Search Results (2,783)

Soft sensors based on deep learning regression models are promising approaches to predict real-time fermentation process quality measurements. However, experimental datasets are generally sparse and may contain outliers or corrupted data. This leads to insufficient model prediction performance. Therefore, datasets with a fully distributed solution space are required that enable effective exploration during model training. In this study, the robustness and predictive capability of the underlying model of a soft sensor was improved by generating synthetic datasets for training. The monitoring of intensified ethanol fermentation is used as a case study. Variational autoencoders were employed to create synthetic datasets, which were then combined with original datasets (experimental) to train neural network regression models. These models were tested on original versus augmented datasets to assess prediction improvements. Using the augmented datasets, the soft sensor predictive capability improved by 34%, and variability was reduced by 82%, based on R² scores. The proposed method offers significant time and cost savings for dataset generation for the deep learning modeling of ethanol fermentation and can be easily adapted to other fermentation processes. This work contributes to the advancement of soft sensor technology, providing practical solutions for enhancing reliability and robustness in large-scale production. Full article

Accurately recognizing apples in complex environments is essential for automating apple picking operations, particularly under challenging natural conditions such as cloudy, snowy, foggy, and rainy weather, as well as low-light situations. To overcome the challenges of reduced apple target detection accuracy due to branch occlusion, apple overlap, and variations between near and far field scales, we propose the Rep-ViG-Apple algorithm, an advanced version of the YOLO model. The Rep-ViG-Apple algorithm features a sophisticated architecture designed to enhance apple detection performance in difficult conditions. To improve feature extraction for occluded and overlapped apple targets, we developed the inverted residual multi-scale structural reparameterized feature extraction block (RepIRD Block) within the backbone network. We also integrated the sparse graph attention mechanism (SVGA) to capture global feature information, concentrate attention on apples, and reduce interference from complex environmental features. Moreover, we designed a feature extraction network with a CNN-GCN architecture, termed Rep-Vision-GCN. This network combines the local multi-scale feature extraction capabilities of a convolutional neural network (CNN) with the global modeling strengths of a graph convolutional network (GCN), enhancing the extraction of apple features. The RepConvsBlock module, embedded in the neck network, forms the Rep-FPN-PAN feature fusion network, which improves the recognition of apple targets across various scales, both near and far. Furthermore, we implemented a channel pruning algorithm based on LAMP scores to balance computational efficiency with model accuracy. Experimental results demonstrate that the Rep-ViG-Apple algorithm achieves precision, recall, and average accuracy of 92.5%, 85.0%, and 93.3%, respectively, marking improvements of 1.5%, 1.5%, and 2.0% over YOLOv8n. Additionally, the Rep-ViG-Apple model benefits from a 22% reduction in size, enhancing its efficiency and suitability for deployment in resource-constrained environments while maintaining high accuracy. Full article

We implement an inversion-based deblending method in the common midpoint gathers (CMP) as an alternative to the standard common receiver gather (CRG) domain methods. The primary advantage of deblending in the CMP domain is that reflections from dipping layers are centred around zero offsets. As a result, CMP gathers exhibit a simpler structure compared to common receiver gathers (CRGs), where these reflections are apex-shifted. Consequently, we can employ a zero-offset hyperbolic Radon operator to process CMP gathers. This operator is a computationally more efficient alternative to the apex-shifted hyperbolic Radon required for processing CRG gathers. Sparse transforms, such as the Radon transform, can stack reflections and produce sparse models capable of separating blended sources. We utilize the Radon operator to develop an inversion-based deblending framework that incorporates a sparse model constraint. The inclusion of a sparsity constraint in the inversion process enhances the focusing of the transform and improves data recovery. Inversion-based deblending enables us to account for all observed data by incorporating the blending operator into the cost function. Our synthetic and field data examples demonstrate that inversion-based deblending in the CMP domain can effectively separate blended sources. Full article

In isolated operation, DC microgrids require multiple distributed energy storage units (DESUs) to accommodate the variability of distributed generation (DG). The traditional control strategy has the problem of uneven allocation of load current when the line impedance is not matched. As the state-of-charge (SOC) balancing proceeds, the SOC difference gradually decreases, leading to a gradual decrease in the balancing rate. Thus, an improved SOC droop control strategy is introduced in this paper, which uses a combination of power and exponential functions to improve the virtual impedance responsiveness to SOC changes and introduces an adaptive acceleration factor to improve the slow SOC balancing problem. We construct a sparse communication network to achieve information exchange between DESU neighboring units. A global optimization controller employing the consistency algorithm is designed to mitigate the impact of line impedance mismatch on SOC balancing and current allocation. This approach uses a single controller to restore DC bus voltage, effectively reducing control connections and alleviating the communication burden on the system. Lastly, a simulation model of the DC microgrid is developed using MATLAB/Simulink R2021b. The results confirm that the proposed control strategy achieves rapid SOC balancing and the precise allocation of load currents in various complex operational scenarios. Full article

