Honey is a sweet substance extensively consumed in the world for its nutritional and healthy properties. However, residues of pesticides and environmental contaminants can compromise its quality. For this reason, the physicochemical parameters, and the organic contamination of monofloral and multifloral honeys from three regions of Algeria (Tiaret, Laghouat and Tindouf) were monitored to evaluate the quality of honey and the safety for consumers. In general, the results obtained from the physicochemical analyses were in line with the EU standards. In terms of contamination, pesticides authorised and used in Algerian agriculture (matalaxyl-M and cyromazine), as well as banned pesticide (carbaryl), were found in almost all the samples. However, only the concentration of cyromazine was higher than the relative EU maximum residue levels. PCB 180, PCB 189, anthracene, fluorene and phenanthrene were mainly detected. All the honeys show traces of DiBP, DBP, DEHP and DEHT, but no traces of bisphenols were found. Moreover, the dietary exposure assessment shows that a small amount of Algerian honey can be safely consumed. Overall, data from this study should motivate the Algerian government to enhance their monitoring activities in beekeeping and to find solutions for implementing more sustainable agricultural practices harmonizing with the international legislation.

Carotid atherosclerotic lesions are a major cause of cerebrovascular disease (CVD). Identification and quantification of carotid plaques are important for categorizing the vulnerability of plaques for rupture and assessing the impact of treatments. The irregularity of plaque surface is associated with previous plaque rupture and plays an important role in the risk of stroke. Thus, the aim of this study is to develop and validate novel vulnerability biomarkers from three-dimensional ultrasound (3DUS) images by analyzing the surface morphological characterization of carotid plaque using fractal geometry features. 3D box-counting and 3D blanket are the two types of 3D fractal dimension that were employed to describe the smoothness of plaques. This fractal dimension analysis tool was used to evaluate the effect of atorvastatin using 3DUS carotid images, which were acquired from 6 patients treated with atorvastatin with 80 mg daily and 5 patients with placebo. The Student's T Test results showed that those two fractal features were effective for detecting the statin-related changes in carotid atherosclerosis with p<0.0068 and p<0.015 respectively, suggesting that 3D fractal dimension measurements can be used effectively to analyze the surface characteristics of carotid plaques, especially for evaluating the impact of the treatment.

Cloud-induced atmospheric extinction and occlusion have a major effect on the effectiveness and quality of telescope observations. Real-time cloud cover distribution and long-term statistical data are essential for astronomical siting and telescope operations. At ground-based astronomical telescope sites, cloud cover distribution is currently analyzed using manual observation methods. However, the main disadvantages of manual observation methods are human subjective, heavy workloads and poor real-time performance. Therefore, a real-time automatic cloud images classification method is desperately needed. This paper presents a new cloud identification method, which is named PSO+XGBoost model by combining eXtreme Gradient Boosting (XGBoost) and Particle Swarm Optimization (PSO). The entire cloud image is divided into 37 sub-regions in order to more precisely identify the distribution of the clouds. 19 features are then extracted, including the sky background, star density, lighting conditions, and subregion grayscale values. The experimental results have shown that the overall classification accuracy of 96.91% and our model is able to outperform several state-of-the-art baseline methods. Our approach achieves high accuracy in comparison with the manual observation methods. Moreover, this method meets telescope real-time scheduling requirements.

ABSTRACT Purpose: Tongue image analysis for disease diagnosis is an ancient traditional non-invasive diagnosis technique widely used by traditional medicine practitioners. Deep Learning based multi-label disease detection models offer tremendous potential to clinical decision support systems, by facilitating preliminary diagnosis. Methods: In this work, we propose a multi-label disease detection pipeline, where in observation and analysis of tongue images captured and received via smartphones assist in predicting the health status of an individual. All images are voluntarily given by subjects consulting collaborating physicians. Images thus acquired are first and foremost classified either into a diseased or a normal category by a 5-fold cross-validation algorithm using a convolution neural network (MobileNetV2) model for binary classification. Once the diseased label is predicted, the image is used to diagnose multiple diseases using the prediction algorithm based on DenseNet121. Results: Average accuracy of 93 % was achieved in classifying diseased from normal healthy tongue by detection model with MobileNetV2 architecture. Multilabel disease classification produced more than 90% accurate results for the seven class labels considered. Conclusion: AI based image analysis shows promising results and an extensive dataset could provide further improvements to this approach. Rather than employing high-cost sophisticated image capturing setup, experimenting with smartphone images opens opportunity to provide preliminary health status to individuals on smartphone prior to further line of treatment and diagnosis.

Image stitching is a crucial aspect of image processing. However, factors like perspective and environment often lead to irregular shapes in stitched images. Cropping or completion methods typically result in substantial loss of information. This paper proposes a method for rectifying irregularly images into rectangles using deformable meshes and residual networks. The method utilizes a convolutional neural network to quantify rigid structures of images. Choosing the most suitable mesh structure based on the extraction results, offering options such as triangular, rectangular, and hexagonal. Subsequently, the irregularly image, predefined mesh structure, and predicted mesh structure are input into a wide residual neural network for regression. The loss function comprises local and global, aimed at minimizing the loss of image information within the mesh and global target. This method not only significantly reduces information loss during rectification but also adapting to different images with various rigid structures. Validation on the DIR-D dataset shows this method outperforms state-of-the-art methods in image rectification.

Pavement crack detection is one of the key links in highway pavement maintenance management. In the current road crack detection, the network model easily ignores the shallow geometric features of cracks and cannot extract the key feature information of cracks. Aiming at the above problems, an EC-YOLOX network model is proposed. The CFPN is constructed to cross-fuse different scale features to solve the feature scale invariance. In the strengthen feature extraction layer, ECANet is fused to enhance the ability to pay attention to key information. The WEIoU loss function is proposed, which assigns different penalty terms to the target box in the vertical and horizontal directions. In terms of data, geometric distortion constraint correction is performed on the single-point perspective pavement image. The experimental data show that the accuracy of EC-YOLOX on the self-built data set reaches 76.34% [email protected], which is 4.45% higher than that of YOLOX. The loss curve of EC-YOLOX is smoother than that of YOLOX, and the minimum training loss value is 2.136, which is 0.648 lower than the minimum loss value 2.784 of YOLOX. Many experiments have verified the effectiveness of EC-YOLOX in improving the detection effect in pavement crack detection.

Reservoirs created by damming rivers significantly modify the environment and problems may arise: the storage of nutrients and organic matter in reservoirs often leads to decreased water quality due to eutrophication. Reservoirs built on lowland rivers are exposed to significant nutrient loads, often leading to eutrophication. In this article, the Sulejów Reservoir, in Central Poland, was the subject of the research, which focused on the use of Sentinel-2 remote sensing data to detect phytoplankton blooms, and on understanding the pattern of wind-driven surface currents using the CCHE2D model (a depth-averaged hydrodynamic model). The calculation of the total phosphorus load has shown that the reservoir mainly acts as nutrient storage. Still, during the low flow and intensive phytoplankton bloom, it can be a source of nutrients. The distribution of phytoplankton bloom on the Sulejów Reservoir was documented using eight Sentinel-2 satellite images from the vegetation season of 2020 and the Normalized Difference Chlorophyll Index (NDCl). Coupling remote sensing data and numerical modelling helps to interpret the hydrodynamic model results and understand nutrients and sediment dynamics within the reservoir. Hydrodynamic modelling combined with remote sensing data can be a valuable tool for selecting appropriate locations for water quality measurements at monitoring stations.

The application of artificial intelligence to point-of-care testing (POCT) disease detection has become a hot research field, in which breath detection which detects the patient's exhaled VOCs combined with sensor arrays of Convolutional Neural Network (CNN) algorithms as a new lung cancer detection is attracting more researchers' attention. However, due to low accuracy, high complexity computation and large number of parameters make the CNN algorithms difficult transplant to the embedded system of POCT devices. A lightweight neural network (LTNet) in this work is proposed to deal with this problem, and meanwhile, achieve high-precision classification of acetone and ethanol gases, which are respiratory markers for lung cancer patients. Compared to currently popular lightweight CNN models such as EfficientNet, LTNet has fewer parameters (32K) and its training weight size is only 0.155 MB. LTNet achieved an overall classification accuracy of 99.06% and 99.14% in the own mixed gas dataset and the University of California (UCI) dataset, which are both higher than that of the five existing models’, and it also offers the shortest training (844.38s and 584.67s) and inference time (23s and 14s) in the same validation set. Compared to the existing CNN models, LTNet is more suitable for resource-limited POCT devices.

Surface water distribution extracted from remote sensing data has been used in water resource assessment, coastal management, and environmental change studies. Traditional manual methods for extracting water bodies cannot satisfy the requirements for mass processing of remote sensing data; therefore, accurate automated extraction of such water bodies has remained a challenge. The histogram bimodal method (HBM) is a frequently used objective tool for threshold selection in image segmentation. The threshold is determined by seeking twin peaks, and the valley values between them; however, automatically calculating the threshold is difficult because complex surfaces and image noise which lead to not perfect twin peaks (single or multiple peaks). We developed an operational automated water extraction method, the modified histogram bimodal method (MHBM). The MHBM defines the threshold range of water extraction through mass static data; therefore, it does not require the identification of twin histogram peaks. It then seeks the minimum values in the threshold range to achieve automated threshold. We calibrated the MHBM for many lakes in China using Landsat 8 Operational Land Imager (OLI) images, for which the relative error (RE) and squared correlation coefficient (R2) for threshold accuracy were found to be 2.1% and 0.96, respectively. The RE and root-mean-square error (RMSE) for the area accuracy of MHBM were 0.59% and 7.4 km2. The results show that the MHBM could easily be applied to mass time-series remote sensing data to calculate water thresholds within water index images and successfully extract the spatial distribution of large water bodies automatically.

For multi-sensor integrated systems, such as a mobile mapping system (MMS), data fusion at sensor-level, i.e., the 2D-3D registration between optical camera and LiDAR, is a prerequisite for higher level fusion and further applications. This paper proposes a line-based registration method for panoramic images and LiDAR point cloud collected by a MMS. We first introduce the system configuration and specification, including the coordinate systems of the MMS, the 3D LiDAR scanners, and the two panoramic camera models. We then establish the line-based transformation model for panoramic camera. Finally, the proposed registration method is evaluated for two types of camera models by visual inspection and quantitative comparison. The results demonstrate that the line-based registration method can significantly improve the alignment of the panoramic image and LiDAR datasets under either the ideal spherical or the rigorous panoramic camera model, though the latter is more reliable.

This study introduces SkinLiTE, a lightweight supervised contrastive learning model, tailored to enhance the detection and typification of skin lesions in dermoscopic images. The core of SkinLiTE lies in its unique integration of supervised and contrastive learning approaches, which leverages labeled data to learn generalizable representations. This approach is particularly adept at handling the complexities and imbalances inherent in skin lesions datasets. The methodology encompasses a two-phase learning process. In the first phase, SkinLiTE utilizes an encoder network and a projection head to transform and project dermoscopic images into a feature space where contrastive loss is applied, focusing on minimizing intra-class variations while maximizing inter-class differences. The second phase freezes the encoder's weights, leveraging the learned representations for classification through a series of dense and dropout layers. The model was evaluated using three datasets from Skin Cancer ISIC 2019-2020, covering a wide range of skin conditions. SkinLiTE demonstrated superior performance across various metrics—accuracy, AUC , and F1 score—particularly when compared with traditional supervised learning models. Notably, SkinLiTE achieved an accuracy of 0.9087 using AugMix augmentation for binary classification of skin lesions. It also showed comparable results with the state-of-the-art approaches of ISIC challenge without relying on external data, underscoring its efficacy and efficiency. The results highlight the potential of SkinLiTE as a significant step forward in the field of dermatological AI, offering a robust, efficient, and accurate tool for skin lesion detection and classification. Its lightweight architecture and ability to handle imbalanced datasets make it particularly suited for integration into Internet of Medical Things environments, paving the way for enhanced remote patient monitoring and diagnostic capabilities. This research contributes to the evolving landscape of AI in healthcare, demonstrating the impact of innovative learning methodologies in medical image analysis.

