Search

This study proposes an innovative deep learning algorithm for pose estimation based on point clouds, aimed at addressing the challenges of pose estimation for objects affected by the environment. Previous research on using deep learning for pose estimation has primarily been conducted using RGB-D data. This paper introduces an algorithm that utilizes point cloud data for deep learning-based pose computation. The algorithm builds upon previous work by integrating PointNet + + technology and the classical Point Pair Features algorithm, achieving accurate pose estimation for objects across different scene scales. Additionally, an adaptive parameter-density clustering method suitable for point clouds is introduced, effectively segmenting clusters in varying point cloud density environments. This resolves the complex issue of parameter determination for density clustering in different point cloud environments and enhances the robustness of clustering. Furthermore, the LineMod dataset is transformed into a point cloud dataset, and experiments are conducted on the transformed dataset to achieve promising results with our algorithm. Finally, experiments under both strong and weak lighting conditions demonstrate the algorithm's robustness. [ABSTRACT FROM AUTHOR]

This review paper provides a comprehensive overview of machine vision pose measurement algorithms. The paper focuses on the state-of-the-art algorithms and their applications. The paper is structured as follows: the introduction in provides a brief overview of the field of machine vision pose measurement. Describes the commonly used algorithms for machine vision pose measurement. Discusses the factors that affect the accuracy and reliability of machine vision pose measurement algorithms. Summarizes the paper and provides future research directions. The paper highlights the need for more robust and accurate algorithms that can handle varying lighting conditions and occlusion. It also suggests that the integration of machine learning techniques may improve the performance of machine vision pose measurement algorithms. Overall, this review paper provides a comprehensive overview of machine vision pose measurement algorithms, their applications, and the factors that affect their accuracy and reliability. It provides a valuable resource for researchers and practitioners working in the field of computer vision. [ABSTRACT FROM AUTHOR]

In new energy battery manufacturing, machine vision is widely used in automated assembly scenarios for key parts. To improve the accuracy and real‐time multi‐target positioning recognition of flexible thin parts, this paper proposes a multi‐level feature fusion template matching algorithm based on the Gaussian pyramid. Firstly, the algorithm constructs a Gaussian pyramid by multi‐scale image construction. Secondly, considering the image features of each layer of the pyramid, this paper uses the grey‐based Fast Normalized Matching algorithm to obtain coarse positioning coordinates on the upper layer, and the improved Linemod‐2D algorithm is applied to the bottom layer image to get accurate positioning coordinates. Finally, the positioning coordinates returned from each layer are fused to obtain the final positioning coordinate. The experimental results show that the proposed algorithm achieves excellent performance in nickel plate positioning and recognition. It exhibits satisfactory performance in nickel sheet localization and recognition. In terms of angular error, repeat accuracy, and matching speed, it competes favourably with Halcon, VisionMaster, and SCISmart. Its positioning error closely approximates that of Halcon, effectively meeting the practical production demands for high‐speed feeding and high‐precision positioning. [ABSTRACT FROM AUTHOR]

Traffic sign detection is critical for autonomous driving technology. However, accurately detecting traffic signs in complex traffic environments remains challenge despite the widespread use of one‐stage detection algorithms known for their real‐time processing capabilities. In this paper, the authors propose a traffic sign detection method based on YOLO v8. Specifically, this study introduces the Space‐to‐Depth (SPD) module to address missed detections caused by multi‐scale variations of traffic signs in traffic scenes. The SPD module compresses spatial information into depth channels, expanding the receptive field and enhancing the detection capabilities for objects of varying sizes. Furthermore, to address missed detections caused by complex backgrounds such as trees, this paper employs the Select Kernel attention mechanism. This mechanism enables the model to dynamically adjust its focus and more effectively concentrate on key features. Additionally, considering the uneven distribution of training data, the authors adopted the WIoUv3 loss function, which optimizes loss calculation through a weighted approach, thereby improving the model's detection performance across various sizes and frequencies of instances. The proposed methods were validated on the CCTSDB and TT100K datasets. Experimental results demonstrate that the authors' method achieves substantial improvements of 3.2% and 5.1% on the mAP50 metric compared to YOLOv8s, while maintaining high detection speed, significantly enhancing the overall performance of the detection system. The code for this paper is located at https://github.com/dusongjie/TSD‐YOLO‐Small‐Traffic‐Sign‐Detection‐Based‐on‐Improved‐YOLO‐v8 [ABSTRACT FROM AUTHOR]

