantic segmentation network based on BiseNetV2. The new auxiliary branch makes full use of spatial details and context information to cover the long object in the field of view. Meanwhile, in order to ensure the inference speed of the model, the asymmetric convolution is used in each stage of the auxiliary branch to design a structure with low computational complexity. In the multi-branch fusion stage, the alignment and fusion module is designed to provide guidance information for deep and shallow feature mapping, so as to make up for the problem of feature misalignment in the fusion of information at different scales, and thus reduce the loss of small target information. In order to further improve the model’s awareness of key information, a global context module is designed to capture the most important features in the input data. The proposed network uses NVIDIA GeForce RTX 3080 Laptop GPU experiment, on the road street view data set Cityscapes and CamVid average occurring simultaneously ratio reached 77.1% and 78.4% respectively, with running speed of 127 frames/s respectively and 112 frames/s. The experimental results show that the proposed algorithm can achieve real-time segmentation and improve the accuracy significantly, showing good semantic segmentation performance

The exponential growth of population and the consistent food demand has compelled humanity to seek alternatives to traditional farming and innovative technologies to increase production. Exploring offshore for natural resources and alleviating pressure on land has been an ongoing research field, especially in the energy and aquaculture sector. However, the idea of floating farming is still in its infancy and requires significant innovations. The work presented here shed further light on this area by proposing a comprehensive model of ‘Integrated, multicultural, Multileveled Floating Farm (MFF).’ Various aspects of planning, design, constructions and operations of such MFF are discussed. An integrated waste management system is proposed to improve sustainability. The conceptual design and associated financial analysis demonstrated that such integration of various modes of farming could be profitable and sustainable at the same time. The cost estimation and profit analysis are presented in the context of Singapore, and a conservative approach is followed for the calculation. However, the model can easily be extended for application in other countries.

Street view images can help us better understand the city environment and potential characteristics. With the development of computer vision and deep learning, the technology of semantic segmentation algorithms has become more mature. However, DeeplabV3+, which is commonly used in semantic segmentation, has shortcomings such as a large number of parameters, high requirements for computing resources, and easy loss of detailed information. Therefore, this paper proposes LM-DeeplabV3+, which aims to greatly reduce parameters and computations of the model while ensuring the segmentation accuracy. Firstly, the lightweight network MobileNetV2 is selected as the backbone network, and the ECA attention mechanism is introduced after MobileNetV2 extracts shallow features to improve the ability of feature representation; secondly, the ASPP module is improved, and on its basis, the EPSA attention mechanism is introduced to achieve cross-dimensional channel attention and important feature interaction; thirdly, a loss function named CL loss is designed to balance the training offset of multiple categories and better indicate the segmentation quality. This paper conducted experimental verification on the Cityspaces dataset, and the results showed that the mIoU reached 74.9%, which was an improvement of 3.56% compared to DeeplabV3+; the mPA reached 83.01%, which was an improvement of 2.53% compared to DeeplabV3+.

One control challenge in prosthetic legs is seamless transition from one gait mode to another. User intent recognition (UIR) is a high-level controller that tells a low-level controller to switch to the identified activity mode, depending on the user’s intent and environment. We propose a new framework to design an optimal UIR system with simultaneous maximum performance and parsimony for gait mode recognition. We use multi-objective optimization (MOO) to find an optimal feature subset that creates a trade-off between these two conflicting objectives. The main contribution of this paper is two-fold: (1) a new gradient-based multi-objective feature selection (GMOFS) method for optimal UIR design; and (2) the application of advanced evolutionary MOO methods for UIR. GMOFS is an embedded method that simultaneously performs feature selection and classification by incorporating an elastic net in multilayer perceptron neural network training. Experimental data are collected from six subjects, including three able-bodied subjects and three transfemoral amputees. We implement GMOFS and four variants of multi-objective biogeography-based optimization (MOBBO) for optimal feature subset selection, and we compare their performances using normalized hypervolume and relative coverage. GMOFS demonstrates competitive performance compared to the four MOBBO methods. We achieve a mean classification accuracy of 97.14% ± 1.51% and 98.45% ± 1.22% with the optimal selected subset for able-bodied and amputee subjects, respectively, while using only 23% of the available features. Results thus indicate the potential of advanced optimization methods to simultaneously achieve accurate, reliable, and compact UIR for locomotion mode detection of lower-limb amputees with prostheses.

The robust fusion estimation problem of a multi-sensor system with communication constraints, random packet drops, d-step state delays, and deterministic control inputs is investigated. The state augmentation method is used to transform the time delay system into a non-time delay system, and then a robust fusion estimation algorithm based on sensitivity penalty is proposed and the pseudo-cross-covariance matrix is updated. An event-triggered robust fusion estimator with state delay and deterministic control inputs is considered to make the model closer to reality while effectively reducing network congestion and lowering communication costs. Conditions are given to ensure that the estimation errors of the robust fusion estimator are uniform boundedness. Two sets of numerical simulations are provided to illustrate the effectiveness of the derived fusion estimator.

This study presents a modest attempt to interpret, formulate, and manipulate emotion of robots within the precepts of quantum mechanics. Our proposed framework encodes the emotion information as a superposition state whilst unitary operators are used to manipulate the transition of the emotion states which are recovered via appropriate quantum measurement operations. The framework described provides essential steps towards exploiting the potency of quantum mechanics in a quantum affective computing paradigm. Further, the emotions of multi-robots in a specified communication scenario are fused using quantum entanglement thereby reducing the number of qubits required to capture the emotion states of all the robots in the environment, and fewer quantum gates are needed to transform the emotion of all or part of the robots from one state to another. In addition to the mathematical rigours expected of the proposed framework, we present a few simulation-based demonstrations to illustrate its feasibility and effectiveness. This exposition is an important step in the transition of formulations of emotional intelligence to the quantum era.