Semantic segmentation of high-resolution remote sensing images provides detailed and precise feature information, playing a pivotal role in decision-making and analysis in sectors such as water management, agriculture, military, and environmental protection. However, most methods merely combine features from various branches directly, without a mechanism for spatial location feature screening, and treat all extracted features as equally important. To overcome these limitations, we introduce a novel spatially adaptive interaction network (SAINet) for dynamic interaction across different features in remote sensing semantic segmentation task. Specifically, we propose a spatial refined module that leverages local context information to filter spatial locations and extract salient regions. Following this, we introduce an innovative spatial interaction module that uses a spatial adaptive modulation mechanism to dynamically select and allocate spatial position weights. This facilitates interaction between local salient areas and global information, allowing the network to focus more on relevant regions. This flexible approach enables the network to capture more informative features, thereby enhancing segmentation accuracy. Experiments conducted on the DeepGlobe, Vaihingen, and Potsdam public datasets confirm the effectiveness and capability of SAINet. Our code and models will be publicly available

The advent of deep learning (DL) has revolutionized medical imaging, offering unprecedented avenues for accurate disease classification and diagnosis. DL models have shown remarkable promise for classifying brain tumor from Magnetic Resonance Imaging (MRI) scans. However, despite their impressive performance, the opaque nature of DL models poses challenges in understanding their decision-making processes, particularly crucial in medical contexts where interpretability is essential. This paper explores the intersection of medical image analysis and DL interpretability, aiming to elucidate the decision-making rationale of DL models in brain tumor classification. Leveraging state-of-the-art DL frameworks with transfer learning, we conduct a comprehensive evaluation encompassing both classification accuracy and interpretability. Using state-of-the-art DL frameworks with transfer learning, we conduct a thorough evaluation covering both classification accuracy and interpretability. We employ adaptive path-based techniques to understand the underlying decision-making mechanisms of these models. Grad-CAM and Grad-CAM++ highlight critical image regions where the tumors are located.

During the last 25 years the scientific community has coexisted with the most fascinating protocol due to Quantum Physics: quantum teleportation (QTele), which would have been impossible if quantum entanglement, so questioned by Einstein, did not exist. In this work, a complete architecture for the teleportation of Computational Basis States (CBS) is presented. Such CBS will represent each of the possible 24 classical bits commonly used to encode every pixel of a 3-color-channel-image (red-green-blue, or cyan-yellow-magenta). For this purpose, a couple of interfaces: classical-to-quantum (Cl2Qu) and quantum-to-classical (Qu2Cl) are presented with two versions of the teleportation protocol: standard and simplified.

Sentinel-2 can provide multi-spectral optical remote sensing images in RGBN bands with a spatial resolution of 10m, but the spatial details provided are not enough for many applications. WorldView can provide HR multi-spectral images less than 2m, but it is a commercial paid resource with relatively high usage costs. In this paper, without any available reference images, Sentinel-2 images at 10m resolution are improved to a resolution of 2.5m through super-resolution (SR) based on deep learning technology. Our model, named DKN-SR-GAN, uses degradation kernel estimation and noise injection to construct a dataset of near-natural low-high-resolution (LHR) image pairs, with only low-resolution (LR) images and no high-resolution (HR) prior information. DKN-SR-GAN uses the Generative Adversarial Networks (GAN) combined of ESRGAN-type generator, PatchGAN-type discriminator and the VGG-19-type feature extractor, using perceptual loss to optimize the network, so as to obtain SR images with clearer details and better perceptual effects. Experiments demonstrate that in the quantitative comparison of the non-reference image quality assessment (NR-IQA) metrics like NIQE, BRISQUE and PIQE, as well as the intuitive visual effects of the generated images, compared with state-of-the-art models such as EDSR8-RGB, RCAN and RS-ESRGAN, our proposed model has obvious advantages.

Medical brain image analysis is a necessary step in the Computers Assisted /Aided Diagnosis (CAD) systems. Advancements in both hardware and software in the past few years have led to improved segmentation and classification of various diseases. In the present work, we review the published literature on systems and algorithms that allow for classification, identification, and detection of White Matter Hyperintensities (WMHs) of brain MRI images specifically in cases of ischemic stroke and demyelinating diseases. For the selection criteria, we used the bibliometric networks. Out of a total of 140 documents we selected 38 articles that deal with the main objectives of this study. Based on the analysis and discussion of the revised documents, there is constant growth in the research and proposal of new models of deep learning to achieve the highest accuracy and reliability of the segmentation of ischemic and demyelinating lesions. Models with indicators (Dice Score, DSC: 0.99) were found, however with little practical application due to the uses of small datasets and lack of reproducibility. Therefore, the main conclusion is to establish multidisciplinary research groups to overcome the gap between CAD developments and their complete utilization in the clinical environment.

This paper provides orthogonal moments (OM) such as, Zernike Moments(ZM), Psuedo Zernike Moments(PZM) and Orthogonal Fourier Mellin Moments(OFMM) for the analysis of melanoma images. The moment invariants may vary with respect to geometric variations. For the analysis of orthogonal moments hundred random melanoma images and hundred non-melanoma images have been taken into consideration from the database of 570 melanoma images and 250 non-melanoma images respectively. Orthoganal moments have been computed by varying the phase angles from 10° to 40° with an equal interval of 10° degree for the orders 2, 4,8,16,32,64,128,256 respectively. For the optimal OMs Particle Swarm Optimization (PSO) technique have been used. These set of extracted optimal OMs have been further applied to classify melanoma images. Support Vector Machine (SVM) has been used for the classification of [1]sensitivity=88.78%.

The Fractal Analysis of Medical Images is an investigation field in those that numerous results of important value have been obtained. This work describes the basic foundations of the Fractal Analysis and its application has more than enough medical images, leaning on in five scientific investigations.

An image consisting of (m, n) pixels can be seen as a matrix of m × n size. Based on the formula of angles between two subspaces, a pair of angles can be defined between two matrices, by utilizing column spaces and row spaces of the two matrices. The singular values of each matrix can be used to calculate distances. Thus, a distance representation is obtained, and a pair of angles between 2 matrices of the same size.

Hyperspectral imaging provides vital information about the objects and elements present inside the image. That’s why they are very useful in satellite imagery as well as image forensics. Hyperspectral document analysis (HSDI) can be used for document authentication using ink analysis which can provide sufficient information about the composition and type of ink. In this project, we have implemented HSDI based ink classification technique using Principle Component Analysis for dimensionality reduction and K-means clustering for ink classification. This is unsupervised learning approach and it is very simple and efficient in order to classify limited number of bands. We have used this technique to classify 33 different bands of ink.

Endoscopic medical images can suffer from uneven illumination, low contrast, and lack of texture information due to the use of point directional light sources and the presence of narrow tissue structures, posing diagnostic difficulties for physicians. In this paper, a deep learning-based su-pervised illumination enhancement network is designed for low-light endoscopic images, aiming to improve both global illumination and local details. Initially, a global illumination enhancement module is formulated utilizing a higher-order curve function to improve global illumination. Sec-ondly, a local feature extraction module incorporating dual attention is designed to capture local detailed features. Considering the significance of color fidelity in biomedical scenarios, the designed loss function prioritizes introducing color loss to alleviate image color distortion. Compared with seven state-of-the-art enhancement algorithms on Endo4IE endoscopic datasets, experimental re-sults show that the proposed method can better enhance low-light endoscopic images and avoid image color distortion. It provides an efficient method to enhance images captured by endoscopes which can effectively assist clinical diagnosis and treatment.

The evolution in imaging technologies and artificial intelligence algorithms, coupled with improvements in UAV technology, has enabled the use of unmanned aircraft in a wide range of applications. The feasibility of this kind of approach for cattle monitoring has been demonstrated by several studies, but practical use is still challenging due to the particular characteristics of this application, such as the need to track mobile targets and the extensive areas that need to be covered in most cases. The objective of this study was to investigate the feasibility of using a tilted angle to increase the area covered by each image. Deep Convolutional Neural Networks (Xception architecture) were used to generate the models for the experiments, which covered aspects like ideal input dimensions, effect of the distance between animals and sensor, effect of classification error on the overall detection process, and impact of physical obstacles on the accuracy of the model. Experimental results indicate that oblique images can be successfully used under certain conditions, but some practical limitations need to be addressed in order to make this approach appealing.

Event-based vision is an emerging field of computer vision that offers unique properties such as asynchronous visual output, high temporal resolutions, and dependence on brightness changes to generate data. These properties can enable robust high-temporal-resolution object detection and tracking when combined with frame-based vision. In this paper, we present a hybrid, high-temporal-resolution, object detection and tracking approach, that combines learned and classical methods using synchronized images and event data. Off-the-shelf frame-based object detectors are used for initial object detection and classification. Then, event masks, generated per each detection, are used to enable inter-frame tracking at varying temporal resolutions using the event data. Detections are associated across time using a simple low-cost association metric. Moreover, we collect and label a traffic dataset using the hybrid sensor DAVIS 240c. This dataset is utilized for quantitative evaluation using state-of-the-art detection and tracking metrics. We provide ground truth bounding boxes and object IDs for each vehicle annotation. Further, we generate high-temporal-resolution ground truth data to analyze the tracking performance at different temporal rates. Our approach shows promising results with minimal performance deterioration at higher temporal resolutions (48 – 384 Hz) when compared with the baseline frame-based performance at 24 Hz.

Taekwondo has evolved from a traditional martial art into an official Olympic sport. This study introduces a novel action recognition model tailored for Taekwondo unit actions, utilizing joint motion data acquired via wearable inertial measurement unit (IMU) sensors. The utilization of IMU sensor-measured motion data facilitates the capture of the intricate and rapid movements characteristic of Taekwondo techniques. The model, underpinned by a conventional convolutional neural network (CNN)-based image classification framework, synthesizes action images to represent individual Taekwondo unit actions. These action images are generated by mapping joint motion profiles onto the RGB color space, thus encapsulating the motion dynamics of a single unit action within a solitary image. To further refine the representation of rapid movements within these images, a time-warping technique was applied, adjusting motion profiles in relation to the velocity of the action. The effectiveness of the proposed model was assessed using a dataset compiled from 40 Taekwondo experts, yielding remarkable outcomes: an accuracy of 0.998, precision of 0.983, recall of 0.982, and an F1-score of 0.982. These results underscore the time-warping technique's contribution to enhancing feature representation, as well as the proposed method's scalability and effectiveness in recognizing Taekwondo unit actions.

The visualization of hyperspectral images in display devices, having RGB colour composition channels is quite difficult due to the high dimensionality of these images. Thus, principal component analysis has been used as a dimensionality reduction algorithm to reduce information loss, by creating uncorrelated features. To classify regions in the hyperspectral images, K-means clustering has been used to form clusters/regions. These two algorithms have been implemented on the three datasets imaged by AVIRIS and ROSIS sensors.

Recent state-of-the-art scene text recognition methods have primarily focused on horizontal text in images. However, in several Asian countries, including China, large amounts of text in signs, books, and TV commercials are vertically directed. Because the horizontal and vertical texts exhibit different characteristics, developing an algorithm that can simultaneously recognize both types of text in real environments is necessary. To address this problem, we adopted the direction encoding mask (DEM) and selective attention network (SAN) methods based on supervised learning. DEM contains directional information to compensate in cases that lack text direction; therefore, our network is trained using this information to handle the vertical text. The SAN method is designed to work individually for both types of text. To train the network to recognize both types of text and to evaluate the effectiveness of the designed model, we prepared a new synthetic vertical text dataset and collected an actual vertical text dataset (VTD142) from the Web. Using these datasets, we proved that our proposed model can accurately recognize both vertical and horizontal text and can achieve state-of-the-art results in experiments using benchmark datasets, including the street view test (SVT), IIIT-5k, and ICDAR. Although our model is relatively simple as compared to its predecessors, it maintains the accuracy and is trained in an end-to-end manner.

This article presents an automated vision-based algorithm for the die-scale inspection of wafer images captured using scanning acoustic tomography (SAT). This algorithm can find defective and abnormal die-scale patterns, and produce a wafer map to visualize the distribution of defects and anomalies on the wafer. The main procedures include standard template extraction, die detection through template matching, pattern candidate prediction through clustering, and pattern classification through deep learning. To conduct the template matching, we first introduce a two-step method to obtain a standard template from the original SAT image. Subsequently, a majority of the die patterns are detected through template matching. Thereafter, the columns and rows arranged from the detected dies are predicted using a clustering method; thus, an initial wafer map is produced. This map is composed of detected die patterns and predicted pattern candidates. In the final phase of the proposed algorithm, we implement a deep learning-based model to determine defective and abnormal patterns in the wafer map. The experimental results verified the effectiveness and efficiency of our proposed algorithm. In conclusion, the proposed method performs well in identifying defective and abnormal die patterns, and produces a wafer map that presents important information for solving wafer fabrication issues.