In this paper, we propose a Local Global Union Network (LGUN), which effectively combines the strengths of Transformers and Convolutional Networks to develop a lightweight and high-performance network suitable for Single Image Super-Resolution (SISR). Specifically, we make use of the advantages of Transformers to provide input-adaptation weighting and global context interaction. We also make use of the advantages of Convolutional Networks to include spatial inductive biases and local connectivity. In the shallow layer, the local spatial information is encoded by Multi-order Local Hierarchical Attention (MLHA). In the deeper layer, we utilize Dynamic Global Sparse Attention (DGSA), which is based on the Multi-stage Token Selection (MTS) strategy to model global context dependencies. Moreover, we also conduct extensive experiments on both natural and satellite datasets, acquired through optical and satellite sensors, respectively, demonstrating that LGUN outperforms existing methods. Full article

Precise prediction of wireless communication network traffic is indispensable in the operational deployment of base station resources and improvement of the user experience. Cellular wireless network traffic has both spatial and temporal characteristics. The existing modeling algorithms have achieved good results in extracting the spatial features, but there are still deficiencies in the extraction models for the time dependencies. To resolve these problems, this paper proposes a novel hybrid neural network prediction model, called WVETT-Net. Firstly, variational mode decomposition (VMD) is used to preprocess network traffic, and the whale optimization algorithm (WOA) is used to select the optimal parameters for VMD. Secondly, the local and global features are extracted from each subsequence by a temporal convolutional network (TCN) and an improved Transformer network with a multi-head ProbSparse self-attention mechanism (Pe-Transformer), respectively. Finally, the extracted feature representation is enhanced by using an efficient channel attention (ECA) mechanism to achieve accurate wireless network traffic predictions. Experimental results on two wireless network traffic datasets show that the proposed model (WVETT-Net) outperforms the traditional single or combined models in wireless network traffic prediction. Full article

Incorporating silver wires into propellant has emerged as a highly effective strategy for enhancing propellant burning rates, a technique extensively deployed in the construction of numerous fielded sounding rockets and tactical missiles. Our research, employing a multi-faceted approach encompassing thermogravimetric-differential scanning calorimetry measurements (TG-DSC), combustion diagnoses, burning rate tests, and meticulous collection of condensed combustion products, sought to elucidate how variations in silver wire quantity and winding configuration impact the combustion properties of propellants. Our findings underscore the remarkable efficacy of double tightly twisted silver wire in significantly boosting propellant burning rates under ambient conditions. Moreover, at lower temperatures, the reduced gap between the propellant and silver wire further magnifies the influence of silver wire on burning rates. However, it is noteworthy that the relationship between burning speed and combustion efficiency is not deterministic. While a smaller cone angle of the burning surface contributes to heightened burning rates, it concurrently exacerbates the polymerization effect of vapor phase aluminum particles, consequently diminishing propellant combustion efficiency. Conversely, propellants configured with sparsely twinned silver wires exhibit notable enhancements in combustion efficiency, despite a less pronounced impact on the burning rate attributed to the larger cone angle of the burning surface. Remarkably, these trends persist at lower temperatures. Based on the principle of heat transfer balance, a theoretical model for the combustion of propellants with wire inserts is developed. The reliability of this theoretical model is validated through a comparison of calculated values with experimental data. Our research outcomes carry significant implications for guiding the application and advancement of the silver wire method in solid propellants for solid rocket motors, offering valuable insights to inform future research and development endeavors in this domain. Full article

Stress has various impacts on the health of human beings. Recent success in wearable sensor development, combined with advancements in deep learning to automatically detect features from raw data, opens several interesting applications related to detecting emotional states. Being able to accurately detect stress-related emotional arousal in an acute setting can positively impact the imminent health status of humans, i.e., through avoiding dangerous locations in an urban traffic setting. This work proposes an explainable deep learning methodology for the automatic detection of stress in physiological sensor data, recorded through a non-invasive wearable sensor device, the Empatica E4 wristband. We propose a Long-Short Term-Memory (LSTM) network, extended through a Deep Generative Ensemble of conditional GANs (LSTM DGE), to deal with the low data regime of sparsely labeled sensor measurements. As explainability is often a main concern of deep learning models, we leverage Integrated Gradients (IG) to highlight the most essential features used by the model for prediction and to compare the results to state-of-the-art expert-based stress-detection methodologies in terms of precision, recall, and interpretability. The results show that our LSTM DGE outperforms the state-of-the-art algorithm by 3 percentage points in terms of recall, and 7.18 percentage points in terms of precision. More importantly, through the use of Integrated Gradients as a layer of explainability, we show that there is a strong overlap between model-derived stress features for electrodermal activity and existing literature, which current state-of-the-art stress detection systems in medical research and psychology are based on. Full article