Background: Arabic image captioning (AIC) is the automatic generation of text descriptions in the Arabic language for images. Applies a transfer learning approach in deep learning to enhance computer vision and natural language processing. There are many datasets in English reverse other languages. Instead of, the Arabs researchers unanimously agreed that there is a lack of Arabic databases available in this field. Objective: This paper presents the improvement and processing of the available Arabic textual database using Google spreadsheets for translation and creation of AR. Flicker8k2023 dataset is an extension of the Arabic Flicker8k dataset available, it was uploaded to GitHub and made public for researches. Methods: An efficient model proposed using deep learning techniques by including two pre-training models (VGG16 and VGG19), to extract features from the images and build (LSTM and GRU) models to process textual prediction sequence. In addition to the effect of pre-processing the text in Arabic. Results: The adopted model outperforms better compared to the previous study in BLEU-1 from 33 to 40. Conclusions: This paper concluded that the biggest problem is the database available in the Arabic language. This paper has worked to increase the size of the text database from 24,276 to 32,364 thousand captions, where each image contains 4 captions. [ABSTRACT FROM AUTHOR]

The freshness of aquatic products is directly related to the safety and health of the people. Traditional methods of detecting the freshness of Conger myriaster rely on manual operations, which are labor‐intensive, inefficient, and highly subjective. This paper combines computer vision and the DWG‐YOLOv8 network model to establish an intelligent classification method for C. myriaster freshness. Through image augmentation, 484 C. myriaster samples were expanded to 2904 samples. The YOLOv8n model was improved by simplifying the network backbone, introducing Ghost convolution and the new DW‐GhostConv, thereby reducing the number of parameters and computational load. Test results show that the recognition accuracy of the DWG‐YOLOv8 model reached 98.958%, outperforming models such as ResNet18, Mobilenetv3 small, and Swin transformer v2 tiny. The model's parameter count is 16.609 K, the inference time is 57.80 ms, and the model size is only 102 KB. The research provides a reliable method for online intelligent and nondestructive detection of C. myriaster freshness. [ABSTRACT FROM AUTHOR]

Timely and accurate detection and recognition of traffic lights are critical for Autonomous Vehicles (AVs) to avoid crashes due to red light running. This paper integrates a new robust machine learning based solution by combining a Convolutional Neural Network (CNN) with computer vision techniques to achieve a real-time traffic light detector. The proposed detection and recognition algorithm is capable of recognizing traffic lights on low-power small-form platforms, which are lightweight, portable, and can be mounted on AVs in daylight scenarios. The LISA open-source dataset is utilized with augmentation methods to increase the accuracy of the solution. The proposed approach achieves 93.42% of accuracy at a speed of 30.01 Frames Per Second (FPS) on an NVIDIA Jetson Xavier platform without using hardware accelerators such as FPGA. This solution is expected to promote the quicker adoption and wider deployment of AVs by increasing the chances of avoiding crashes and ultimately saving lives. [ABSTRACT FROM AUTHOR]

Research into explainable artificial intelligence (XAI) methods has exploded over the past five years. It is essential to synthesize and categorize this research and, for this purpose, multiple systematic reviews on XAI mapped out the landscape of the existing methods. To understand how these methods have developed and been applied and what evidence has been accumulated through model training and analysis, we carried out a tertiary literature review that takes as input systematic literature reviews published between 1992 and 2023. We evaluated 40 systematic literature review papers and presented binary tabular overviews of researched XAI methods and their respective characteristics, such as the scope, scale, input data, explanation data, and machine learning models researched. We identified seven distinct characteristics and organized them into twelve specific categories, culminating in the creation of comprehensive research grids. Within these research grids, we systematically documented the presence or absence of research mentions for each pairing of characteristic and category. We identified 14 combinations that are open to research. Our findings reveal a significant gap, particularly in categories like the cross-section of feature graphs and numerical data, which appear to be notably absent or insufficiently addressed in the existing body of research and thus represent a future research road map. [ABSTRACT FROM AUTHOR]