Nowadays, semantic segmentation is increasingly used in exploration by underwater robots, for example in autonomous navigation, so that the robot can recognise the nature and elements of its environment during the mission and act according to this classification to avoid collisions. Other applications can be found in the search for archaeological artefacts, in the inspection of underwater structures or in species monitoring. Therefore, it is necessary to try to improve the performance in these tasks as much as possible. To this end, we compare some methods for improving image quality and for data augmentation and test whether higher performance metrics can be achieved with both strategies. The experiments are performed with the SegNet implementations and the SUIM dataset with 8 common underwater classes to compare the obtained results with the already known ones. The results obtained with both strategies show that they are beneficial and lead to better performance results by achieving a mean IoU of 56% and an increased overall accuracy of 81.8%. The single result shows that there are 5 classes with an IoU value above 60% and only one class with an IoU value below 30%, which is a more reliable result and easier to use in real contexts.

Synthetic Aperture Radar (SAR) is an active sensor that uses microwave for sense, it is unrestricted by weather and illumination conditions, and it can observe targets all day and weather. Aircraft targets are important monitoring objects in military and civilian fields, and how to efficiently detect and recognize aircraft targets is an important topic in the field of SAR image interpretation. Based on the features of SAR images, such as complex background, high resolution, and multi-scale, we proposed an improved method based on YOLOv5s. Firstly, this paper proposed the structure of the multi-scale receptive field and channel attention fusion, which is applied to the shallow layer of the backbone of YOLOv5s, it can adjust the weights of the multi-scale receptive field during the training process to enhance the extraction ability of feature information. Secondly, we proposed four de-coupled detection heads to replace the original part in YOLOv5s, which can improve the efficiency and accuracy of SAR image interpretation for small targets. Thirdly, in the case of the limited amount of SAR images, this paper proposed multi methods of data augmentation, which can enhance the diversity and generalization of the network. Fourthly, this paper proposed the K-means++ to re-place the original K-means to improve the network convergence speed and detection accuracy. Finally, Experiments demonstrate that the improved YOLOv5s can enhance the accuracy of SAR image interpretation by 9.3%, and the accuracy of small targets is improved more obviously, reaching 13.1%.

A morphological feature estimation method is presented, that can be used to monitor fish from low contrast underwater images and videos. The fish orientation is classified in order to apply a shape alignment technique that is based on the Ensemble of Regression Trees machine learning method. Shape alignment allows the estimation of fish dimensions (length, height, mass) and the location of fish body parts of particular interest such as the eyes and gills. The developed system can be exploited both for fish monitoring in open sea as well as aquacultures where non-invasive, low cost morphological feature extraction is essential. Fish health can be assessed from malformations in body shape, the color of the eyes or the gills as well as the fish behavior. The proposed method can estimate the position of 18 landmarks with an accuracy of about 95% from low resolution and low contrast underwater images where the fish can be hardly distinguished from its background. Hardware acceleration can be applied at various stages of the proposed method for real time video processing. As a case study, the developed system has been trained and tested with several Mediterranean fish species in the category of seabream or similar: diplodous sargus annularis (white seabream), diplodus annularis seabream (spawn), oblada melanura (saddled seabream), pagrus pagrus (common seabream), etc. A large dataset with low resolution underwater videos and images has been created in the context of this work.

Big Data in medicine includes possibly fast processing of large data sets, both current and historical in purpose supporting the diagnosis and therapy of patients' diseases. Support systems for these activities may include pre-programmed rules based on data obtained from the interview medical and automatic analysis of test results diagnostic results will lead to classification of observations to a specific disease entity. The current revolution using Big Data significantly expands the role of computer science in achieving these goals, which is why we propose a Big Data computer data processing system using artificial intelligence to analyze and process medical images.

Edge detection is one of the key issues in the field of computer vision and remote sensing image analysis. Although many different edge-detection methods have been proposed for gray-scale, color, and multispectral images, they still face difficulties when extracting edge features from hyperspectral images (HSIs) that contain a large number of bands with very narrow gap in the spectral domain. Inspired by the clustering characteristic of the gravitation, a novel edge-detection algorithm for HSIs is presented in this paper. In the proposed method, we first construct a joint feature space by combining the spatial and spectral features. Each pixel of HSI is assumed to be a celestial object in the joint feature space, which exerts gravitational force to each of its neighboring pixel. Accordingly, each object travels in the joint feature space until it reaches a stable equilibrium. At the equilibrium, the image is smoothed and the edges are enhanced, where the edge pixels can be easily distinguished by calculating the gravitational potential energy. The proposed edge-detection method is tested on several benchmark HSIs and the obtained results were compared with those of three state-of-the-art approaches. The experimental results confirm the efficacy of the proposed method

Despite increased protection globally, climate change and other anthropogenic factors have caused a significant decline in seagrass cover. Identifying the specific causes of this decline is paramount if they are to be addressed. Therefore, we identified the causes of long-term change in seagrass/submerged aquatic vegetation (SAV) percentage cover and extent in the Bluefields Bay Special Fish Conservation Area (BBSFCA), a marine protected area on Jamaica’s southern coast. We used two random forest regression (RFr) models that included 2013 hydroacoustic survey SAV percentage cover data (dependent variable), and auxiliary data, reflectance data from level 2 processed 2013 Landsat 7 and 8 images and image band texture statistics maps (mean and variance) as predictors to generate 24 SAV percentage cover maps for the period 1984 – 2021 (37 years) using data from Landsat 4-5, 7 and 8 images., from which benthic features maps (SAV present, absent and coral reef) were created. Rainfall and map data were used to determine if SAV extent/area (km2) and average percentage cover and annual rainfall changed significantly over time and to evaluate the influence of rainfall. Rainfall impact on the overall spatial patterns of SAV loss, gain, and percentage cover change was assessed using a pixel-based regression. Finally, the most important spatial pattern predictors (two rainfall proxies (distance and direction from river mouth), benthic topography, depth, and hurricane exposure (a measure of hurricane disturbance)) of SAV loss, gain, and percentage cover change during 23 successive 1-to-4-year periods were identified using spatial Bayesian INLA generalized linear mixed models. SAV area/extent was largely stable with &gt; 70% mean percentage cover for multiple years. However, Hurricane Ivan (in 2004) caused a significant decline in SAV area/extent (by 1.62 km2, or a 13%) during 2002 – 2006 and a second hurricane (Dean) in 2007 delayed recovery until 2015. Additionally, rainfall declined significantly by &gt;1000 mm since 1901, and mean monthly rainfall positively influenced SAV percentage cover change. The pixel-based regression highlighted areas where mean monthly rainfall had a positive overall effect on SAV cover percentage change (across the entire bay) and gain (closer to the mouth of a river). The most frequently selected important spatial pattern predictors were the two rainfall proxies (areas closer to the river mouth were more likely to experience SAV loss and gain) and depth, with shallow areas generally having a higher probability of SAV loss and gain. Three hurricanes had significant but different impacts depending on their distance from the southern coastline. Hurricane Gilbert, which made landfall in 1988, resulted in higher SAV percentage cover loss in 1987 - 1988. Benthic locations with a northwestern/northern facing aspect (the predominant direction of Ivan’s leading edge wind bands) experienced higher SAV losses during 2002 – 2006. Although some locations impacted by Ivan and not by Dean recovered during 2006 – 2008, exposure to Ivan explained percentage cover loss during 2006 – 2008 and average exposure to (the cumulative impact of) Ivan and Dean (both with tracks close to the southern coastline) explained SAV loss during 2013 – 2015. Therefore, despite historic lows in annual rainfall, overall, higher rainfall was beneficial, multiple hurricanes impacted the site, and despite two hurricanes in three years SAV recovered within a decade. Hurricanes and a further reduction in rainfall may pose a serious threat to SAV persistence in the future.

A topic of growing interest in urban remote sensing is the automated extraction of geometrical building information for 3D city modeling. Roof geometry information is useful for applications such as urban planning, solar potential estimation and telecommunication installation planning, and wind flow simulations for pollutant diffusion analysis. Recent research has proven that the advance in remote sensing technologies and deep learning methods offer the prospects of deriving the roof structure information accurately and efficiently. In this study, we propose a Vectorized Roof Extractor- method based on Fully Convolutional Networks (FCNs) and advanced polygonization method to extract roof structure from aerial imagery and a normalized Digital Surface Models (nDSM) in a regularized vector format. The roof structure consists of building outlines, external edges of the building roof, inner rooflines, internal intersections of the main roof planes. The methodology is comprised of segmentation, vectorization and post-processing for outer rooflines, external edges of the building roof, and inner rooflines, and internal intersections of the main roof planes. For the comparison, we adapt the Frame field Learning (FFL) method originally designed to extract building polygons [1]. Our experiments are conducted on a custom data set derived for the city of Enschede, The Netherlands, using aerial imagery, nDSM and manually digitized training polygons. The results show that the proposed Vectorized Roof Extractor outperformed adapted FFL on PoLiS distance with values of 3.5 m and 1.2 m for outlines and inner rooflines, respectively. Furthermore, the model surpassed the adapted FFL on PoLiS-thresholded F-score for outlines and inner rooflines, with 0.31 and 0.57, respectively. The Vectorized Roof Extractor produced adequate visual results, with straighter walls and fewer missed inner roofline detections. It can predict buildings with common walls thanks to skeleton graph computation. To summarize, the proposed method is suitable for urban applications and has the potential to be improved further.

Lesion studies are crucial in establishing brain-behavior relationships, and accurately segmenting the lesion represents the first step in achieving this. Manual lesion segmentation is the gold standard for chronic strokes. However, it is labor-intensive, subject to bias, and limits sample size. Therefore, our objective is to develop an automatic segmentation algorithm for chronic stroke lesions on T1-weighted MR images. Methods: To train our model, we utilized an open-source dataset: ATLAS v2.0 (Anatomical Tracings of Lesions After Stroke). We partitioned the dataset of 655 T1 images with manual segmentation labels into five subsets and performed a 5-fold cross-validation to avoid overfitting of the model. We used a deep neural network (DNN) architecture for model training. Results: To evaluate the model performance, we used three metrics that pertain to diverse aspects of volumetric segmentation including shape, location, and size. The Dice similarity coefficient (DSC) compares the spatial overlap between manual and machine segmentation. The average DSC was 0.65 (0.61- 0.67; 95% bootstrapped CI). Average symmetric surface distance (ASSD) measures contour distances between the two segmentations. ASSD between manual and automatic segmentation was 12mm. Finally, we compared the total lesion volumes and the Pearson correlation coefficient (ρ) between the manual and automatically segmented lesion volumes, which was 0.97(p-value &lt; 0.001). Conclusions: We present the first automated segmentation model trained on a large multicentric dataset. This model will enable automated on-demand processing of MRI scans and quantitative chronic stroke lesion assessment.

UAV network operation enables gathering and fusion from disparate information sources for flight control in both manned and unmanned platforms. In this investigation, a novel procedure for detecting runways and horizons as well as enhancing surrounding terrain is introduced based on fusion of enhanced vision system (EVS) and synthetic vision system (SVS) images. EVS and SVS image fusion has yet to be implemented real-world situations due to signal misalignment. We address this through a registration step to align the EVS and SVS images. Four fusion rules combining discrete wavelet transform (DWT) sub-bands are formulated, implemented and evaluated. The resulting procedure is tested on real EVS-SVS image pairs and pairs containing simulated turbulence. Evaluations reveal that runways and horizons can be detected accurately even in poor visibility. Furthermore, it is demonstrated that different aspects of the EVS and SVS images can be emphasized by using different DWT fusion rules. The procedure is autonomous throughout landing, irrespective of weather. We believe the fusion architecture developed holds promise for incorporation into head-up displays (HUDs) and UAV remote displays to assist pilots landing aircraft in poor lighting and varying weather. The algorithm also provided a basis rule selection in other signal fusion applications.

In recent years, ontology plays an important role in facilitating the semantic description of images which consists of describing the image by a set of words to describe the scene which signifies the belonging of the image. This description specified in several archaeological fields such as cultural, religious objects and heritage objects of France. However, the description of archaeological images is not yet exploited because this field is very diverse. In this article, we present a panel of ontology construction methods as well as archaeological ontologies. Also, we show our own OntunAr ontology to describe Tunisian archaeological images. In addition, we integrate OnTunAr into the automatic annotation of Tunisian archaeological images.

The growing amount of data demands methods that can gradually learn from new samples. However, it is not trivial to continually train a network. Retraining a network with new data usually results in a known phenomenon, called “catastrophic forgetting.” In a nutshell, the performance of the model drops on the previous data by learning from the new instances. This paper explores this issue in the table detection problem. While there are multiple datasets and sophisticated methods for table detection, the utilization of continual learning techniques in this domain was not studied. We employed an effective technique called experience replay and performed extensive experiments on several datasets to investigate the effects of catastrophic forgetting. Results show that our proposed approach mitigates the performance drop by 15 percent. To the best of our knowledge, this is the first time that continual learning techniques are adopted for table detection, and we hope this stands as a baseline for future research.