It's challenging to achieve robust lane detection depending on single frame when considering complicated scenarios. In order to detect more credible lane markings by using sequential frames, a novel approach to fusing vision and Inertial Measurement Unit (IMU) is proposed in this paper. The hough space is employed as the space where lane markings are stored and it's calculated by three steps. Firstly, a basic hough space is extracted by Hough Transform and primary line segments are extracted from it. In order to measure the possibility about line segments belong to lane markings, a CNNs based classifier is introduced to transform the basic hough space into a probabilistic space by using the networks outputs. However, this probabilistic hough space based on single frame is easily disturbed. In the third step, a filtering process is employed to smooth the probabilistic hough space by using sequential information. Pose information provided by IMU is applied to align hough spaces extracted at different times to each other. The final hough space is used to eliminate line segments with low possibility and output those with high confidence as the result. Experiments demonstrate that the proposed approach has achieved a good performance.

The proliferation of Unmanned Aerial Vehicle (UAV) networks is increasing, driven by their capacity to deliver automated services tailored to the varied demands of numerous smart city applications. Trust, security, and privacy remain paramount in the public domain. Traditional centralized network designs fall short of ensuring device authentication, data integrity, and privacy within UAV networks’ highly dynamic and adaptable environments. Decentralized reputation systems emerge as a promising solution to enhance the reliability and trustworthiness of data and communications within these networks while safeguarding UAV security. This paper presents an exhaustive survey of trust and reputation systems, exploring existing frameworks and proposed innovations alongside their inherent challenges. It highlights the crucial role of reputation systems in strengthening trust, security, and privacy throughout these networks and discusses various strategies to mitigate existing vulnerabilities. As a useful resource for researchers and practitioners seeking to advance the state of the art in UAV network security, we hope this survey will spark further community discussion and stimulate innovative ideas in this burgeoning field.

Land cover mapping provides spatial information on the physical properties of the Earth’s surface, for various classes of wetlands, artificial surface and constructions, vineyards, water bodies, etc. Having reliable information on land cover is crucial to developing solutions to a variety of environmental problems such as destruction of important wetlands/forests, and loss of fish and wildlife habitats. This has made land cover mapping one of the most widespread application areas in remote sensing computational imaging. However, due to the differences between modalities in terms of resolutions, content, and sensors, integrating complementary information that multi-modal remote sensing imagery exhibits into a robust and accurate system still remains challenging, and classical segmentation approaches generally do not give satisfactory results for land cover mapping. In this paper, we propose a novel dynamic deep network architecture, AMM-FuseNet, that promotes the use of multi-modal remote sensing images for the purpose of land cover mapping. The proposed network exploits the hybrid approach of the Channel Attention mechanism and Densely Connected Atrous Spatial Pyramid Pooling (DenseASPP). In the experimental analysis, in order to to verify the validity of the proposed method, we test AMM-FuseNet applied to four datasets whilst comparing it to the 6 state-of-the-art models of DeepLabV3+, PSPNet, UNet, SegNet, DenseASPP, and DANet. In addition, we also demonstrate the capability of AMM-FuseNet under minimal training supervision (reduced number of training samples) compared to the state-of-the-art, achieving less accuracy loss even for the case with 1/20 of the training samples.

Security has become a critical issue in today’s highly distributed and networked systems. Network intrusion detection systems (NIDSs), especially signature-based NIDSs, are being widely deployed in a distributed network environment with the purpose of defending against a variety of network attacks. Most of the commercially available NIDSs are software based and rely on pattern matching to extract the threat from network traffic. The increase in network speed and traffic may make existing algorithms to become a performance bottleneck. Therefore it is very necessary to develop faster and more efficient pattern matching algorithm in order to overcome the troubles on performance of NIDSs. Therefore, we propose a multi fusion pattern matching algorithm for Network Intrusion Detection Systems. The results obtained in percentages from the proposed fusion algorithm given better values in terms processing time in milliseconds than the existing algorithms when data English text are applied to evaluate the fusion performances.

In order to meet the diversified rehabilitation needs of people, it is necessary to design a product for active rehabilitation of patients. Existing rehabilitation chairs use intelligent massage, which can cause problems such as large massage area, inability to local massage, large chair size, and inability to meet the continuous use of the damaged parts. In this paper, modular design method and multi-layer evaluation method are used to solve the problems related to rehabilitation chairs. The authors use the questionnaire survey method and the functional technology matrix method to determine the functional requirements of the rehabilitation chair, and then use the multilevel evaluation methods, including the AHP method, entropy weight method main and grey correlation analysis, to optimize the functional solutions of the rehabilitation chair, and finally obtain a chair for the rehabilitation of patients with upper and lower limb disorders. Problems such as generalisation of rehabilitation scope and non-durable use of components were solved, and the purpose of active exercise was achieved. This study verifies that the use of multilevel decision evaluation method can effectively improve the efficiency of programme decision-making, and provides theoretical and practical basis for the design of similar products.