Reconstructing 3D indoor scenes from 2D images has always been an important task in computer vision and graphics applications. For indoor scenes, traditional 3D reconstruction methods have problems such as missing surface details, poor reconstruction of large plane textures and uneven illumination areas, and many wrongly reconstructed floating debris noises in the reconstructed models. This paper proposes a 3D reconstruction method for indoor scenes that combines neural radiation field (NeRFs) and signed distance function (SDF) implicit expressions. The volume density of the NeRF is used to provide geometric information for the SDF field, and the learning of geometric shapes and surfaces is strengthened by adding an adaptive normal prior optimization learning process. It not only preserves the high-quality geometric information of the NeRF, but also uses the SDF to generate an explicit mesh with a smooth surface, significantly improving the reconstruction quality of large plane textures and uneven illumination areas in indoor scenes. At the same time, a new regularization term is designed to constrain the weight distribution, making it an ideal unimodal compact distribution, thereby alleviating the problem of uneven density distribution and achieving the effect of floating debris removal in the final model. Experiments show that the 3D reconstruction effect of this paper on ScanNet, Hypersim, and Replica datasets outperforms the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Due to the excellent results achieved by transformers in computer vision, more and more scholars have introduced transformers into the field of medical image segmentation. However, the use of transformers will make the model's parameters very large, which occupies a large amount of the computer's resources, making them very time-consuming during training. In order to alleviate this disadvantage, this paper explores a flexible and efficient search strategy that can find the best subnet from a continuous transformer network. The method is based on a learnable and uniform L1 sparsity constraint, which contains factors that reflect the global importance of the continuous search space in different dimensions, while the search process is simple and efficient, containing a single round of training. At the same time, in order to compensate for the loss of accuracy caused by the search, a pixel classification module is introduced into the model to compensate for the loss of accuracy in the model search process. Our experiments show that the model in this paper compresses 30% of the parameters and FLOPs used, while also showing a slight increase in the accuracy of the model on the Automatic Cardiac Diagnosis Challenge (ACDC) dataset. [ABSTRACT FROM AUTHOR]

In the field of studies on the "Neural Synapses" in the nervous system, its experts manually (or pseudo-automatically) detect the bio-molecule clusters (e.g., of proteins) in many TIRF (Total Internal Reflection Fluorescence) images of a fluorescent cell and analyze their static/dynamic behaviors. This paper proposes a novel method for the automatic detection of the bio-molecule clusters in a TIRF image of a fluorescent cell and conducts several experiments on its performance, e.g., mAP @ IoU (mean Average Precision @ Intersection over Union) and F1-score @ IoU, as an objective/quantitative means of evaluation. As a result, the best of the proposed methods achieved 0.695 as its mAP @ IoU = 0.5 and 0.250 as its F1-score @ IoU = 0.5 and would have to be improved, especially with respect to its recall @ IoU. But, the proposed method could automatically detect bio-molecule clusters that are not only circular and not always uniform in size, and it can output various histograms and heatmaps for novel deeper analyses of the automatically detected bio-molecule clusters, while the particles detected by the Mosaic Particle Tracker 2D/3D, which is one of the most conventional methods for experts, can be only circular and uniform in size. In addition, this paper defines and validates a novel similarity of automatically detected bio-molecule clusters between fluorescent cells, i.e., SimMolCC, and also shows some examples of SimMolCC-based applications. [ABSTRACT FROM AUTHOR]

Featured Application: This paper is applied in the industrial monitoring of the stamping process, where vibrations are common, necessitating the use of image registration and homography estimation methods to align template images with test images. By employing machine-vision and image-processing techniques, the process is monitored in real-time to detect any anomalies, ultimately aiming to protect the stamping molds. Homography estimation is a crucial task in aligning template images with target images in stamping monitoring systems. To enhance the robustness and accuracy of homography estimation against random vibrations and lighting variations in stamping environments, this paper proposes an improved unsupervised homography estimation model. The model takes as input the channel-stacked template and target images and outputs the estimated homography matrix. First, a specialized deformable convolution module and Group Normalization (GN) layer are introduced to expand the receptive field and enhance the model's ability to learn rotational invariance when processing large, high-resolution images. Next, a multi-scale, multi-stage unsupervised homography estimation network structure is constructed to improve the accuracy of homography estimation by refining the estimation through multiple stages, thereby enhancing the model's resistance to scale variations. Finally, stamping monitoring image data is incorporated into the training through data fusion, with data augmentation techniques applied to randomly introduce various levels of perturbation, brightness, contrast, and filtering to improve the model's robustness to complex changes in the stamping environment, making it more suitable for monitoring applications in this specific industrial context. Compared to traditional methods, this approach provides better homography matrix estimation when handling images with low texture, significant lighting variations, or large viewpoint changes. Compared to other deep-learning-based homography estimation methods, it reduces estimation errors and performs better on stamping monitoring images, while also offering broader applicability. [ABSTRACT FROM AUTHOR]