Visual Question Generation (VQG) from images is a rising research topic in both fields of natural language processing and computer vision. Although there are some recent efforts towards generating questions from images in the open domain, the VQG task in the medical domain has not been well-studied so far due to the lack of labeled data. In this paper, we introduce a goal-driven VQG approach for radiology images called VQGRaD that generates questions targeting specific image aspects such as modality and abnormality. In particular, we study generating natural language questions based on the visual content of the image and on additional information such as the image caption and the question category. VQGRaD encodes the dense vectors of different inputs into two latent spaces, which allows generating, for a specific question category, relevant questions about the images, with or without their captions. We also explore the impact of domain knowledge incorporation (e.g., medical entities and semantic types) and data augmentation techniques on visual question generation in the medical domain. Experiments performed on the VQA-RAD dataset of clinical visual questions showed that VQGRaD achieves 61.86% BLEU score and outperforms strong baselines. We also performed a blinded human evaluation of the grammaticality, fluency, and relevance of the generated questions. The human evaluation demonstrated the better quality of VQGRaD outputs and showed that incorporating medical entities improves the quality of the generated questions. Using the test data and evaluation process of the ImageCLEF 2020 VQA-Med challenge, we found that relying on the proposed data augmentation technique to generate new training samples by applying different kinds of transformations, can mitigate the lack of data, avoid overfitting, and bring a substantial improvement in medical VQG.

Several depth image based rendering (DIBR) watermarking methods have been proposed, but they have various drawbacks, such as non-blindness, low imperceptibility, and vulnerability to signal or geometric distortion. This paper proposes a template based DIBR watermarking method that overcomes the drawbacks of previous methods. The proposed method exploits two properties to resist DIBR attacks: the pixel is only moved horizontally by DIBR, and the smaller block is not distorted by DIBR. The one dimensional (1D) discrete cosine transform (DCT) and curvelet domains are adopted to utilize these two properties. A template is inserted in the curvelet domain to restore the synchronization error caused by geometric distortion. A watermark is inserted in the 1D DCT domain to insert and detect a message from the DIBR image. Experimental results of the proposed method show high imperceptibility and robustness to various attacks, such as signal and geometric distortions. The proposed method is also robust to DIBR distortion and DIBR configuration adjustment, such as depth image preprocessing and baseline distance adjustment.

Ship detection in large-scene offshore synthetic aperture radar (SAR) images is crucial in civil and military fields, such as maritime management and wartime reconnaissance. However, the problems of low detection rates, high false alarm rates, and high missed detection rates of offshore ship targets in large-scene SAR images are due to the occlusion of objects or mutual occlusion among targets, especially for small ship targets. To solve this problem, this study proposes a target detection model (TAC_CSAC_Net) that incorporates a multi-attention mechanism for detecting marine vessels in large-scene SAR images. Experiments were conducted on two public datasets, the SAR-Ship-Dataset and high-resolution SAR image dataset (HRSID), with multi-scene and multi-size, and the results showed that the proposed TAC_CSAC_Net model achieves good performance for both small and occluded target detection. Experiments were conducted on a real large-scene dataset, LS-SSDD, to obtain the detection results of subgraphs of the same scene. Quantitative comparisons were made with classical and recently developed deep learning models, and the experiments demonstrated that the proposed model outperformed other models for large-scene SAR image target detection.

Hyperspectral image and LiDAR image fusion plays a crucial role in remote sensing by capturing spatial relationships and modeling semantic information for accurate classification and recognition. However, existing methods, like Graph Convolutional Networks (GCNs), face challenges in constructing effective graph structures due to variations in local semantic information and limited receptiveness to large-scale contextual structures. To overcome these limitations, we proposed a invariant attribute-driven binary bi-branch classification (IABC) method which is a unified network that combines binary Convolutional Neural Network (CNN) and GCN with invariant attributes. Our approach utilizes a joint detection framework that can simultaneously learn features from small-scale regular regions and large-scale irregular regions, resulting in an enhanced structured representation of HSI and LiDAR images in the spectral-spatial domain. This approach not only improves the accuracy of classification and recognition but also reduces storage requirements and enables real-time decision-making, which is crucial for effectively processing large-scale remote sensing data. Extensive experiments demonstrates the superior performance of our proposed method in hyperspectral image analysis tasks. The combination of CNNs and GCNs allows for accurate modeling of spatial relationships and effective construction of graph structures. Furthermore, the integration of binary quantization enhances computational efficiency, enabling real-time processing of large-scale data. Therefore, our approach presents a promising opportunity for advancing remote sensing applications using deep learning techniques.

In this work, the problem is considered for eliminating a non-uniform rectilinear smearing of an image by mathematical and computer-based way, for example, a picture of several cars taken with a fixed camera and moving with different speeds. The problem is described by a set of 1-dimensional Fredholm integral equations (IEs) of the first kind of convolution type with a 1-dimensional point spread function (PSF) at uniform smearing and by a set of new 1-dimensional IEs of a general type (not convolution type) with a 2-dimensional PSF at non-uniform smearing. The problem is also described by one 2-dimensional IE of convolution type with a 2-dimensional PSF at uniform smearing and by a new 2-dimensional IE of a general type with a 4-dimensional PSF at non-uniform smearing. The problem for solving Fredholm IE of the first kind is ill-posed (unstable). Therefore, IEs of convolution type are solved by the Fourier transform (FT) method and Tikhonov's regularization (TR), and IEs of general type are solved by the quadrature/cubature and TR methods. Moreover, the magnitude of the image smear Δ is determined by the original “spectral method”, which increases the accuracy of image restoration. It is shown that the use of a set of 1-dimensional IEs is preferable to one 2-dimensional IE in the case of non-uniform smearing. In the inverse problem (image restoration), the Gibbs effect (the effect of false waves) in the image may occur. It can be edge or inner. The edge effect is well suppressed by the proposed technique “diffusing the edges". In the case of an inner effect, it is eliminated with difficulty, but the image smearing itself plays the role of diffusing and suppresses the inner Gibbs effect to a large extent. It is shown (in the presence of impulse noise in an image) that the well-known Tukey median filter can distort the image itself, and the Gonzalez adaptive filter also distorts the image (but to a lesser extent). We propose a modified adaptive filter. A software package was developed in MatLab and illustrative calculations are performed.

Telemedicine is the use of Information and Communication Technology (ICT) for clinical health care from a distance. The exchange of radiographic images and electronic patient health information/records (ePHI/R) for diagnostic purposes has the risk of confidentiality, ownership identity, and authenticity. In this paper, a data hiding technique for ePHI/R is proposed. The color information in the cover image is used for key generation, and stego-images are produced with ideal case. As a result, the whole stego-system is perfectly secure. This method includes the features of watermarking and steganography techniques. The method is applied to radiographic images. For the radiographic images, this method resembles watermarking, which is an ePHI/R data system. Experiments show promising results for the application of this method to radiographic images in ePHI/R for both transmission and storage purpose.

Containing hundreds of spectral bands (features), hyperspectral images (HSIs) have high ability in discrimination of land cover classes. Traditional HSIs data processing methods consider the same importance for all bands in the original feature space (OFS), while different spectral bands play different roles in identification of samples of different classes. In order to explore the relative importance of each feature, we learn a weighting matrix and obtain the relative weighted feature space (RWFS) as an enriched feature space for HSIs data analysis in this paper. To overcome the difficulty of limited labeled samples which is common case in HSIs data analysis, we extend our method to semisupervised framework. To transfer available knowledge to unlabeled samples, we employ graph based clustering where low rank representation (LRR) is used to define the similarity function for graph. After construction the RWFS, any arbitrary dimension reduction method and classification algorithm can be employed in RWFS. The experimental results on two well-known HSIs data set show that some dimension reduction algorithms have better performance in the new weighted feature space.

Skin image analysis using artificial intelligence (AI) has recently attracted significant research interest, particularly for analyzing skin images captured by mobile devices. Acne is one of the most common skin conditions with profound effects in severe cases. In this study, we developed an AI system called AcneDet for automatic acne object detection and acne severity grading using facial images captured by smartphones. AcneDet includes two models for conducting two tasks: (1) a Faster R-CNN-based deep learning model for the detection of acne lesion objects of four types including blackheads/whiteheads, papules/pustules, nodules/cysts, and acne scars; and (2) a LightGBM machine learning model for grading acne severity using the Investigator’s Global Assessment (IGA) scale. The output of the Faster R-CNN model, i.e., the counts of each acne type, were used as input for the LightGBM model for acne severity grading. A dataset consisting of 1,572 labeled facial images captured by both iOS and Android smartphones was used for training. The results show that the Faster R-CNN model achieves a mAP of 0.54 for acne object detection. The mean accuracy of acne severity grading by the LightGBM model is 0.85. With this study, we hope to contribute to the development of artificial intelligent systems that are able to help acne patients understand more about their conditions and support doctors in acne diagnosis.

In this paper we develop and explore a methodical pipeline for swimmer localisation in outdoor environments. The developed framework is intended to be used for enhancing swimmer safety. A main issue we deal with by the proposed approach is the lack of real world training data in such outdoor environments. Natural lighting changes, dynamic water textures and possibly barely visible swimming persons are key elements to approach. We account for these difficulties by adopting an effective background removal technique with available training data. This allows us to edit swimmers into natural environment backgrounds for the use in subsequent image augmentation. We created 17 training datasets with real images, synthetic images and a mixture of both to investigate different aspects and characteristics of the proposed approach. The datasets are used to train a YOLO architecture for the possible future application in real-time detection. The trained framework is then tested and evaluated on outdoor environment imagery acquired by a safety drone to investigate and confirm the usefulness for outdoor swimmer localisation.

With the rapid advancements in remote sensing technology, the spectral information from hyperspectral remote sensing images has become increasingly rich, facilitating detailed spectral analysis of the Earth's surface objects. However, this abundance of spectral data poses significant challenges in data processing, such as the curse of dimensionality leading to the “Hughes” phenomenon, "strong correlation" due to high resolution, and "non-linear characteristics" caused by varied surface reflectance rates. Therefore, dimensionality reduction of hyperspectral data has become a crucial task. This paper, grounded in manifold theory and manifold learning techniques, and considering the non-linear structures and features in hyperspectral remote sensing data, elucidates the principles and processes of dimensionality reduction in hyperspectral remote sensing images using manifold learning, with a formalized expression of the process. This article introduces spectral information divergence (SID) into the nearest neighbor graph computation of manifold learning algorithms. The principles and computational processes of nearest neighbor graph algorithms based on Euclidean distance (ED), spectral angle mapping (SAM), and SID are studied, and a comparative analysis of the dimensionality reduction effects under these three metrics in hyperspectral data is conducted. Experiments on feature extraction under different metrics were performed using the publicly available Indian Pines hyperspectral dataset. The intrinsic features obtained post-dimensionality reduction were used as inputs for classification algorithms in ground objects classification experiments, with algorithm runtime, overall accuracy, and Kappa coefficient as evaluation metrics for dimensionality reduction quality. The results demonstrate that nearest neighbor graph computation based on SAM and SID outperforms traditional ED methods; SAM-based computation has the lowest time complexity, while SID-based manifold learning yields the highest accuracy in ground objects classification. Thus, manifold learning based on SAM and SID metrics proves to be an effective method for feature extraction in hyperspectral remote sensing data, underscoring the potential of manifold learning techniques in the dimensionality reduction of hyperspectral remote sensing images.

Several studies in computer vision have examined specular removal, which is crucial for object detection and recognition. This research has traditionally been divided into two tasks: specular highlight removal, which focuses on removing specular highlights on object surfaces, and reflection removal, which deals with specular reflections occurring on glass surfaces. In reality, however, both types of specular effects often coexist, making it a fundamental challenge that has not been adequately addressed. Recognizing the necessity of integrating specular components handled in both tasks, we constructed a Specular-Light (S-Light) DB for training single-image-based deep learning models. Moreover, considering the absence of benchmark datasets for quantitative evaluation, the Multi-Scale Normalized Cross Correlation (MS-NCC) metric, which considers the correlation between specular and diffuse components, was introduced to assess the learning outcomes.