Against the background of the sky, imaging and perception of aircraft are cru- cial. Among these, the location of aircrafts is not a very difficult problem in general, because that deep learning algorithms and object detection models based on convolutional neural network(CNN) are ever-evolving. In order to have a better identification of various aircrafts, or to accurately locate different types of flying targets in the sky, it becomes significant to distinguish and recognize different types of aircrafts. One question is that, distinguishing and recognizing between sub-categories of aircrafts pose great challenges. Although fine-grained recognition focuses on exploring and studying such problems, aircrafts under different sub-categories and granularities lead us to rethink the application of features. We noticed that features in swin-transformer demonstrates the under- standing and processing of images, fully showcasing the encoding and indexing of information. Through this research and proposed approach, we discovered a better understanding of features encoding and use, which are inspired by the feature encoding and computation in swin-transformer. In our paper, our ap- proach has achieved effects on features encoded graphically, manifested from the architecture design and convolutional neural network computation, and outper- forms other famous fine-graiend classification models on this issue. Not only approach we proposed has demonstrated superior performance in fine-grained aicraft classification, but also the mechanisms of the feature encoding under different sample space partitions is revealed for this issue, which provides a re- search for the representation of aircraft fineness characteristics. The relationship between representation orientation of aircraft feature information under various grained divisions, shows that the recognition of different categories according to man-made, such as aircraft, can be achieved by means of the feature represen- tations studied in this paper, for more specific defined classification tasks and various man-made targets partition criteria, which may influences the princi- ple of design for calculation and feature extraction in fine-grained classification models.

Drone and aerial remote sensing images are widely used, but their imaging environment is complex and prone to image blurring. Existing CNN deblurring algorithms usually use multi-scale fusion to extract features in order to make full use of aerial remote sensing blurred image information, but images with different degrees of blurring use the same weights, leading to increasing errors in the feature fusion process layer by layer. Based on the physical properties of image blurring, this paper proposes an adaptive multi-scale fusion blind deblurred generative adversarial network (AMD-GAN), which innovatively applies the degree of image blurring to guide the adjustment of the weights of multi-scale fusion, effectively suppressing the errors in the multi-scale fusion process and enhancing the interpretability of the feature layer. The research work in this paper reveals the necessity and effectiveness of a priori information on image blurring levels in image deblurring tasks. By studying and exploring the image blurring levels, the network model focuses more on the basic physical features of image blurring. Meanwhile, this paper proposes an image blurring degree description model, which can effectively represent the blurring degree of aerial remote sensing images. The comparison experiments show that the algorithm in this paper can effectively recover images with different degrees of blur, obtain high-quality images with clear texture details, outperform the comparison algorithm in both qualitative and quantitative evaluation, and can effectively improve the object detection performance of aerial remote sensing blurred images. Moreover, the average PSNR of this paper's algorithm tested on the publicly available dataset RealBlur-R reached 41.02dB, surpassing the latest SOTA algorithm.

The multi-modal data fusion method based on IA-net and CHMM technical proposed is designed to solve the problem that the incompleteness of target behavior information in complex family environment leads to the low accuracy of human behavior recognition.The two improved neural networks(STA-ResNet50、STA-GoogleNet)are combined with LSTM to form two IA-Nets respectively to extract RGB and skeleton modal behavior features in video. The two modal feature sequences are input CHMM to construct the probability fusion model of multi-modal behavior recognition.The experimental results show that the human behavior recognition model proposed in this paper has higher accuracy than the previous fusion methods on HMDB51 and UCF101 datasets. New contributions: attention mechanism is introduced to improve the efficiency of video target feature extraction and utilization. A skeleton based feature extraction framework is proposed, which can be used for human behavior recognition in complex environment. In the field of human behavior recognition, probability theory and neural network are cleverly combined and applied, which provides a new method for multi-modal information fusion.

The quality of digital elevation models (DEMs) is inevitably affected by the limitations of the imaging modes and the generation methods. One effective way to solve this problem is to merge the available datasets through data fusion. In this paper, a fusion-based global DEM dataset (82°S-82°N) is introduced, which we refer to as GSDEM-30. This is a 30-m DEM mainly reconstructed from the unfilled SRTM1, AW3D30, and ASTER GDEM v2 datasets combining the multi-source and multi-scale fusion techniques. A comprehensive evaluation of the GSDEM-30 data, as well as the 30-m ASTER GDEM v2 and AW3D30 DEM, was presented. Global ICESat GLAS data and the local National Elevation Dataset (NED) were used as the reference for the vertical accuracy validation, while GlobeLand30 was introduced for the landscape analysis. Furthermore, we employed the maximum slope approach to detect the potential artefacts in the DEMs. The results show that the GDEM data are seriously affected by noise and artefacts. With the advantage of the multiple datasets and the refined post-processing, the GSDEM-30 are contaminated with fewer anomalies than both ASTER GDEM and AW3D30. The fusion techniques used can also be applied to the reconstruction of other fused DEM datasets.

In this study, we introduce and advanced multi-feature selection technique for bacterial classifi-cation employing the Salp Swarm Algorithm (SSA). We enhance SSA’s effectiveness by incorpo-rating the Opposition-Based Learning (OBL) strategy and the Local Search (LSA) algorithm. The proposed technique encompasses three key stages, streamlining the automated categorization of bacteria based on their distinctive features. The research adopts a multi-feature selection approach bolstered by an enhanced iteration of the Salp Swarm Algorithm (SSA). Enhancements include the utilization of Opposition-Based-Learning (OBL) to increase population diversity during search and Local Search Algorithms (LSA) to tackle local optimization challenges. The ISSA algorithm is designed to optimize the multi-feature selection by increasing the number of selected feature and improving classification accuracy. This study compares its performance with several other algo-rithms across ten distinct test datasets. The comparison results show that ISSA has better perfor-mance in terms of classification accuracy on 3 datasets consisting of 19 features, with a value reaching 73.75%. Additionally, ISSA excels in determining the optimal feature count and producing a better-fit value with a classification error rate of 0,249. Thus, the ISSA technique is expected to make a significant contribution to solving feature selection problems in bacterial analysis