The article discusses the importance of mechanization in agriculture, particularly in developing countries where there is a shortage of agricultural labor due to rapid urbanization. It highlights the development of various farming machinery for soil preparation, sowing, fertilization, pest control, harvesting, and post-harvesting. The article presents research papers from five countries, covering topics such as improved measurement methods for agricultural tractors, optimizing the design of farming machinery, and the interaction between soil and mechanical systems. It also includes studies on soil properties and their impact on agricultural applications. The article concludes by expressing gratitude to the authors and reviewers for their contributions to the special issue. [Extracted from the article]

Introduction: Recently, the Transformer model and its variants have been a great success in terms of computer vision, and have surpassed the performance of convolutional neural networks (CNN). The key to the success of Transformer vision is the acquisition of short-term and long-term visual dependencies through selfattention mechanisms; this technology can efficiently learn global and remote semantic information interactions. However, there are certain challenges associated with the use of Transformers. The computational cost of the global self-attention mechanism increases quadratically, thus hindering the application of Transformers for high-resolution images. Methods: In view of this, this paper proposes a multi-view brain tumor segmentation model based on cross windows and focal self-attention which represents a novel mechanism to enlarge the receptive field by parallel cross windows and improve global dependence by using local fine-grained and global coarse-grained interactions. First, the receiving field is increased by parallelizing the self-attention of horizontal and vertical fringes in the cross window, thus achieving strong modeling capability while limiting the computational cost. Second, the focus on self-attention with regards to local fine-grained and global coarse-grained interactions enables the model to capture short-term and longterm visual dependencies in an efficient manner. Results: Finally, the performance of the model on Brats2021 verification set is as follows: dice Similarity Score of 87.28, 87.35 and 93.28%; Hausdorff Distance (95%) of 4.58 mm, 5.26 mm, 3.78 mm for the enhancing tumor, tumor core and whole tumor, respectively. Discussion: In summary, the model proposed in this paper has achieved excellent performance while limiting the computational cost. [ABSTRACT FROM AUTHOR]

This paper presents a textile defect detection method that utilizes a multi‐proportion spatial attention mechanism and channel memory feature fusion network by addressing the difficulties presented by complicated shapes and large size variations. In particular, a multi‐proportion spatial attention mechanism (MPAM) is introduced, which employs multi‐proportion convolution to improve the backbone network's capacity to detect non‐uniform structural defects. Additionally, the generality and adaptability of the model are enhanced by a multi‐scale spatial pyramid pooling structure (MS‐SPP). Second, a channel attention mechanism‐based memory feature fusion network is developed, which incorporates channel attention to adaptive weight the feature channels, focusing on crucial information channels to efficiently fuse contextual features and enhance the model's memory capacity. Finally, a novel efficient Wise‐IoU (EWIoU) loss function is proposed, which utilizes a dynamic non‐monotonic focusing mechanism to increase the penalty on distance measurement, thus enhancing the model's detection performance. Experiment findings on the ZJU‐Leaper and Tianchi textile datasets reveal that compared to the YOLOv7 baseline, the method in this paper has an increase of 6.5 and 2 percentage points, respectively, and the detection accuracy is better than most existing networks. [ABSTRACT FROM AUTHOR]