With the development of remote sensing satellite technology for Earth observation, remote sensing stereo images have been used for three-dimensional reconstruction in various fields, such as urban planning and construction. However, remote sensing images often include noise, occluded regions, weakly textured areas and repeated textures, which can lead to reduced accuracy in stereo matching and affect the quality of the 3D reconstruction results. To reduce the impact of complex scenes in remote sensing images on stereo matching and to ensure both speed and accuracy, we propose a new end-to-end stereo matching network based on convolutional neural networks (CNNs). The proposed stereo matching network can learn features at different scales from the original images and construct cost volumes with varying scales to obtain richer scale information. Additionally, when constructing the cost volume, we introduce negative disparity to adapt to the common occurrence of both negative and nonnegative disparities in remote sensing stereo image pairs. For cost aggregation, we employ a 3D convolution-based encoder-decoder structure that allows the network to adaptively aggregate information. Before feature aggregation, we also introduce an attention module to retain more valuable feature information, enhance feature representation, and obtain a higher-quality disparity map. By training on the publicly available US3D dataset, we obtain the accuracy that 1.115 pixel in end-point error (EPE) and 5.32% in the error pixel ratio (D1) on the test dataset, and the inference speed is 92 ms. Comparing our model with existing state-of-the-art models, we achieve higher accuracy, and the network is beneficial for the three-dimensional reconstruction of remote sensing images.

The corrugated board is a versatile and durable material that is widely used in the packaging industry. Its unique structure provides strength and cushioning, while its recyclability and bio-degradability make it an environmentally friendly option. The strength of corrugated board depends on many factors, including the type of individual papers on flat and corrugated layers, the geometry of the flute, temperature, humidity, etc. This paper presents a new approach to analysis geometric features of corrugated boards based on the images. It allows for precise determination of the geometry of the corrugated layer, as well as the thickness of individual papers in all cardboard layers. The experimental set used in the work and the created software are characterized by high reliability and precision of measurement thanks to the use of an identification procedure based on image analysis and evolutionary algorithms. This allows to quickly find the main geometric characteristics of the corrugated board, which in turn leads to the possibility of performing numerical analyzes taking into account the real model of the corrugated board. Calculations made on a real (correctly mapped geometrically) model allow, for example, a more reliable analysis of the composition of the cardboard, and thus also to obtain greater savings in the optimization process.

Manual interpretation of these huge amounts of image volumes are susceptible to inter-reader variability and human error. Thus, accurate automated CAD scheme is highly desirable in clinical pathological diagnosis. In this research, plethora of machine learning paradigms (e.g. feature extraction, dimensionality reduction and supervised classification methods) were explored, evaluated, compared and analyzed to identify the optimal pathway for brain MR images (normal vs neoplastic) binary classification task. External validation dataset was used to test the generalizability of the optimal predictive models implemented. Relevant and informative features were selected to construct cross-validated decision tree and eventually simple rule set was built based on the decision tree. The experimental results show that almost all pattern recognition paradigms achieve high accuracy with careful selection of number of attributes. LDA+ELM with 55 features are the optimal pipelines which achieve perfect classification when training and test data are of same source; and achieving (accuracy=97.5%, AUC=0.989, sensitivity=95% and specificity=100%) under balanced test dataset; (accuracy=99.5%, AUC=0.988, sensitivity=95% and specificity=100%). Cross-validated decision tree model also shows comparable performance: accuracy=98.8%, AUC=99.1%, sensitivity=99.6% and specificity=98.2%. Three highly relevant and robust attributes are visualized and selected for construction of decision tree models and finally a rule sets are read directly off the decision tree. This rule sets can potentially serve as fast and accurate classification algorithm.

Αbstract: To date, countless satellite image fusions have been made, mainly with panchromatic spatial resolution to a multispectral image ratio of 1/4, fewer fusions with lower ratios, and relatively recently fusions with much higher spatial resolution ratios have been published. Apart from this, there is a small number of publications studying the fusion of aerial photographs with satellite images, with the year of image acquisition varying and the dates of acquisition not mentioned. In addition, in these publications, either no quantitative controls are performed on the composite images produced, or the aerial photographs are recent and colorful and only the RGB bands of the satellite images are used for data fusion purposes. The objective of this paper is the study of the addition of multispectral information from satellite images to black and white aerial photographs of the 80s decade (1980–1990) with small difference (just a few days) in their image acquisition date, the same year and season. Quantitative tests are performed in two case studies and the results are encouraging, as the accuracy of the classification of the features and objects of the Earth’s surface is improved and the automatic digital extraction of their form and shape from the archived aerial photographs is now allowed. This opens up a new field of use for the black and white aerial photographs and archived multispectral satellite images of the same period in a variety of applications, such as the temporal changes of cities, forests and archaeological sites.

In recent years, remote sensing technology has been widely used to predict forest stand parameters. In order to compare the effects of different features of remote sensing images and topographic information on the prediction of forest stand parameters, multivariate stepwise regression analysis method was used to build estimation models for important forest stand parameters by using textural and spectral features as well as topographic information of SPOT-5 satellite images in northeastern Heilongjiang Province in China as independent variables. The study results show that the optimal window to predict forest stand parameters using textural features of SPOT-5 satellite image is 9×9; the ability of textural features was better than that of spectral features in terms of predicting forest stand parameters; with the inclusion of topographic information, the accuracy of prediction of all models was improved, of which elevation has the most significant effect. The highest accuracy was achieved when predicting the stand volume (SV) (R²_adj=0.820), followed by basal area (BA) (R²_adj =0.778), accuracy of both above models exceeded 75%. The results show that models combined use of textural, spectral features and topographic information of SPOT-5 images have a good application prospect in predicting forest stand parameters.

Recent advancements in computer vision, especially deep learning models, have shown considerable promise in tasks related to plant image object detection. However, the efficiency of these deep learning models heavily relies on input image quality, with low-resolution images significantly hindering model performance. Therefore, reconstructing high-quality images through specific techniques will help extract features from plant images, thus improve model performance. In this study, we explored the value of super-resolution technology for improving object detection model performance on plant images. Firstly, we built a comprehensive dataset comprising 1030 high-resolution plant images, named the PlantSR dataset. Subsequently, we developed a super-resolution model using the PlantSR dataset and benchmarked it against several state-or-the-art models designed for general image super-resolution tasks. Our proposed model demonstrated superior performance on the PlantSR dataset, indicating its efficacy in enhancing the super-resolution of plant images. Furthermore, we explored the effect of super-resolution on two specific object detection tasks: apple counting and soybean seed counting. By incorporating super-resolution as a pre-processing step, we observed a significant reduction in mean absolute error. Specifically, on the YOLOv7 model employed for apple counting, the mean absolute error decreased from 13.085 to 5.71. Similarly, on the P2PNet-Soy model utilized for soybean seed counting, the mean absolute error decreased from 19.159 to 15.085. These findings underscore the substantial potential of super-resolution technology in improving the performance of object detection models for accurately detecting and counting specific plants from images. The source codes and associated datasets are available at https://github.com/SkyCol/PlantSR.

Vegetation identification using remote sensing is essential for understanding terrestrial ecosystems and monitoring climate change. This technology by providing high-resolution images of the Earth's surface from satellites, aircraft, and drones has revolutionized the way scientists study and un-derstand vegetation patterns, providing information on the distribution, structure, and health of plant communities in different regions and biomes. However, identifying vegetation in multi and hyperspectral images remains a difficult, time-consuming, and costly process. Traditional methods for vegetation mapping involve the in-terpretation of spectral indices such as the Normalized Difference Vegetation Index (NDVI), alt-hough they often suffer from limitations such as manual interpretation and limited spatial resolution of satellite images. Furthermore, machine learning techniques, artificial neural networks, decision trees, support vector machines, and random forests have shown promising results in identifying and classifying vegetation cover from hyperspectral images. However, due to the need for per-son-nel with specialized expertise, their application on a large scale is still challenging. On the other hand, correlation methods between curves have emerged as powerful and simplified tool for comparing spectral signatures of hyperspectral image pixels and efficiently classifying vegetation in a study area. Pearson's correlation coefficient and spectral angle mapper are among the most common correlation methods used in vegetation mapping. These correlation methods, among many others available, offer an easily implemented and computationally efficient approach, making them particularly useful for applications in developing countries or regions with limited resources. In this article, a comparative study of correlation/distance metrics was conducted for the detection of vegetation pixels in hyperspectral images. The study allowed for the comparison of five distance and/or correlation metrics: direct correlation, cosine similarity, normalized Euclidean distance, Bray-Curtis’s distance, and Pearson correlation. The two metrics that yielded the best accuracy results in vegetation pixel detection were direct correlation and Pearson correlation. Based on the selected methods, a vegetation detection algorithm was implemented and validated on a hyper-spectral image of the Manga neighborhood in Cartagena de Indias, Colombia. The spectral library was utilized for image processing, while the numpy and scipy libraries in the Python programming language were used for the mathematical calculation of correlations. Both the study's approach and the implemented algorithm aim to serve as a reference for conducting detection studies of various material types in hyperspectral images using open-access programming platforms.

There is great interest in automatically detecting road weather and understanding its impacts on the overall safety of the transport network. This can, for example, support road condition-based maintenance or even serve as detection systems that assist safe driving during adverse climate conditions. In computer vision, previous work has demonstrated the effectiveness of deep learning in predicting weather conditions from outdoor images. However, training deep learning models to accurately predict weather conditions using real-world road-facing images is difficult due to: (1) the simultaneous occurrence of multiple weather conditions; (2) imbalanced occurrence of weather conditions throughout the year; and (3) road idiosyncrasies, such as road layouts, illumination, road objects etc. In this paper, we explore the use of focal loss function to force the learning process to focus on weather instances that are hard to learn with the objective to help address data imbalance. In addition, we explore the attention mechanism for pixel based dynamic weight adjustment to handle road idiosyncrasies using state-of-the-art vision transformer models. Experiments with a novel multi-label road weather dataset show that focal loss significantly increases the accuracy of computer vision approaches for imbalanced weather conditions. Furthermore, vision transformers outperforms current state-of-the-art convolutional neural networks in predicting weather conditions with a validation accuracy of 92% and F1-score of 81.22%, which is impressive considering the imbalanced nature of the dataset.

Tables in the document image are one of the most important entities since they contain crucial information. Therefore, accurate table detection can significantly improve information extraction from tables. In this work, we present a novel end-to-end trainable pipeline, HybridTabNet, for table detection in scanned document images. Our two-stage table detector uses the ResNeXt-101 backbone for feature extraction and Hybrid Task Cascade (HTC) to localize the tables in scanned document images. Moreover, we replace conventional convolutions with deformable convolutions in the backbone network. This enables our network to detect tables of arbitrary layouts precisely. We evaluate our approach comprehensively on ICDAR-13, ICDAR-17 POD, ICDAR-19, TableBank, Marmot, and UNLV. Apart from the ICDAR-17 POD dataset, our proposed HybridTabNet outperforms earlier state-of-the-art results without depending on pre and post-processing steps. Furthermore, to investigate how the proposed method generalizes unseen data, we conduct an exhaustive leave-one-out-evaluation. In comparison to prior state-of-the-art results, our method reduces the relative error by 27.57% on ICDAR-2019-TrackA-Modern, 42.64% on TableBank (Latex), 41.33% on TableBank (Word), 55.73% on TableBank (Latex + Word), 10% on Marmot, and 9.67% on UNLV dataset. The achieved results reflect the superior performance of the proposed method.

This paper develops a detail image signal enhancement that makes images perceived as clearer and more resolved and so is more effective for higher resolution displays. We observe that the local variant signal enhancement makes images more vivid, and the more revealed granular signals harmonically embedded on the local variant signals make images more resolved. Based on this observation, we develop a method that not only emphasizes the local variant signals by scaling up the frequency energy in accordance with human visual perception, but also strengths up the granular signals by embedding the alpha-rooting enhanced frequency components. The proposed energy scaling method emphasizes the detail signals in texture images and rarely boosts noisy signals in plain images. In addition, to avoid the local ringing artifact, the proposed method adjusts the enhancement direction to be parallel to the underlying image signal direction. It was verified through the subjective and objective quality evaluations that the developed method makes images perceived as clearer and highly resolved.

This document introduces a new cryptosystem mixing two improvement standards generally used for text encryption, in order to give birth a new color image encryption algorithm capable of dealing with known attacks. Firstly, two substitution matrixes attached to a strong replacement function will be generated for advanced Vigenere technique application. At the end of this first round, the output vector is subdivided into size blocks according to the used chaotic map, for acting a single enhanced Hill circuit insured by a large inversible matrix. A detailed description of such a large involutive matrix constructed using Kronecker products will be given. accompanied by a dynamic translation vector to eliminate any linearity. A solid chaining is established between the encrypted block and the next clear block to avoid any differential attack. Simulations carried out on a large volume of images of different sizes and formats ensure that our approach is not exposed to any known attacks.

As people grow a custom to effortless outdoor navigation there is a rising demand for similar possibility indoors as well. Unfortunately, indoor localization, being one of the necessary requirements for navigation, continues to be problem without a clear solution. In this article we are proposing a method for an indoor positioning system using a single image. This is made possible using small preprocessed database of images with known control points as the only preprocessing needed. Using feature detection with SIFT algorithm we can look through the database and find image which is the most similar to the image taken by user. Pair of images is then used to find coordinates of database image using PnP problem. Furthermore, projection and essential matrices are determined allowing for the user image localization ~ determining the position of the user in indoor environment. Benefits of this approach lies in the single image being the only input from user and no requirements for new onsite infrastructure and thus enables a simpler realization for the building management.