Purpose: Emotional voice conveys important social cues that demand listeners’ attention and timely processing. This event-related potential study investigated the feasibility of a multi-feature oddball paradigm to examine adult listeners’ neural responses to detecting emotional prosody changes in non-repeating naturally spoken words. Method: Thirty-three adult listeners completed the experiment by passively listening to the words in neutral and three alternating emotions while watching a silent movie. Previous research documented pre-attentive change-detection electrophysiological responses (e.g., MMN, P3a) to emotions carried by fixed syllables or words. Given that the MMN and P3a have also been shown to reflect extraction of abstract regularities over repetitive acoustic patterns, the current study employed a multi-feature oddball paradigm to compare listeners’ MMN and P3a to emotional prosody change from neutral to angry, happy, and sad emotions delivered with hundreds of non-repeating words in a single recording session. Results: Both MMN and P3a were successfully elicited by the emotional prosodic change over the varying linguistic context. Angry prosody elicited the strongest MMN compared to happy and sad prosodies. Happy prosody elicited the strongest P3a in the centro-frontal electrodes, and angry prosody elicited the smallest P3a. Conclusions: The results demonstrated that listeners were able to extract the acoustic patterns for each emotional prosody category over constantly changing spoken words. The findings confirm the feasibility of the multi-feature oddball paradigm in investigating emotional speech processing beyond simple acoustic change detection, which may potentially be applied to pediatric and clinical populations.

Speech emotion recognition is a critical component for achieving natural human-robot interaction. The modulation-filtered cochleagram is a feature based on auditory modulation perception, which contains multi-dimensional spectral-temporal modulation representation. In this study, we propose an emotion recognition framework that utilizes a multi-level attention network to recognize emotions from the modulation-filtered cochleagram. The channel-level attention and spatial-level attention modules are used to capture emotional saliency maps of channel and spatial feature representations from the 3D convolution feature maps, respectively. Furthermore, the temporal-level attention module captures significant emotional regions from the concatenated feature sequence of the emotional saliency maps. Our experiments on the IEMOCAP dataset demonstrate that the modulation-filtered cochleagram significantly improves the prediction performance of categorical emotion compared to other evaluated features. Moreover, our emotion recognition framework achieves a better unweighted accuracy of 71% in categorical emotion recognition than several existing approaches in the experiments. In summary, our study demonstrates the effectiveness of the modulation-filtered cochleagram in speech emotion recognition, and our proposed multi-level attention framework provides a promising direction for future research in this field.

This study aims to understand the current state of research in urban resilience and to open a discussion about multi-level perspectives for this concept. Starting with the history of the concept of resilience, we identify three main stages in resilience concept’s evolution: conceptualization, contextualization and operationalization. Confusion occurs between sustainability and resilience, therefore we clearly separate these two concepts by creating conceptual maps. Such maps also underline the specificities of urban and regional resilience discourses. We illustrate that urban resilience research, operating within intra-urban processes, is oriented towards natural disasters, while regional resilience research, operating mostly within inter-urban processes, is oriented towards economic shocks. We show that these two approaches to resilience – urban and regional – are complementary, and we propose to integrate them into a multi-level perspective. By combining these two discourses, we propose a multi-level approach to urban resilience that takes into account both top-down and bottom-up resistance processes. In the discussion section, we propose to take the panarchy perspective as a theoretical framework for multi-level urban resilience, that explains the interactions between different levels through adaptive cycles, relationships between which can help to explain urban resilience.

The purpose of this paper is to describe a feature selection algorithm and its application to enhance the accuracy of the reconstruction of phylogenetic trees by improving the efficiency of tree construction. Applying machine learning models for Arabic and Aramaic scripts such as Deep Neural Networks (DNN), Support Vector Machines (SVM) and Random Forests (RF) each model was used to compare the phylogenies. The methodology was utilized with a dataset containing Arabic and Aramaic scripts, demonstrating its relevance in a range of phylogenetic analyses. Results emphasize the essential role of feature selection by DNNs is outperforming other models in terms of Area Under the Curve (AUC) and Equal Error Rate (EER) across various datasets and fold sizes. Additionally, SVM and RF models show important understandings, advantages and limits of each approach within the context of phylogenetic analysis. This method not only simplifies the tree structures but also enhances their Consistency Index. Therefore offering a robust framework for evolutionary studies. The findings highlight the applications of machine learning in phylogenetics suggesting a path toward accurate and efficient evolutionary analyses and adding a deeper understanding of evolutionary relationships.

Analyzing horse behavior is crucial for assessing training quality, particularly in accurately identifying actions like standing, walking, and running. Research has been conducted both domestically and internationally in this regard; however, challenges persist, including reliance on single sensors, poor real-time performance, and low identification accuracy. In view of these challenges, this study investigated real-time identification of horse behavior based on wearable devices. The system, centered around a microcontroller and utilizing a 4G network as a carrier, employs multi-axis IMU sensors as input sources to perceive horse posture. The study proposed a behavior classification method based on analysis of acceleration thresholds. The method consists primarily of two sequential stages: first, the resultant acceleration interval counting method, which employs a nonlinear segmentation approach for initial behavior classification; and second, the statistical analysis of variance parameters between segments, which when coupled with multi-level threshold processing, achieves a refined classification. Experimental results indicated that the interval counting method alone achieved an accuracy of 87.55%, while for the variance analysis method alone the accuracy was 90.87%. The proposed method, comprising the two stages, reached a classification accuracy of 91.57%, underpinning its usefulness in supporting future equine research.