Image segmentation is a key problem in the field of machine vision. Its core goal is to separate the target and background in the region of interest from the image and directly affect the accuracy of subsequent operations such as target recognition and image understanding. In the past decades, there have been many good image segmentation algorithms. In recent years, the deep learning method represented by deep learning has made great progress in the field of image segmentation. In this paper, some commonly used image segmentation algorithms based on machine learning were reviewed, and their theoretical and experimental studies were carried out. In this paper, the application prospect of machine learning in image segmentation was prospected. The existing image segmentation methods are mainly divided into the following categories: threshold-based segmentation methods, region-based segmentation methods, edge-based segmentation methods, and segmentation methods based on specific theories. In recent years, with the rapid progress of computer vision technology, the requirements for the accuracy of object image edge information segmentation have become increasingly high. The main reason for image segmentation is to better obtain object information. However, due to interference conditions such as lighting and noise, image blurry edge information segmentation has become the most difficult point in the development of computer vision technology. In the comparative experiment of algorithms, the results showed that in the training set, the response time of Deep Neural Network (DNN) algorithm, Cluster Analysis (CA) algorithm, and Support Vector Machine (SVM) algorithm was 13.72 s, 16.88 s and 17.29 s when the number of samples was 150. In the test set, when the sample number was 50, the recognition rate of DNN algorithm was 93.7%; the recognition rate of CA algorithm was 87.9%; the recognition rate of SVM algorithm was 84.3%. Therefore, the research of image fuzzy edge information segmentation based on computer vision and machine learning is essential. [ABSTRACT FROM AUTHOR]

• Methodology to automatically generate geo-located waste container maps. • Use of Computer Vision algorithms to detect waste containers. • Automatic division into locations with and without containers in city maps. • Robust model with consistent performance disregarding the container type. • System evaluated in eleven Spanish cities with an average accuracy of 89%. This paper describes the scientific achievements of a collaboration between a research group and the waste management division of a company. While these results might be the basis for several practical or commercial developments, we here focus on a novel scientific contribution: a methodology to automatically generate geo-located waste container maps. It is based on the use of Computer Vision algorithms to detect waste containers and identify their geographic location and dimensions. Algorithms analyze a video sequence and provide an automatic discrimination between images with and without containers. More precisely, two state-of-the-art object detectors based on deep learning techniques have been selected for testing, according to their performance and to their adaptability to an on-board real-time environment: EfficientDet and YOLOv5. Experimental results indicate that the proposed visual model for waste container detection is able to effectively operate with consistent performance disregarding the container type (organic waste, plastic, glass and paper recycling,...) and the city layout, which has been assessed by evaluating it on eleven different Spanish cities that vary in terms of size, climate, urban layout and containers' appearance. [ABSTRACT FROM AUTHOR]

The objective of the study was to develop and test an automatic machine vision-based spraying robot for the detection, tracking, and spraying of artificial weeds by using LabVIEW programming language. The greenness method was used to distinguish green objects in the image. A time-controlled spray nozzle was run according to the presence of an artificial weed and its coordinates. A mobile test bench was built and the spraying system with a webcam was operated at speeds of 0.42, 0.54, 0.66, 0.78, and 0.90 km h-1, so as to be able to see the performance of the system. The amount of deposits on the ground in the spray pattern was evaluated on the test area and used in comparisons for site specific and broadcast spraying methods. A spraying solution containing brilliant sulpho-flavin (BSF) tracer (0.4 g L-1) and filter papers were used to compare the deposition of spray pattern achieved on the ground with both methods. According to the results, site-specific spraying application saved on average 89.48%, 79.98%, and 73.93% application volumes for 500 ms, 1000 ms, and 1500 ms spraying durations, respectively, at all spraying speeds is compared to broadcast spraying application. As one would expect, deposits on the filter papers decreased with increasing spraying speed. In addition, operating the system with 1000 ms nozzle controlled site specific spraying at different speeds did not cause a significant difference in the amount of deposits in the spray pattern and spraying accuracy as compared to the broadcast spraying method. [ABSTRACT FROM AUTHOR]

[Display omitted] • Two colorimetric paper-based sensors for creatinine determination are developed. • Signal readout is achieved with a smartphone with a self-written application. • App relies on computer-vision to enable automatic detection of the zone of interest. • The utility of the platform to determine creatinine in undiluted urine is proved. Creatinine is a clinically significant analyte used to diagnose kidney condition. However, the literature still lacks in creatinine sensors fulfilling point-of-care testing requirements. In this paper, we have developed colorimetric paper-based creatinine sensors adhering to point-of-care testing principles. The signal readout is accomplished with a smartphone modified with 3D-printed elements and processed with a self-written application compromising computer vision algorithm for automatic detection of the colored zone. Two colorimetric methods – routinely used Jaffé method and an alternative one with 3,5-dinitrobenzoic acid were both tested and compared. Hue channel intensity from HSV color space and green channel from RGB color space was used as the analytical signal in Jaffé and 3,5-dinitrobenzoate method, respectively. For both kinds of sensors the linear range of the response covered the range significant for urinary analysis, with precision, expressed as RSD, below 5%. Limit of quantification for Jaffé method was 1.05 mmol·L−1 whereas it was 0.82 mmol·L−1 for 3,5-dinitrobenzoate method. The utility of the developed sensors to selectively quantify creatinine in undiluted urine was proved using artificial urine samples and the obtained recoveries were in the range from 70 to 129 %. [ABSTRACT FROM AUTHOR]