In this paper, a technique of image watermarking using multiple images as watermarks is presented. The technique is based on transform domain functions including discrete wavelet transform (DWT), discrete cosine transform (DCT) and singular value decomposition (SVD) with an image as the host signal i.e. the watermarks will be used as proofs of the authenticity of the host image. The technique is executed by performing multilevel DWT followed by applying DCT and SVD to both the host and watermark. Multiple watermarks are used for the insurance of better security level. The scheme is immune to common image processing operations & some attacks and exhibits PSNR of 108.3781dB, normalized cross correlation (NCC) over 0.99 and normalized correlation (NC) over 0.99.

Spike shape and morphometric characteristics are among the key characteristics of cultivated cereals associated with their productivity. Identification of the genes controlling these traits requires morphometric data at harvesting and analysis of numerous plants, which could be automatically done using technologies of digital image analysis. A method for wheat spike morphometry utilizing 2D image analysis is proposed. Digital images are acquired in two variants: a spike on a table (one projection) or fixed with a clip (four projections). The method identifies spike and awns in the image and estimates their quantitative characteristics (area in image, length, width, circularity, etc.). Section model, quadrilaterals, and radial model are proposed for describing spike shape. Parameters of these models are used to predict spike shape type (spelt, normal, or compact) by machine learning. The mean error in spike density prediction for the images in one projection is 4.61 (~18%) versus 3.33 (~13%) for the parameters obtained using four projections.

Spike shape and morphometric characteristics are among the key characteristics of cultivated cereals associated with their productivity. Identification of the genes controlling these traits requires morphometric data harvesting and analysis for numerous plants, which is automatable using technologies of digital image analysis. A method for wheat spike morphometry utilizing 2D image analysis is proposed. Digital images are acquired in two variants: a spike on a table (one projection) or fixed with a clip (four projections). The method identifies spike and awns in the image and estimates their quantitative characteristics (area in image, length, width, circularity, etc.). Models of sections, quadrilaterals, and radial model are proposed for describing spike shape. Parameters of these models are used to predict spike shape type (spelt, normal, or compact) by machine learning. The mean error in spike density prediction for the images in one projection is 4.61 (~18%) versus 3.33 (~13%) for the parameters obtained using four projections. F1 measure in automated spike classification into three types is 0.78 using logistic regression (one projection) and 0.85 using random forest method (four projections). The proposed method is implemented in Java; examples of images and user guide are available at http://wheatdb.org/werecognizer.

With an increased interest in applications that require a clean background image, such as video surveillance, object tracking, street view imaging and location-based services on web-based maps, multiple algorithms have been developed to reconstruct a background image from cluttered scenes. Traditionally, statistical measures and existing image quality techniques have been applied for evaluating the quality of the reconstructed background images. Though these quality assessment methods have been widely used in the past, their performance in evaluating the perceived quality of the reconstructed background image has not been verified. In this work, we discuss the shortcomings in existing metrics and propose a full reference Reconstructed Background image Quality Index (RBQI) that combines color and structural information at multiple scales using a probability summation model to predict the perceived quality in the reconstructed background image given a reference image. To compare the performance of the proposed quality index with existing image quality assessment measures, we construct two different datasets consisting of reconstructed background images and corresponding subjective scores. The quality assessment measures are evaluated by correlating their objective scores with human subjective ratings. The correlation results show that the proposed RBQI outperforms all the existing approaches. Additionally, the constructed datasets and the corresponding subjective scores provide a benchmark to evaluate the performance of future metrics that are developed to evaluate the perceived quality of reconstructed background images.

Advancements in computer vision are rapidly revolutionizing the way traffic agencies gather roadway geometry data, leading to significant savings in both time and money. Utilizing aerial and satellite imagery for data collection proves to be more cost-effective, more accurate, and safer compared to traditional field observations, considering factors such as equipment cost, crew safety, and data collection efficiency. Consequently, there is a pressing need to develop more efficient methodologies for promptly, safely, and economically acquiring roadway geometry data. While image processing has previously been regarded as a time-consuming and error-prone approach for capturing these data, recent developments in computing power and image recognition techniques have opened up new avenues for accurately detecting and mapping various roadway features from a wide range of imagery data sources. This research introduces a novel approach combining image processing with a YOLO-based methodology to detect turning lane pavement markings from high-resolution aerial images, specifically focusing on Florida's public roadways. Upon comparison with ground truth data from Leon County, Florida, the developed model achieved an average accuracy of 87% at a 25% confidence threshold for detected features. Implementation of the model in Leon County identified approximately 3,026 left turn, 1,210 right turn, and 200 center lane features automatically. This methodology holds paramount significance for transportation agencies in facilitating tasks such as identifying deteriorated markings, comparing turning lane positions with other roadway features like crosswalks, and analyzing intersection-related accidents. The extracted roadway geometry data can also be seamlessly integrated with crash and traffic data, providing crucial insights for policymakers and road users.

Generative AI has gained enormous interest nowadays due to new applications like chatGPT, DALL E, Stable Difussion and Deep Fake. Particularly DALL E, Stable Difussion and others (Adobe Firefly, ImagineArt...) are able to create images from a text prompt and are also able to recreate real photographs. Due to this fact, intense research has arisen to create new image forensics applications able to distinguish between real captured images and videos and artificial ones. Detecting forgeries made with Deep Fake is one of the most researched issues. This paper is about another kind of forgery detection. The purpose of this research aims to detect photo realistic AI created images versus real photos coming from a physical camera. For this purpose, techniques that perform a pixel level feature extraction are used. First one is Photo Response Non-Uniformity (PRNU). PRNU is a special noise due to imperfections on the camera sensor that is used for source camera identification. The underlying idea is that AI images will have a different PRNU pattern. Second one is Error level analysis (ELA). This is other type of feature extraction traditionally used for detecting image editions. In fact, ELA is being used nowadays by photographers to detect manually AI created images. Both kinds of features are used to train Convolutional Neural Networks to differentiate between AI images and real photographs. Good results are obtained achieving accuracy rates over 95%. Both extraction methods are carefully assessed by computing precision/recall and F1-score measurements.

Lung cancer is a prevalent malignancy that impacts individuals of all genders and is often diagnosed late due to delayed symptoms. To catch it early, researchers are developing algorithms to study lung cancer images. One such method uses computer aided detection to classify as benign or malignant from the histopathological images. Standard histopathological images were used from a Lung and Colon Cancer Histopathological Image Dataset (LC25000) which contains two classes of benign and malignant of 5000 each. Images were preprocessed and features extracted using Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO). Feature selection methods used are KL Divergence and Invasive Weed Optimization (IWO). Seven different classifiers like SVM, KNN, Random Forest, Decision Tree, Softmax Discriminant, Multilayer Perceptron, and BLDC were used to analyze and classify the images as benign or malignant. Results were compared using standard metrics, and kappa analysis assessed classifier agreement. The Decision Tree Classifier with GWO feature extraction achieved good accuracy of 85.01% without Feature selection and Hyperparameter Tuning approaches. Furthermore, we present a methodology to enhance the accuracy of the classifiers by employing hyperparameter tuning algorithms based on Adam and RAdam. By combining features from GWO and IWO, and using the RAdam algorithm, the Decision Tree classifier achieves the commendable accuracy of 91.57%

With growing global concern of food and water insecurity, an efficient method to monitor irrigation projects is essential, especially in the developing world, where irrigation performance is often suboptimal. In Nepal, the irrigated area has not been objectively recorded, although their assessment has substantial implications on national policy, project’s annual budgets, and donor fund-ing. Here we present the application of Landsat images to measure irrigated areas in Nepal for the past 17 years to contribute to the assessment of the irrigation performance. Landsat 5 TM (2006-2011) and Landsat 8 OLI (2013-2022) images were used to develop a machine-learning model which classifies irrigated and non-irrigated areas in study areas. The random forest classification achieved overall accuracy of 82.2% and kappa statistics of 0.72. For the class of irrigation areas, the producer’s accuracy and the consumer’s accuracy were 79% and 96%, respectively. Our regionally trained machine-learning model outperforms the existing global cropland map, highlighting the need of such models for local irrigation project evaluations. We assess irrigation project performance and its drivers by combining long-term changes in satellite-derived irrigated area with local data related to irrigation performance, such as annual budget, irrigation service fee, crop yield, precipitation, and main canal discharge.

In this study, we aimed to develop a deep learning model for identifying bacterial keratitis (BK) and fungal keratitis (FK) by using slit-lamp images. We retrospectively collected slit-lamp images of patients with culture-proven microbial keratitis between January 1, 2010, and December 31, 2019, from two medical centers in Taiwan. We constructed a deep learning algorithm, consisting of a segmentation model for cropping cornea images and a classification model that applies convolutional neural networks to differentiate between FK and BK. The model performance was evaluated and presented as the area under the curve (AUC) of the receiver operating characteristic curves. A gradient-weighted class activation mapping technique was used to plot the heatmap of the model. By using 1330 images from 580 patients, the deep learning algorithm achieved an average diagnostic accuracy of 80.00%. The diagnostic accuracy for BK ranged from 79.59% to 95.91% and that for FK ranged from 26.31% to 63.15%. DenseNet169 showed the best model performance, with an AUC of 0.78 for both BK and FK. The heat maps revealed that the model was able to identify the corneal infiltrations. The model showed better diagnostic accuracy than the previously reported diagnostic performance of both general ophthalmologists and corneal specialists.

The analysis of microscopy images has always been an important yet time consuming process in in materials science. Convolutional Neural Networks (CNNs) have been very successfully used for a number of tasks, such as image segmentation. However, training a CNN requires a large amount of hand annotated data, which can be a problem for material science data. We present a procedure to generate synthetic data based on ad-hoc parametric data modelling for enhancing generalization of trained neural network models. Especially for situations where it is not possible to gather a lot of data, such an approach is beneficial and may enable to train a neural network reasonably. Furthermore, we show that targeted data generation by adaptively sampling the parameter space of the generative models gives superior results compared to generating random data points.

This paper aims at tackling with the task of fusion feature from images and its corresponding point clouds for 3D object detection in autonomous driving scenarios basing on AVOD, an Aggregate View Object Detection network. The proposed fusion algorithms fuse features targeted from Bird’s Eye View (BEV) LIDAR point clouds and its corresponding RGB images. Differs in existing fusion methods, which are simply the adoptions of concatenation module, element-wise sum module or element-wise mean module, our proposed fusion algorithms enhance the interaction between BEV feature maps and its corresponding images feature maps by designing a novel structure, where single level feature maps and another utilizes multilevel feature maps. Experiments show that our proposed fusion algorithm produces better results on 3D mAP and AHS with less speed loss comparing to existing fusion method used on the KITTI 3D object detection benchmark.

Virtual Histology- Intravascular Ultrasound (VH-IVUS) image is an available method for visualizing plaque component to detect thin cap fibroatheroma. Nevertheless, this imaging modality has considerable limitations to extract the plaque component features and identifying the TCFA plaque. The aim of this paper is to improve the identification of TCFA using fusion of IVUS and VH-IVUS images. In order to generate the automatic technique for reducing the human interaction, a new method namely Active Contour based Plaque Border Detection (ACPB) is proposed. In order to perform the pixel wise classification, hybrid of K-means algorithm with Particle Swarm Optimization and Plaque based Minimum Euclidean Distance (KMPSO-PMED) method is presented to classify the plaque region as well. Moreover, to obtain more significant information of imaging modalities, fusion of two different images consisting of VH-IVUS and IVUS is performed. Therefore, geometric features are extracted from the plaque region and combine with IVUS features. Furthermore, different group of plaque features are divided by means of the histopathological studies. SVM classifiers is applied to detect the TCFA and non-TCFA plaques. The proposed method is applied on 566 in-vivo IVUS and their matching VH-IVUS images obtained from 9 patients. The best result of SVM illustrates the accuracy rates of 99.41% for classification of TCFA plaque. The results prove that the highest accuracy is achieved by integrated features of IVUS and VH-IVUS images.