Providing safely-managed sanitation/hygiene requires key competencies for education, training, service delivery, enterprise development and management, product and infrastructure design and development, construction and installation, governance, financing, research, etc. These strategic sanitation capacities will have to be built from higher education’s academic and professional programmes structured and designed to produce skilled and knowledgeable professionals and practitioners. This study aims to investigate the quality of the sanitation/hygiene management content of Environmental Health programmes; the adequacy of the existing central curricula; and the perception of environmental health sanitation/hygiene-trained professionals to determine the next phase for building knowledge and capacity of sanitation professionals in Nigeria through higher education institutions. A multilevel mixed method concurrent study was used for sampling and data collection with a multilevel perception analysis to examine the perceptions of students, lecturers and graduate alumni as well as employers/supervisors and clients/service-users of sanitation/hygiene related graduates. The main findings of this paper show limited understanding of the concepts of contemporary issues of sanitation/hygiene management; the central/national teaching and curricula were found to be seriously out of date; and clients/service-users were dissatisfied with the skills and knowledge levels of the graduates. The study concludes by recommending a national sanitation management higher education pathway to guide the provision of integrated sanitation/hygiene management education at higher education institutions (HEIs), to build effective, efficient, competent, and sufficient workforce for the country.

Insulin-like Growth Factor 2 mRNA binding Protein 1 (IGF2BP1) is a promising candidate gene for influencing economic traits in livestock and poultry. The second intronic region of IGF2BP1 is closely related to growth traits, although the exact mechanism remains unclear. The preliminary resequencing results of goats in our laboratory demonstrated that the IGF2BP1 gene region was subject to strong selection, then we selected multiple single nucleotide polymorphisms (SNPs) to identify. Seven SNPs were located in the conserved region of the second intron, and their functions and relationship with growth traits require further clarification. This study selected 348 Nanjiang Yellow goats as experimental subjects, and chose Time of Flight Mass Spectrometer (MALDI-TOF) as the main method. Our results showed that six SNPs were significantly correlated with the growth traits of the Nanjiang yellow goat by association analysis. Among them, the rs652062749(A>G) was identified as the primary locus affecting the later stage of growth traits of Nanjiang yellow goat. Moreover, there was a strong linkage among three SNPs, and the rs638185407 (T>A) mutant type was observed to increase the luciferase activity in H293T cells extremely significantly. The TTAACT, TTCCCC, and ATCACT combination genotypes were dominant combination genotypes, which provided theoretical references for further genetic co-breeding. The results of our study demonstrate the potential role of the IGF2BP1 gene as a functional genetic marker in the breeding programs of Nanjiang Yellow goats.

The colossal depths of the deep neural network sometimes suffer from ineffective backpropagation of the gradients through all its depths. Whereas, The strong performance of shallower multilayer neural structures prove their ability to increase the gradient signals in the early stages of training which easily gets backpropagated for global loss corrections. Shallow neural structures are always a good starting point for encouraging the sturdy feature characteristics of the input. In this research, a shallow, deep neural structure called PrimeNet is proposed. PrimeNet is aimed to dynamically identify and encourage the quality visual indicators from the input to be used by the subsequent deep network layers and increase the gradient signals in the lower stages of the training pipeline. In addition to this, the layerwise training is performed with the help of locally generated errors which means the gradient is not backpropagated to previous layers, and the hidden layer weights are updated during the forward pass, making this structure a backpropagation free variant. PrimeNet has obtained state-of-the-art results on various image datasets, attaining the dual objective of (1) compact dynamic deep neural structure, which (2) eliminates the problem of backwards-locking. The PrimeNet unit is proposed as an alternative to traditional convolution and dense blocks for faster and memory-efficient training, outperforming previously reported results aimed at adaptive methods for parallel and multilayer deep neural systems.

To enhance the performance of visual SLAM in underwater environments, this paper presents an enhanced front-end method based on visual feature enhancement. The method comprises three modules aimed at optimizing and improving the matching capability of visual features from different perspectives. Firstly, to address issues related to insufficient underwater illumination and uneven distribution of artificial light sources, a brightness consistency recovery method is pro-posed. This method employs an adaptive histogram equalization algorithm to balance the brightness of images. Secondly, a method for denoising underwater suspended particulates is introduced to filter out noise from the images. After image-level processing, a combined underwater acousto-optic feature association method is proposed, which associates acoustic features from sonar with visual features, thereby providing distance information for the visual features. Finally, utilizing the AFRL dataset, the improved system incorporating the proposed enhancement methods is evaluated for performance against the OKVIS framework. The system achieves better trajectory estimation accuracy compared to OKVIS and demonstrates robustness in underwater environments.

To ensure consumers can purchase high-quality rice, accurate identification of rice varieties is particularly important. Methods: This article conducts research based on machine learning algorithms from the perspectives of image, spectrum, and spectrogram fusion. Six types of hyperspectral image data of rice were preprocessed using convolutional smoothing (SG) and multiple scatter correction (MSC). Texture information of the images was extracted using gray-level co-occurrence matrix. Spectra, texture, and spectrogram data were fused into a new matrix. Spectral, texture, and spectrogram fusion data were used as inputs for the model, and support vector machines (SVM), logistic regression (LR), and K-nearest neighbors (KNN) classification models were constructed and compared. Results: From the classification results, the spectrogram fusion classification performance was better than classification models using only spectra or texture. Conclusion: The research results showed that the accuracy of SVM and LR classification models exceeded 90%, and the LR model performed the best, effectively classifying rice varieties.