The text is a preface to the Journal of Computer Science & Technology's special section on the Computational Visual Media (CVM) conference series. The CVM conference serves as an international platform for the exchange of research ideas and computational methodologies related to visual media. The conference encourages collaboration across disciplines such as computer graphics, computer vision, machine learning, and more. The 12th international conference in the series, CVM 2024, was held in Wellington, New Zealand, and attracted attention from researchers worldwide. The special section includes six outstanding papers covering various topics, as well as an invited survey paper. The editors express gratitude to the authors and reviewers for their contributions. The preface also provides brief biographies of the leading editor and guest editors. [Extracted from the article]

This document is a guest editorial for a special issue of the CAAI Transactions on Intelligence Technology journal. The special issue focuses on advances in representation learning for computer vision. The editorial highlights the success of deep learning methods in deriving powerful representations from visual data, but also acknowledges the challenges of conducting representation learning with deep models, especially with large and noisy datasets. The document provides summaries of several research papers included in the special issue, covering topics such as cancellable biometrics, medical image analysis, watermarking for medical images, facial pattern description, multi-biometric strategies, semantic segmentation, image enhancement, image classification, and hyperspectral image super-resolution. The authors express their hope that these papers will enhance readers' understanding of current trends and guide future research in the field. The document also includes brief biographies of the authors. [Extracted from the article]

Waste segregation is an essential aspect of a smoothly functioning waste management system. Usually, various recyclable waste types are disposed of together at the source, and this brings in the necessity to segregate them into their categories. Dry waste needs to be separated into its own categories to ensure that the proper procedures are implemented to treat and process it, which leads to an overall increased recycling rate and reduced landfill impact. Paper, plastics, metals, and glass are just a few examples of the many dry waste materials that can be recycled or recovered to create new goods or energy. Over the past years, much research has been conducted to devise effective and productive ways to achieve proper segregation for the waste that is being produced at an ever-increasing rate. This article introduces a multi-class garbage segregation system employing the YOLOv5 object detection model. Our final prototype demonstrates the capability of classifying dry waste categories and segregating them into their respective bins using a 3D-printed robotic arm. Within our controlled test environment, the system correctly segregated waste classes, mainly paper, plastic, metal, and glass, eight out of 10 times successfully. By integrating the principles of artificial intelligence and robotics, our approach simplifies and optimizes the traditional waste segregation process. [ABSTRACT FROM AUTHOR]

Unmanned Aerial Vehicles (UAVs) are revolutionising diverse computer vision use case domains, from public safety surveillance to Search and Rescue (SAR), and other emergency management and disaster relief operations. The growing need for accurate face verification algorithms has prompted an exploration of synergies between UAVs and face verification. This promises cost-effective, wide-area, non-intrusive person verification. Real-world human-centric use cases such as a ”Drone Guard Angel” for vulnerable people can contribute to public safety management and offload significant police resources. These scenarios demand efficient face verification to distinguish correctly the end users for authentication, authorisation and customised services. This paper investigates the suitability of existing solutions, and analyses five state-of-the-art candidate face verification algorithms. Informed by the advantages and disadvantages of existing solutions, the paper proposes an extended dataset and a refined face verification pipeline. Subsequently, it conducts empirical evaluation of these algorithms using the proposed pipeline and dataset in terms of inference times and the distribution of the similarity indexes. Furthermore, this paper provides essential guidance for algorithm selection and deployment in UAV-based applications. Two candidate algorithms, ArcFace and FaceNet512, have emerged as the top performers. The choice between them will depend on the specific use case requirements. [ABSTRACT FROM AUTHOR]