Hyperspectral (HS) imaging plays a pivotal role in various fields, including medical diagnostics, where precise human vein detection is crucial. Hyperspectral image data are very large and can cause computational complexities. Dimensionality reduction techniques are often employed to streamline HS image data processing. This paper investigates the effectiveness of three dimensionality reduction techniques, namely: Principal Component Analysis (PCA), Folded PCA (FPCA), and Ward’s Linkage Strategy using Mutual Information (WaLuMI) for vein detection using HS images. A HS image dataset, encompassing left and right-hand images captured from 100 subjects with varying skin tones was created and annotated using anatomical data to represent vein and non-vein areas within the images. To generate experimental results, the HS image dataset was divided into train and test datasets. Optimum performing parameters for each of the dimensionality reduction techniques in conjunction with The Support Vector Machine binary classification were determined using the Training dataset. The performance of the three dimensionality reduction based vein detection methods was then assessed and compared using the test image dataset. Results show that the FPCA-based method outperforms the other two methods in terms of accuracy. For visualization purposes, the classification prediction image for each technique is post-processed using morphological operators, and results show the significant potential of HS imaging in vein detection.

Track fasteners play a pivotal role in infrastructure inspection for high-speed rail. Yet, images taken by drones often capture shadows cast by electrical towers flanking the high-speed rail tracks. These shadows can hinder the visibility of the track fasteners, thereby impacting detection efficiency and accuracy considerably. The present paper introduces an end-to-end shadow removal algorithm, rooted in generative adversarial network training. The comprehensive network framework is segmented into three sub-networks: pseudo-mask generation, shadow removal, and result refinement. We have integrated a Fourier convolutional residual module to bolster the feature extraction capability of the generator network. This integration ensures the network retains a global receptive field, even in its more superficial layers. By employing an overall weighted loss function, we enhance the quality of the images produced without shadows. Further, a perceptual loss function has been incorporated to retain the structural information of objects, setting the stage for subsequent defect detection. Our results highlight that Pse-ShadowNet adeptly eradicates fastener shadows while maintaining vital visual features, including object position, structure, texture, edges, and other key visual elements. Consequently, the reconstructed images are detailed and showcase superior image quality.

As a high-quality forestry product with excellent performance, particleboard is widely used in many fields and has great market potential. At present, it is a research difficulty to accurately identify small, fine and different shapes of defects on the whole particleboard through machine vision technology. The application of image super-resolution reconstruction technology on particleboard can improve the surface image quality of particleboard, which is conducive to the subsequent improvement of defect detection accuracy. In this study, super-resolution dense attention generative adversarial network (SRDAGAN) model was improved to solve the problem that super-resolution generative adversarial network (SRGAN) reconstructed image would produce artifacts and its performance needed to be improved. The Batch Normalization (BN) layer was removed, the convolutional block attention module (CBAM) was optimized to construct dense block, and the dense blocks were constructed by densely skip connection. Then, the corresponding 52400 image blocks with high resolution and low resolution were trained, verified and tested according to the ratio of 3:1:1. The model was comprehensively evaluated from the effect of image reconstruction and the two indexes of PSNR and SSIM. It was found that compared with BICUBIC and SRGAN, PSNR index of SRDAGAN increased by 4.88dB and 3.25dB respectively, and SSIM increased by 0.0507 and 0.1122 respectively. The reconstructed images not only had clearer texture, but also had more realistic expression of various features, and the performance of the model had been greatly improved. At the same time, this study also emphatically discussed on the image reconstruction effect with defects. The result showed that the SRDAGAN proposed in this study can complete the super-resolution reconstruction of particleboard images with high quality. In the future, it can also be further combined with defect detection for actual production to improve the quality of forestry products and increase economic benefits.

We discuss the optimal control formulation for enhancement and denoising of satellite multiband images and propose to take it in the form of an L1-control problem for quasi-linear parabolic equation with non-local p[u]-Laplacian and with a cost functional of a tracking type. The main characteristic feature of the considered class of parabolic problems is the fact that the variable exponent p(t,x) and the anisotropic diffusion tensor D(t,x) are not well predefined a priori, but instead these characteristic non-locally depend on a solution of this problem, i.e., pu=p(t,x,u) and Du=D(t,x,u). We prove the existence of optimal pairs with sparse L1-controls using for that the indirect approach and a special family of approximation problems.

Deep convolutional neural networks (CNN) are favored methods widely used in medical image processing due to their assured shown performance. Recently, the emergence of new lung diseases and the possibility of early detection of their symptoms has attracted many researchers to classify diseases by training deep CNNs on lung CT images. The trained networks are expected to distinguish between lung indications in different diseases, especially at the early stages of them. With the hope of achieving this purpose, we proposed an efficient deep CNN called AFEX-Net with adaptive feature extraction layers that successfully extract distinguishing features and classify chest CT images. The efficiency of the proposed network has two aspects: it is a lightweight network with low number of parameters and fast training and it has adaptive pooling layers and adaptive activation functions to increase its level of compatibility to the input data. The proposed network has been evaluated on a dataset with more than 10K chest CT slices, while an efficient pre-processing method is developed to remove any bias from the images. Additionally, we evaluated the performance of the proposed model on the public COVID-CTset dataset to prove the generalisability of our model. The obtained results confirm the competence of the proposed network in confronting medical images, where prompt and accurate learning is required.

In Taiwan, peanuts are classified by hand. The kernel quality is judged based on the appearance of peanuts. The identification takes considerable time and labor. Fatigue can induce misrecogni-tion or grade cognition instability due. Peanut defects are generally divided into health, underde-velopment, insect bite, and rupture. This study employed two imaging methods push-boom FX10 and Snapshot. Deep learning was used to detect peanut defects. A push-broom instrument was used for testing. There were 1,560 peanuts, including 240 good and 240 bad ones. There were a total of 1,080 good peanuts and bad peanuts in the test set data, the band selection was used, and a CNN model was built. Based on the results from three methods, 3DCNN could classify with 97%accuracy. The Snapshot was used to achieve the real-time design of a lightweight CNN model. Finally, five bands were selected using PCA, and the screening could be accurate and efficient. The maximum overall accuracy was about 98.5%. Kappa was 97.3% with real-time recognition. This will be a major advantage in the future practical application and commercialization processes. This technique can reduce labor costs, achieve intelligent detection and intelligent grading, and bring breakthroughs to smart agriculture for other crops as well.

Machine Learning has completely transformed health care system, which transmits medical data through IOT sensors. So it is very important to encrypt them to protect patient data. encrypting medical images from a performance perspective consumes time; hence the use of an auto encoder is essential. An auto encoder is used in this work to compress the image as a vector prior to the encryption process. The digital image passes across description function and a decoder to get back the image in the proposed work; various experiments are carried out on hyper parameters to achieve the highest outcome of the classification. The findings demonstrate that the combination of Mean Square Logarithmic Error as the loss function, ADA grad as an optimizer, two layers for the encoder, and another reverse for the decoder, RELU as the activation function generates the best auto encoder results. The combination of Mean square error (lose function), RMS prop (optimizer), three layers for the encoder and another reverse for the decoder, and RELU (activation function) has the best classification result. All the experiments with different hyper parameter has run almost very close to each other even when changing the number of layers. The running time is between 9 and 16 second for each epoch.

Abstract: Imaging devices of less than 300,000 pixels are mostly used for sewage conduit exploration due to the petty nature of the survey industry in Korea. P articular ly , devices of less than 100,000 pixels are still widely used, and the environment for image processing is very bitter . Since the sewage conduit image s covered in this study ha ve a very low resolution (240 × 320 = 76,800 pixels), it is very difficult to detect cracks. Because most of the resolution of the sewe r conduit images are very low in Korea, this problem of low resolution was selected as the subject of study. Cracks were detected through a total of six steps of improving the crack in Step 2, finding the optimal threshold value in Step 3, and applying an algorithm to detect cracks in Step 5. Cracks were effectively detected by the optimal parameters in Steps 2 and 3 and the user algorithm in Step 5. Desp ite the very low resolution, the cracked image s showed 96.4% accuracy of detection, and the non cracked image s showed 94.5% accuracy . Moreover, the analysis was excellent in quality , also . It is believed that the findings of this study can be effectively u sed for crack detection with low resolution images.

CNN-based methods have dominated the field of aerial scene classification for the past few years. While achieving remarkable success, CNN-based methods suffer from excessive parameters and notoriously rely on large amounts of training data. In this work, we introduce few-shot learning to the aerial scene classification problem. Few-shot learning aims to learn a model on base-set that can quickly adapt to unseen categories in novel-set, using only a few labeled samples. To this end, we proposed a meta-learning method for few-shot classification of aerial scene images. First, we train a feature extractor on all base categories to learn a representation of inputs. Then in the meta-training stage, the classifier is optimized in the metric space by cosine distance with a learnable scale parameter. At last, in the meta-testing stage, the query sample in the unseen category is predicted by the adapted classifier given a few support samples. We conduct extensive experiments on two challenging datasets: NWPU-RESISC45 and RSD46-WHU. The experimental results show that our method yields state-of-the-art performance. Furthermore, several ablation experiments are conducted to investigate the effects of dataset scale, the impact of different metrics and the number of support shots; the experiment results confirm that our model is specifically effective in few-shot settings.

It has remained a hard nut for years to segment sonar images, most of which are noisy images with inevitable blur after noise reduction. For the purpose of solutions to this problem, a fast segmentation algorithm is proposed on the basis of the gray value characteristics of sonar images. This algorithm is endowed with the advantage in no need of segmentation thresholds to be calculated. To realize this goal, it follows the undermentioned steps: first, calculate the gray matrix of the fuzzy image background. After adjusting the gray value, segment the region into the background region, buffer region and target regions. After filtering, reset the pixels with gray value lower than 255 to binarize images and eliminate most artifacts. Finally, remove the remaining noise from images by means of morphological image processing. The simulation results of several sonar images show that the algorithm can segment the fuzzy sonar image quickly and effectively, with no problem of incomplete image target shape. Thus, the stable and feasible method is testified.

In the oil palm industry, stands age is an important parameter to monitor the sustainability of cultivation, to develop the growth yield model, to identify the disease or stressed area, and to estimate the carbon storage capacity. This research is focused to estimate and distinguish oil palm stands age based on crown/ canopy density obtained using Forest Canopy Density (FCD) model derived from four indices as follows; Advanced Vegetation Index, Bare Soil Index, Shadow Index, and Thermal Index. FCD model employs multi temporal image analysis resulting four classes of oil palm stands age categorized as seed with FCD value of 29–56% (0 years), young with FCD value of 56–63% (1–9 years), teen with FCD value of 63–80% (10–15 years), and mature with FCD value of >80% (>15 years). Minimum canopy density value is 29% even in the zero years old indicates incomplete land clearance or the type of seed planted in the land.

Speckle noise corrupt the major part of ultrasound image, because of which the quality deteriorate and loss of valuable information leads to false diagnosis. A large community of images like synthetic aperture radar (SAR) image, Synthetic image, and simulated ultrasound image, require despeckling at pre-processing stage for better processing. Cleaning the speckle from image and preserving the edge details is a vital task. Nowadays not only despeckling is considered as an important process but also preserving information at boundary and edges of image is also important. As most of the algorithms able to remove speckle noise but fails to preserve the details of edges. This paper covers several recent methods for removal of speckle noise along with various metrics opted for comparisons. The distinctive part of this paper is, a mathematical and parametric review has been done. Also a table is also included which summarizes the entire paper.

Growth monitoring of crops is a crucial aspect of precision agriculture, essential for optimal yield prediction and resource allocation. Traditional crop growth monitoring methods are labor-intensive and prone to errors. This study introduces an automated segmentation pipeline utilizing multi-date aerial images and orthomosaics to monitor the growth of cauliflower crops (Brassica Oleracea var. Botrytis). The methodology employs YOLOv8, Grounding DINO, and the Segment Anything Model (SAM) for automatic annotation and segmentation. The YOLOv8 model was trained using initial datasets, which then facilitated the training of the Grounded SAM framework. This approach generated automatic annotations and segmentation masks, classifying crop rows for temporal monitoring and growth estimation. The study's findings highlight the efficiency of this automated system in providing accurate crop growth analysis, promoting informed decision-making in crop management and sustainable agricultural practices. Results indicate consistent and comparable growth patterns between aerial images and orthomosaics, with significant periods of rapid expansion and minor fluctuations over time. The integration of these advanced techniques demonstrates the potential for enhancing precision agriculture through reduced labor and increased accuracy in crop growth monitoring.