Yttrium oxide (Y2O3) resistive random-access memory (RRAM) devices were fabricated using the sol–gel process on indium tin oxide/glass substrates. These devices exhibited conventional bipolar RRAM characteristics, without requiring a high voltage forming process. The effect of current compliance on the Y2O3 RRAM devices was investigated and revealed that as the set current compliance values increased, the resistance values gradually decreased. Consequently, intermediate resistance values were obtained, which are suitable for multi-level cell (MLC) switching. The fabricated Y2O3 RRAM devices were capable of functioning as an MLC with a capacity of 2 bits in one cell, utilizing three low resistance states and one common high resistance state. The potential of the Y2O3 RRAM for neural networks was further explored through numerical simulations. Hardware neural networks based on the Y2O3 RRAM demonstrated effective digit image classification with a high accuracy rate of approximately 88%, comparable to the ideal software-based classification (~92%). This suggests that the proposed RRAM has the potential to be utilized as a memory component in practical neuromorphic systems.

As the world’s second largest economy, China ranks amount the world’s top nations when it comes to carbon emission, and therefore its attitude towards climate change is closely followed by all parties concerned. There have been few researches on the role of environmental governance in low-carbon city transformation process, especially the Chinese one. This paper analyses the role of government environmental regulation played in the low-carbon city transformation process by taking Shenzhen as the research object. One of the world's youngest super cities, it also happens to be the lowest carbon emission intensity city in China. Striving to explore green low-carbon development path for the whole country, Shenzhen provides practical experience for countries to cope with global climate change. However, its efforts to reduce the total carbon emissions failed, but it emphasized the carbon emission intensity, which is consistent with the international commitments made by the central government. China’s policy towards handling climate change relies on hierarchical governance arrangement. The strength of the NGOs in the country is weak and incomparable with the government’s, which has mastered most of the resources and is just a reality in China.

This study extends the CGAN-based MODIS cloud vertical-profile (64×64-scene, about 70km width×15km height) retrieval technique developed by Leinonen et al. (2019) [1] to construct seamless 3D cloud fields for the MODIS granules [2]. Firstly, the accuracy and spatial continuity of the Leinonen et al. [1] retrievals are statistically evaluated. Then, according to the characteristics of the retrieval error, a spatially overlapping-scene ensemble generation method and a bi-directional Ensemble Binning Probability Fusion (CGAN-BEBPF) technique are developed, which improves the CGAN retrieval accuracy and support to construct seamless 3D clouds. The CGAN-BEBPF technique involves three steps: cloud masking, intensity scaling, and optimal value selection. It ensures adequate coverage of the low reflectivity areas while preserving the high reflectivity cloud cores. The technique is applied to retrieve the 3D cloud fields of Typhoon Chaba and a multi-cell convective system. The cloud structures of the CGAN-BEBPF results are highly consistent with the ground-based radar observations. It can retrieve weak clouds at the top levels of convective storms and hurricanes that were missed by ground-based radars and fill the gaps of the ground-based radar lower-level PPI scan. Overall, CGAN-BEBPF can be automated to retrieve the rich informative 3D structures of typhoons and severe convective systems over the oceans along the MODIS swath, which certainly presents great value for improving maritime and near-shore typhoons and convection prediction and other cloud-sensitive applications.

Ambulatory heart rate (HR) monitors that acquire electrocardiogram (ECG) or/and photoplethysmographm (PPG) signals from the torso, wrists, or ears are notably less accurate in tasks associated with high levels of movement compared to clinical measurements. However, reliable estimation of HR can be obtained through data fusion from different sensors. These methods are especially suitable for multimodal hearable devices, where heart rate can be tracked from different modalities, including electrical ECG, optical PPG, and sounds (heart tones). Combined information from different modalities can compensate for single source limitations. In this paper, we evaluate the possible application of data fusion methods in hearables. We assess data fusion for heart rate estimation from simultaneous in-ear ECG and in-ear PPG, recorded on eight subjects while performing 5-minute sitting and walking tasks. Our findings show that data fusion methods provide a similar level of mean absolute error as the best single-source heart rate estimation but with much lower intra-subject variability, especially during walking activities. We conclude that data fusion methods provide more robust HR estimation than a single cardiovascular signal. These methods can enhance the performance of wearable devices, especially multimodal hearables, in heart rate tracking during physical activity.

Link prediction is a fundamental problem in graphs. Currently, SubGraph Representation Learning (SGRL) methods provide state-of-the-art solutions for link prediction by transforming the task into a graph classification problem. However, existing SGRL solutions suffer from high computational costs and lack scalability. In this paper, we propose a novel SGRL framework called Augmented Feature Diffusion on Sparsely Sampled Subgraph (AFD3S). The AFD3S first uses a conditional variational autoencoder to augment the local features of the input graph, effectively improving the expressive ability of downstream Graph Neural Networks. Then, based on a random walk strategy, sparsely sampled subgraphs are obtained from the target node pairs, reducing computational and storage overhead. Graph diffusion is then performed on the sampled subgraph to achieve specific weighting. Finally, the diffusion matrix of the subgraph and its augmented feature matrix are used for feature diffusion to obtain operator-level node representations as inputs for the SGRL-based link prediction. Feature diffusion effectively simulates the message-passing process, simplifying subgraph representation learning, thus accelerating the training and inference speed of subgraph learning. Our proposed AFD3S achieves optimal prediction performance on several benchmark datasets, with significantly reduced storage and computational costs.