The point clouds obtained directly from three-dimensional scanning devices are often sparse and noisy. Therefore, point cloud upsampling plays an increasingly crucial role in fields such as point cloud reconstruction and rendering. However, point cloud upsampling methods are primarily supervised and fixed-rate, which restricts their applicability in various scenarios. In this paper, we propose a novel point cloud upsampling method, named RE-PU, which is based on the point cloud reconstruction and achieves self-supervised upsampling at arbitrary rates. The proposed method consists of two main stages: the first stage is to train a network to reconstruct the original point cloud from a prior distribution, and the second stage is to upsample the point cloud data by increasing the number of sampled points on the prior distribution with the trained model. The experimental results demonstrate that the proposed method can achieve comparable outcomes to supervised methods in terms of both visual quality and quantitative metrics. Full article

In Indonesia, the monitoring of rainfall requires an estimation system with a high resolution and wide spatial coverage because of the complexities of the rainfall patterns. This study built a rainfall estimation model for Indonesia through the integration of data from various instruments, namely, rain gauges, weather radars, and weather satellites. An ensemble learning technique, specifically, extreme gradient boosting (XGBoost), was applied to overcome the sparse data due to the limited number of rain gauge points, limited weather radar coverage, and imbalanced rain data. The model includes bias correction of the satellite data to increase the estimation accuracy. In addition, the data from several weather radars installed in Indonesia were also combined. This research handled rainfall estimates in various rain patterns in Indonesia, such as seasonal, equatorial, and local patterns, with a high temporal resolution, close to real time. The validation was carried out at six points, namely, Bandar Lampung, Banjarmasin, Pontianak, Deli Serdang, Gorontalo, and Biak. The research results show good estimation accuracy, with respective values of 0.89, 0.91, 0.89, 0.9, 0.92, and 0.9, and root mean square error (RMSE) values of 2.75 mm/h, 2.57 mm/h, 3.08 mm/h, 2.64 mm/h, 1.85 mm/h, and 2.48 mm/h. Our research highlights the potential of this model to accurately capture diverse rainfall patterns in Indonesia at high spatial and temporal scales. Full article

Classifieds pose numerous challenges for recommendation methods, including the temporary visibility of ads, the anonymity of most users, and the fact that typically only one user can consume an advertised item. In this work, we address these challenges by choosing models and evaluation procedures that are considered accurate, diverse, and efficient (in terms of memory and time consumption during training and prediction). This paper aims to benchmark various recommendation methods for job classifieds, using OLX Jobs as an example, to enhance the conversion rate of advertisements and user satisfaction. In our research, we implement scalable methods and represent different approaches to the recommendations: Alternating Least Square (ALS), LightFM, Prod2Vec, RP3Beta, and Sparse Linear Methods (SLIM). We conducted A/B tests by sending millions of messages with recommendations to perform online evaluations of selected methods. In addition, we have published the dataset created for our research. To the best of our knowledge, this is the first dataset of its kind. It contains 65,502,201 events performed on OLX Jobs by 3,295,942 users who interacted with (displayed, replied to, or bookmarked) 185,395 job ads over two weeks in 2020. We demonstrate that RP3Beta, SLIM, and ALS perform significantly better than Prod2Vec and LightFM when tested in a laboratory setting. Online A/B tests also show that sending messages with recommendations generated by the ALS and RP3Beta models increases the number of users contacting advertisers. Additionally, RP3Beta had a 20% more significant impact on this metric than ALS. Full article

In this paper, we present a novel sparse Bayesian learning (SBL) method for image reconstruction. We integrate the generalized approximate message passing (GAMP) algorithm and Laplacian hierarchical priors (LHP) into a basic SBL model (called LHP-GAMP-SBL) to improve the reconstruction efficiency. In our SBL model, the GAMP structure is used to estimate the mean and variance without matrix inversion in the E-step, while LHP is used to update the hyperparameters in the M-step.The combination of these two structures further deepens the hierarchical structures of the model. The representation ability of the model is enhanced so that the reconstruction accuracy can be improved. Moreover, the introduction of LHP accelerates the convergence of GAMP, which shortens the reconstruction time of the model. Experimental results verify the effectiveness of our method. Full article