Background and Objective: Segmentation of the femur in Dual Energy X-ray (DXA) images poses challenges due to reduced contrast, noise, bone shape variations, and inconsistent X-ray beam penetration. In this study, we investigate the relationship between noise and certain deep learning (DL) techniques for semantic segmentation of the femur to enhance segmentation and bone mineral density (BMD) accuracy by incorporating noise reduction methods into DL models. Methods: Convolutional neural network (CNN) based models were employed to segment femurs in DXA images and evaluate the effects of noise reduction filters on segmentation accuracy and its effect on BMD calculation. Various noise reduction techniques were integrated into DL-based models to enhance image quality before training. We assessed the performance of the fully convolutional neural network (FCNN) in comparison to noise reduction algorithms and manual segmentation methods. Results: Our study demonstrated that FCNN outperformed noise reduction algorithms in enhancing segmentation accuracy and enabling precise calculation of BMD. The FCNN-based segmentation approach achieved a segmentation accuracy of 98.84% and a correlation coefficient of 0.9928 for BMD measurements, indicating its effectiveness in the clinical diagnosis of osteoporosis. Conclusions: In conclusion, integrating noise reduction techniques into DL-based models significantly improves femur segmentation accuracy in DXA images. The FCNN model, in particular, shows promising results in enhancing BMD calculation and clinical diagnosis of osteoporosis. These findings highlight the potential of DL techniques in addressing segmentation challenges and improving diagnostic accuracy in medical imaging.

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.

Effective control of domain walls or magnetic textures in antiferromagnets promises to enable robust, fast, and non-volatile memories. The lack of net magnetic moment in antiferromagnets implies the need for creative ways to achieve such a manipulation. Here, we investigate changes in magnetic force microscopy (MFM) imaging and in magnon-related mode in Raman spectroscopy of virgin NiO films under a microwave pump. After the MFM and Raman studies, a combined action of broadband microwave (0.01-20 GHz, power scanned from −20 to 5 dBm) and magnetic field (up to 3 kOe) were applied to virgin epitaxial (111) NiO and (100) NiO films grown on (0001) Al2O3 and (100) MgO substrates, following which the MFM and Raman studies were repeated. We observed a suppression of the magnon-related Raman mode subsequent to the microwave exposure. Based on MFM imaging, this effect appears to be caused by the suppression of large antiferromagnetic domain walls due to the possible excitation of antiferromagnetic spin oscillations localized within the antiferromagnetic domain walls.

The changing of land use and land cover (LULC) are both affected by climate and human activity and affect climate, biological diversity, and human well-being. Accurate and timely information about the LULC pattern and change is crucial for land management decision-making, ecosystem monitoring, and urban planning, especially in developing economies undergoing industrialization, urbanization, and globalization. Biodiversity degradation and urban expansion in eastern China are research hot-spots. However, the influence of LULC changes on the region remains largely unexplored. Here, an object-based and multi-temporal image analysis approach was developed to detect how LULC changes during 1985-2015 in the Tiaoxi watershed (Zhejiang province, eastern China) using Landsat TM and OLI data. The main objective of this study is to improve the accuracy of unsupervised change detection from object-based and multi-temporal images. To this end, a total of seven LULC maps are generated with multi-temporal images. A random stratified sample design was used for assessing change detection accuracy. The proposed method achieved an overall accuracy of 91.86%, 92.14%, 92.00%, and 93.86% for 2000, 2005, 2010, and 2015, respectively. Nevertheless, the proposed method, in conjunction with object-oriented and multi-temporal satellite images, offers a robust and flexible approach to LULC changes mapping that helps with emergency response and government management. Urbanization and agriculture efficiency are the main reasons for LULC changes in the region. We anticipate that this freely available data will improve the modeling for surface forcing, provide evidence of changes in LULC, and inform water-management decision-making.

In this article, we analyse the computer aid screening method of COVID19 using Xray and CT scan images. The main objective is to set an analytical closure about the computer aid screening of COVID19 among the X-ray image and CT scan image. The computer aid screening method includes deep feature extraction, transfer learning and traditional machine learning image classification approach. The deep feature extraction and transfer learning method considered 13 pre-trained CNN model. The machine learning approach includes three sets of features and three classifiers. The pre-trained CNN models are alexnet, googlenet, vgg16, vgg19, densenet201, resnet18, resnet50, resnet101, inceptionv3, inceptionresnetv2, xception, mobilenetv2 and shufflenet. The features and classifiers in machine learning approaches are GLCM, LBP, HOG and KNN, SVM, Naive bay’s respectively. In addition, we also analyse the different paradigms of classifiers. In total, the comparative analysis is carried out in 65 classification models, i.e. 13 in deep feature extraction, 13 in transfer learning and 39 in machine learning approaches. Finally, all the classification models perform better in X-ray image set compare to CT scan image set.

The primary goal of picture compression is to reduce the amount of unused image data while still storing or transmitting it in a format that is appropriate. The compression of raw binary data is quite different from the compression of a picture, and these differences may be rather substantial. In light of this, compression is often regarded as an essential technique for the purposes of both data storage an d transmission in order to mitigate the excessive amounts of data that are generated by these images. In order to transmit enormous datasets, particularly for the purposes of telemedicine and teleradiology, one needs a significant amount of storage capacity as well as an expansive network. As a result, compression is an important aspect of medical imaging. In addition to the importance of compression, the quality of the photos themselves is also an essential factor in the success of analysis. In addition to this, the amount of time necessary to compress the photographs before sending them should be reduced.

The creation of crop-type maps from satellite data has proven challenging, often impeded by a lack of accurate in-situ data. This paper aims to demonstrate a method for crop-type (ie. Maize, Wheat and Other) recognition based on Convolutional Neural Networks using a bottom-up approach. We trained the model with a highly accurate dataset of crowdsourced labelled street-level imagery. Classification results achieved an AUC of 0.87 for wheat, 0.85 for maize and 0.73 for other. Given that wheat and maize are the two most common food crops globally, combined with an ever-increasing amount of available street-level imagery, this approach could help address the need for improved crop-type monitoring globally. Challenges remain in addressing the noisy aspect of street-level imagery (ie. buildings, hedgerows, automobiles, etc.), where a variety of different objects tend to restrict the view and confound the algorithms

The fast and reliable processing of medical images is of paramount importance to adequately generate data to feed machine learning algorithms that can prevent and diagnose health issues. Here we benchmark different compressed sensing techniques applied to magnetic resonance imaging as a means to reduce the acquisition time spent in the collection of data and signals that form the image. We show that by using these techniques, it is possible to reduce the number of signals needed and, therefore, substantially decrease the time to acquire the measurements. To this end, we have considered and compared different algorithms: the Iterative Re-Weighted Least Squares, the Iterative Soft Thresholding Algorithm, the Iterative Hard Thresholding Algorithm, the Primal Dual Algorithm and the Log-Barrier Algorithm. We have implemented such algorithms in different analysis programs that have been used to perform the reconstruction of the images and found that the Iterative Soft Thresholding Algorithm gives the optimal results. We found that the images obtained with this algorithm have less quality than the original ones, but the quality is good enough to distinguish each body structure and detect any health problems.

Land Surface Temperature (LST) is significant for climatological and environmental studies. LST products from satellites, however, suffer from the tradeoff between spatial and temporal resolution. Spatial downscaling has emerged as a well explored field aiming to overcome limitations arising from this tradeoff. Previous research on regression based LST downscaling models focused on utilizing predictors derived from optical imagery. Weather-dependency of optical imagery data, however, can influence downscaling models by the weather conditions. To cope this issue, in this study, we involve predictors derived from the weather-independent Sentinel-1 Synthetic Aperture Radar (SAR) imagery to downscale Landsat-8 LST data. In this context, we propose to use machine learning techniques, namely Random Forest (RF) and Convolutional Neural Networks (CNN). To demonstrate the applicability and performance of the proposed method, extensive experimental analyses were conducted over Zuid-Holland in the Netherlands. From the experiments, we found that the results obtained with radar predictors were comparable to those achieved using optical predictors. This confirms that the proposed method indeed paves a new way for mapping land surface temperature using SAR images.

Soil moisture maps are essential for hydrological, agricultural and risk assessment applications. To best meet these requirements, it is essential to develop soil moisture products at high spatial resolution which is now made possible using the free Sentinel-1 (S1) SAR (Synthetic Aperture Radar) data. Some soil moisture retrieval techniques using S1 data relied on the use of a priori weather information in order to increase the precision of soil moisture estimates, which required access to a weather forecasting framework. This paper presents an improved and fully automated solution for high-resolution soil moisture mapping in bare agricultural areas. The proposed solution derives a priori weather information directly from the original Sentinel images, thus bypassing the need for a weather forecasting framework. For soil moisture estimation, the neural network technique was implemented to ensure the optimum integration of radar information. The neural networks were trained using synthetic data generated by the modified Integral Equation Model (IEM) model and validated on real data from two study sites in France and Tunisia. Main findings showed that the use of radar signal averaged over grids of a few km2 in addition to radar signal at plot scale instead of a priori weather information, provides good soil moisture estimations. The accuracy is even slightly better comparatively to the accuracy obtained using a priori weather information.

Longitudinal image registration of pulmonary computed tomography (PCT) images may serve as an essential tool for investigating the relationship between radiation dose distribution and the occurrence and phenotype of radiation-induced lung disease (RILD). Although numerous longitudinal registration algorithms have been developed for PCT, most similarity-based approaches are not suitable for PCT involving RILD due to the complex tissue variation between two PCT images. Moreover, conventional feature-based approaches might fail to find a sufficient number of matched pairs of feature points due to the disparate lung deformation caused by breathing and RILD. To overcome the challenges resulting from RILD, component structure coherence point drift (CSCPD) was proposed to establish a deformation model by decomposing the chest into several components and matching them with individual parameters based on coherence point drift (CPD). Moreover, a regional vascular point matching (RVPM) algorithm was proposed to generate a vascular subtree and to substantially increase the number of corresponding pairs between two images. Eventually, the components were recomposed and aligned by a thin plate spline algorithm. A performance assessment on 15 pairs of PCT images of patients with RILD yielded recall and precision values of 0.85 and 0.89 for RVPM, respectively. Moreover, the target registration error of CSCPD with RVPM (2.3 ± 1.79) was significantly better than that of conventional CPD with RVPM (2.95 ± 1.89) and conventional CPD (5.04 ± 2.87). Therefore, the proposed registration system is robust enough to address the disparate deformation of lungs with RILD, and it improves registration accuracy within the parenchyma.

While thunderstorms can pose severe risks to property and life, forecasting remains challenging, even at short lead times, as these often arise in meta-stable atmospheric conditions. In this paper, we examine the question of how well we could perform short-term (up to 180min) forecasts using exclusively multi-spectral satellite images and past lighting events as data. We employ representation learning based on deep convolutional neural networks in an “end-to-end” fashion. Here, a crucial problem is handling the imbalance of the positive and negative classes appropriately in order to be able to obtain predictive results (which is not addressed by many previous machine-learning-based approaches). The resulting network outperforms previous methods based on physically-based features and optical flow methods (similar to operational prediction models) and generalizes across different years. A closer examination of the classifier performance over time and under masking of input data indicates that the learned model actually draws most information from structures in the visible spectrum, with infrared imaging sustaining some classification performance during the night.

This work presents an approach for detecting mathematical formulas in scanned document images. The proposed approach is end-to-end trainable. Since many OCR engines cannot reliably work with the formulas, it is essential to isolate them to obtain the clean text for information extraction from the document. Our proposed pipeline comprises a hybrid task cascade network with deformable convolutions and a Resnext101 backbone. Both of these modifications help in better detection. We evaluate the proposed approaches on the ICDAR-2017 POD and Marmot datasets and achieve an overall accuracy of 96% for the ICDAR-2017 POD dataset. We achieve an overall reduction of error of 13%. Furthermore, the results on Marmot datasets are improved for the isolated and embedded formulas. We achieved an accuracy of 98.78% for the isolated formula and 90.21% overall accuracy for embedded formulas. Consequently, it results in an error reduction rate of 43% for isolated and 17.9% for embedded formulas.

Single sample face recognition (SSFR) is a computer vision challenge. In this scenario, there is only one example from each individual on which to train the system, making it difficult to identify persons in unconstrained environments, particularly when dealing with changes in facial expression, posture, lighting, and occlusion. This paper suggests a different method based on a variant of the Binarized Statistical Image Features (BSIF) descriptor called Multi-Block Color-Binarized Statistical Image Features (MB-C-BSIF) to resolve the SSFR Problem. First, the MB-C-BSIF method decomposes a facial image into three channels (e.g., red, green, and blue), then it divides each channel into equal non-overlapping blocks to select the local facial characteristics that are consequently employed in the classification phase. Finally, the identity is determined by calculating the similarities among the characteristic vectors adopting a distance measurement of the k-nearest neighbors (K-NN) classifier. Extensive experiments on several subsets of the unconstrained Alex &amp; Robert (AR) and Labeled Faces in the Wild (LFW) databases show that the MB-C-BSIF achieves superior results in unconstrained situations when compared to current state-of-the-art methods, especially when dealing with changes in facial expression, lighting, and occlusion. Furthermore, the suggested method employs algorithms with lower computational cost, making it ideal for real-time applications.