This note makes available five years of data gathered in a measurement site equipped with a micrometeorological station and two monitoring wells. Series of data of hydrological and atmospheric variables allow to estimate the flux of water across the atmosphere-land interface and to calculate the water budget, that are crucial topics in climate and environmental sciences. The measures of water-table began during 2017, one of the driest year of the whole instrumental period of climate history for the Central Mediterranean. Data from the micrometeorological station have been used to construct two more datasets of daily and monthly cumulative values of different terms of the surface water budget, from which the groundwater infiltration has been estimated. A significant decreasing trend characterizes both the data time series of piezometric heights and estimated infiltration in the considered period. All the data are collected in 2017-2022 datasets that are available in the supplementary material.

Theoretical studies propose an association between family violence and low self-control in adolescence, yet empirical findings of this association are inconclusive. The aim of the present research was to systematically summarize available findings on the relation between family violence and self-control across adolescence. We included 27 studies with 143 effect sizes, representing more than 25,000 participants of eight countries from early to late adolescence. Applying a multi-level meta-analyses, taking dependency between effect sizes into account while retaining statistical power, we examined the magnitude and direction of the overall effect size. Additionally, we investigated whether theoretical moderators (e.g., age, gender, country), and methodological moderators (cross-sectional/longitudinal, informant) influenced the magnitude of the association between family violence and self-control. Our results revealed that family violence and self-control have a small to moderate significant negative association (r = -.191). This association did not vary across gender, country, and informants. The strength of the association, however, decreased with age and in longitudinal studies. This finding provides evidence that researchers and clinicians may expect low self-control in the wake of family violence, especially in early adolescence. Recommendations for future research in the area are discussed.

To address the problems of insufficient semantic fusion between text and questions and the lack of consideration of global semantic information encountered in machine reading comprehension models, we proposed a machine reading comprehension model called BERT_hybrid based on BERT and hybrid attention mechanism. In this model, BERT is utilized to separately map the text and questions into the feature space. Through the integration of Bi-LSTM, attention mechanism, and self-attention mechanism, the proposed model achieves comprehensive semantic fusion between text and questions. The probabilities distribution of answers is computed using Softmax. Experimental results on the public dataset DuReader demonstrate that the proposed model achieves improvements in BLEU-4 and ROUGE-L scores compared to existing models. Furthermore, to validate the effectiveness of the proposed model design, we analyze the factors influencing the model’s performance.

In pattern recognition domain, deep architectures are widely used nowadays and they have achieved fine grades. However, these deep architectures need special demands, especially big datasets and GPU. Aiming to gain better grades without deep networks, we propose a simplified algorithm framework using fusion features extracted from the salient areas of faces. Furthermore, the proposed algorithm has achieved a better result than some deep architectures. For extracting more effective features, this paper firstly defines the salient areas on the faces. This paper normalizes the salient areas of the same location in the faces to the same size, therefore it can gain more similar features from different subjects. LBP and HOG features are extracted from the salient areas, fusion features’ dimensions are reduced by Principal Component Analysis( PCA) and we apply softmax to classify the six basic expressions at once. This paper proposes a salient areas definitude method which uses peak expressions frames to compare with their neutral faces. This paper also proposes and applies the idea of normalizing the salient areas to align the specific areas which express the different expressions. This makes the salient areas found from different subjects have the same size. Besides, gamma correction method is firstly applied on LBP features in our algorithm framework which improves our recognition rates significantly. By applying this algorithm framework, our research has gained state-of-the-art performances on CK+ database and JAFFE database.

The task of weakly supervised point cloud semantic segmentation has received widespread attention and also has been widely used in autonomous driving, robotics, and modern industries. Due to the high dimensionality and large volume of point cloud data, many technical difficulties appeal to weakly supervised semantic processing, especially segmentation. However, weakly-supervised schemes usually provide only partial labelling information of the underlying point cloud data and thus need to effectively extract local features, geometric information, and contextual relationships only using these limited labelled data for supervised learning. To improve weakly-supervised semantic segmentation, we propose a novel segmentation network through the boundary-based feature aggregation based on a K-NN algorithm with down-sampling operation, and also introduce the smoothness loss and Siamese loss for effective segmentation. The experiments on public datasets demonstrate that our presented segmentation method exceeds most of the existing fully supervised and weakly-supervised methods in terms of mIoU. Specifically, our network has high segmentation accuracy on the labels of objects with similar geometrical structures, such as ceiling, wall, floor, chair and table, reaching 91.2%, 98.8%, 83.3%, 75.3% and 80.2%, respectively. Extensive experiments also illustrate its robustness, effectiveness, and generalization of the proposed weakly supervised segmentation network.

This research introduces a hybrid feature extraction approach that combines Linear Discriminant Analysis (LDA) and Multilayer Perceptron (MLP) methods to address the challenges of reducing feature vector dimensionality and accurately classifying smartphone-based human activities. Moreover, to refine activity classification accuracy, Support Vector Machine (SVM) optimization with Stochastic Gradient Descent (SGD) is employed. LDA, a statistical tool, is leveraged to derive a new feature space for data projection, enhancing class separation and test feature label prediction. The proposed approach, named LMSS, was evaluated using the UCI-HAR dataset and compared with state-of-the-art models. The results demonstrate that the proposed approach outperformed the best-performing method over this dataset. It achieved an accuracy rate of 99.52%, precision of 99.55%, recall of 99.53%, and an F1-score of 99.54%, highlighting the effectiveness of the proposed method in accurately classifying the data.