The Auckland Harbour Bridge (AHB) utilises a movable concrete barrier (MCB) to regulate the uneven bidirectional flow of daily traffic. In addition to the risk of human error during regular visual inspections, staff members inspecting the MCB work in diverse weather and light conditions, exerting themselves in ergonomically unhealthy inspection postures with the added weight of protection gear to mitigate risks, e.g., flying debris. To augment visual inspections of an MCB using computer vision technology, this study introduces a hybrid deep learning solution that combines kernel manipulation with custom transfer learning strategies. The video data recordings were captured in diverse light and weather conditions (under the safety supervision of industry experts) involving a high-speed (120 fps) camera system attached to an MCB transfer vehicle. Before identifying a safety hazard, e.g., the unsafe position of a pin connecting two 750 kg concrete segments of the MCB, a multi-stage preprocessing of the spatiotemporal region of interest (ROI) involves a rolling window before identifying the video frames containing diagnostic information. This study utilises the ResNet-50 architecture, enhanced with 3D convolutions, within the STENet framework to capture and analyse spatiotemporal data, facilitating real-time surveillance of the Auckland Harbour Bridge (AHB). Considering the sparse nature of safety anomalies, the initial peer-reviewed binary classification results (82.6%) for safe and unsafe (intervention-required) scenarios were improved to 93.6% by incorporating synthetic data, expert feedback, and retraining the model. This adaptation allowed for the optimised detection of false positives and false negatives. In the future, we aim to extend anomaly detection methods to various infrastructure inspections, enhancing urban resilience, transport efficiency and safety. Full article

Vehicle classification technology is one of the foundations in the field of automatic driving. With the development of deep learning technology, visual transformer structures based on attention mechanisms can represent global information quickly and effectively. However, due to direct image segmentation, local feature details and information will be lost. To solve this problem, we propose an improved vision transformer vehicle classification network (IND-ViT). Specifically, we first design a CNN-In D branch module to extract local features before image segmentation to make up for the loss of detail information in the vision transformer. Then, in order to solve the problem of misdetection caused by the large similarity of some vehicles, we propose a sparse attention module, which can screen out the discernible regions in the image and further improve the detailed feature representation ability of the model. Finally, this paper uses the contrast loss function to further increase the intra-class consistency and inter-class difference of classification features and improve the accuracy of vehicle classification recognition. Experimental results show that the accuracy of the proposed model on the datasets of vehicle classification BIT-Vehicles, CIFAR-10, Oxford Flower-102, and Caltech-101 is higher than that of the original vision transformer model. Respectively, it increased by 1.3%, 1.21%, 7.54%, and 3.60%; at the same time, it also met a certain real-time requirement to achieve a balance of accuracy and real time. Full article

Iron in soil exists in various valence states and is prone to changes with alterations in soil environmental conditions. Its migration and transformation are crucial for soil formation and understanding soil evolution. This study focuses on Quaternary red soils found in woodland, sparse forest grassland, grassland, and cultivated land located in the semi-humid region of the middle temperate zone. For comparison, buried Quaternary red soil was also examined. A soil reconstruction model was used to quantitatively calculate the variation of different forms of iron in order to analyze various forms of iron composition, migration, and transformation within the soil profile, as well as the evolutionary traits of Quaternary red soils influenced by diverse land use activities. This study found that after exposure and use, iron from the topsoil of buried Quaternary red soil migrated to the subsoil, altering the iron distribution. Free iron and crystalline oxides decreased in the topsoil but increased in specific subsoil layers, with woodland and grassland showing the most significant changes. Silicate-bound iron pooled in the soil weathered to form free iron under different land uses, and poorly crystalline iron oxides transformed into crystalline oxides, with grassland exhibiting the highest transformation intensity. Conversion processes predominated over iron migration in the Quaternary red soils. The evolution of Quaternary red soils can be divided into three stages, marked by changes in iron composition and crystallization due to anthropogenic land use activities. Initially, during 140−94 ka BP, iron composition was stable. Then, between 94–24 ka BP, plant decomposition formed iron–metal complexes, releasing and crystallizing poorly crystalline iron oxides. Finally, from 24 ka BP to the present, anthropogenic activities intensified, increasing the formation and conversion rates of these oxides. This study quantifies iron migration and transformation in Quaternary red soils, providing insights for sustainable soil management, especially in regions where human activities have accelerated iron evolution. Based on these findings, the following policy recommendations are proposed: implement sustainable land use practices, encourage land management strategies that preserve natural vegetation, promote research on soil management techniques, develop and implement regulatory policies, and support educational programs to maintain the health and stability of Quaternary red soils, particularly in regions prone to accelerated iron evolution due to anthropogenic activities. Full article