The use of Hyperspectral (HS) and LiDAR acquisitions has a great potential to enhance mapping and monitoring practices of endangered grasslands habitats, beyond conventional botanical field surveys. In this study we assess the potentiality of Recursive Feature Elimination (RFE) in combination with Random Forest (RF) classification in extracting the main HS and LiDAR features needed to map selected Natura 2000 grasslands along Polish lowland river valleys, in particular alluvial meadows 6440, lowland hay meadows 6510 and xeric and calcareous grasslands 6120. We developed an automated RFE-RF system capable to combine the potentials of both techniques and applied it to multiple acquisitions. Several LiDAR-based products and different Spectral Indexes (SI) were computed and used as input in the system, with the aim of shedding light on the best-to-use features. Results showed a remarkable increase in classification accuracy when LiDAR and SI products are added to the HS dataset, strengthening in particular the importance of employing LiDAR in combination with HS. Using only the 24 optimal features selection generalized over the three study areas, strongly linked to the highly heterogeneous characteristics of the habitats and landscapes investigated, it was possible to achieve rather high classification results (K around 0.7-0.77 and habitats F1 accuracy around 0.8-0.85), indicating that the selected Natura 2000 meadows and dry grasslands habitats can be automatically mapped by airborne HS and LiDAR data. Similar approaches might be considered for future monitoring activities in the context of habitats protection and conservation

Multispectral and 3D LiDAR remote sensing data sources are valuable tools for characterizing the 3D vegetation structure and thus understanding the relationship between forest structure, biodiversity and microclimate. This study focuses on mapping riparian forest species in the canopy strata using a fusion of Airborne LiDAR data and multispectral multi-sources and multi-resolution satellite imagery; Sentinel-2 and Pleiades at tree level. The idea is to assess the contribution of each data source in the tree species classification at the considered level. The data fusion was processed at feature-level and decision level. At feature level, LiDAR 2D attributes were derived and combined with vegetation indices. At decision level, LiDAR data was used for 3D tree crown delimitation providing unique trees or groups of trees that are used as a support for the species classification. Data augmentation techniques were used to improve the training process. Best results were obtained by the fusion of Sentinel-2 time series and LiDAR data with a Kappa of 0.66 thanks to red-edge based indices that better discriminate vegetation species and the temporal resolution of Sentinel-2 images that allows monitoring the phenological stages helping to discriminate the species.

Side-scan sonar (SSS) detection is a key method in applications such as underwater environmental security and subsea resource development. The use of acoustic images for seabed target detection has gradually become a mainstream underwater detection method. However, many existing detection approaches primarily concentrate on tracking the evolution path of optical image object detection tasks, resulting in complex structures and limited versatility. To tackle this issue, we introduce a pioneering Dual-Domain Multi-Frequency Network (D2MFNet) meticulously crafted to harness the distinct characteristics of SSS image detection. In D2MFNet, aiming at the underwater detection requirements of small scenes, we introduce a novel method for optimize and improve the detection sensitivity of different frequency ranges and propose a Multi-Frequency Combined Attention Mechanism (MFCAM). This mechanism amplifies the relevance of dual-domain features across different channels and space. Moreover, recognizing that SSS images can provide richer insights after frequency domain conversion, we introduce a Dual-Domain Feature Pyramid Network (D2FPN). By incorporating frequency domain information representation, D2FPN significantly augments the depth and breadth of feature information in underwater small datasets. Our methods are seamlessly designed for integration into existing networks, offering plug-and-play functionality with substantial performance enhancements. We have conducted extensive experiments to validate the efficacy of our proposed techniques, and the results showcase their state-of-the-art performance. MFCAM improves the mAP by 16.9% in the KLSG dataset and 15.5% in the SCTD dataset. The mAP of D2FPN was improved by 8.4% in the KLSG dataset and by 9.8% in the SCTD dataset. We will make our code and models publicly available at https://dagshub.com/estrellaww00/D2MFNet.

This paper presents a system for identification of wind features, such as gusts and wind shear. These are of particular interest in the context of energy-efficient navigation of Small Unmanned Aerial Systems (UAS). The proposed system generates real-time wind vector estimates and a novel algorithm to generate wind field predictions. Estimations are based on the integration of an off-the-shelf navigation system and airspeed readings in a so-called direct approach. Wind predictions use atmospheric models to characterize the wind field with different statistical analyses. During the prediction stage, the system is able to incorporate, in a big-data approach, wind measurements from previous flights in order to enhance the approximations. Wind estimates are classified and fitted into a Weibull probability density function. A Genetic Algorithm (GA) is utilized to determine the shaping and scale parameters of the distribution, which are employed to determine the most probable wind speed at a certain position. The system uses this information to characterize a wind shear or a discrete gust and also utilizes a Gaussian Process regression to characterize continuous gusts. The knowledge of the wind features is crucial for computing energy-efficient trajectories with low cost and payload. Therefore, the system provides a solution that does not require any additional sensors. The system architecture presents a modular decentralized approach, in which the main parts of the system are separated in modules and the exchange of information is managed by a communication handler to enhance upgradeability and maintainability. Validation is done providing preliminary results of both simulations and Software-In-The-Loop testing. Telemetry data collected from real flights, performed in the Seville Metropolitan Area in Andalusia (Spain), was used for testing. Results show that wind estimation and predictions can be calculated at 1 Hz and a wind map can be updated at 0.4 Hz. Predictions show a convergence time with a 95% confidence interval of approximately 30 s.