3D human motion prediction, predicting future poses from a given sequence, is an issue of great significance and challenge in computer vision and machine intelligence, which can help machines in understanding human behaviors. Due to the increasing development and understanding of Deep Neural Networks (DNNs) and the availability of large-scale human motion datasets, the human motion prediction has been remarkably advanced with a surge of interest among academia and industrial community. In this context, a comprehensive survey on 3D human motion prediction is conducted for the purpose of retrospecting and analyzing relevant works from existing released literature. In addition, a pertinent taxonomy is constructed to categorize these existing approaches for 3D human motion prediction. In this survey, relevant methods are categorized into three categories: human pose representation , network structure design , and prediction target. We systematically review all relevant journal and conference papers in the field of human motion prediction since 2015, which are presented in detail based on proposed categorizations in this survey. Furthermore, the outline for the public benchmark datasets, evaluation criteria, and performance comparisons are respectively presented in this paper. The limitations of the state-of-the-art methods are discussed as well, hoping for paving the way for future explorations. [ABSTRACT FROM AUTHOR]

The wire harness assembly process is a complicated manufacturing activity, which is becoming more complex because of the evolving nature of mechatronic and electronic products that require more connectors, sensors, controllers, communication networking, etc. Furthermore, the demand for wire harnesses continues to grow in all industries worldwide as the majority of equipment, appliances, machinery, vehicles, etc., are becoming "smart" (i.e., more mechatronic or electronic). Moreover, most of the wire harness assembly process tasks are done manually, and most of these are considered non-ergonomic for human assembly workers. Hence, the wire harness manufacturing industry is faced with the challenge of increasing productivity while improving the occupational health of its human assembly workers. The purpose of this paper is to conduct a literature review exploring the state of the use of collaborative robots in the wire harness assembly process due to their potential to reduce current occupational health problems for human assembly workers and increase the throughput of wire harness assembly lines, and to provide main findings, discussion, and further research directions for collaborative robotics in this application domain. Eleven papers were found in the scientific literature. All papers demonstrated the potential of collaborative robots to improve the productivity of wire harness assembly lines, and two of these in particular on the ergonomics of the wire harness assembly process. None of the papers reviewed presented a cost–benefit or a cycle time analysis to qualitatively and/or quantitatively measure the impact of the incorporation of collaborative robots in the wire harness assembly process. This represents an important area of opportunity for research with relevance to industry. Three papers remark on the importance of the integration of computer vision systems into a collaborative wire harness assembly process to make this more versatile as many types of wire harnesses exist. The literature review findings call for further research and technological developments in support of the wire harness manufacturing industry and its workers in four main categories: (i) Collaborative Robotics and Grippers, (ii) Ergonomics, (iii) Computer Vision Systems, and (iv) Implementation Methodologies. [ABSTRACT FROM AUTHOR]

In reviewing this Special Issue, various topics have been addressed, predominantly machine learning techniques and heuristic search algorithms. Machine learning and computational intelligence are currently high-profile research areas attracting the attention of many researchers. In the first paper, L. Zhao and H. Jin improved the traditional vector-weighted optimization algorithm (INFO) and designed a promising optimization algorithm (IDEINFO) [[8]]. [Extracted from the article]

Machine vision is a driver for novel imaging solutions covering wavelength range from visible to SWIR light. Emberion produces high‐performance industrial cameras based on its own monolithically CMOS integrated image sensors that are built on two unique innovations: 1) using a combination of colloidal quantum dots and graphene, and 2) optimising the measurement principles and ROICs according to the requirements of these nanomaterials. Characteristics and performance of our photodetector structures, image sensors and cameras are discussed, highlighting the benefits of graphene electronics. Our devices have photoresponse from 400 nm to 2100 nm, with low noise at high frame rates. [ABSTRACT FROM AUTHOR]

This paper presents a high‐performance capacitive touch system enhanced by a 3D vision‐based engine, the "AirTouch". The system is based on multiple 3D cameras that analyze the 50 cm near‐field volume above the screen surface. A real‐time 50 Hz 3D reconstruction engine with millimetric precision enables the sensing and tracking of several users and multiple hands or objects. The system identifies a preliminary set of hand poses: open palm, closed fist and pointing. For the pointing pose, the pointing target is estimated. It also determines the hand location, angle and movement of the fingers through advanced algorithms. The system aims to characterize the skeleton key elements including joints and segments of the hand, and reconstruct its mechanical properties (movement restrictions, angle correlations, etc.). [ABSTRACT FROM AUTHOR]

This document is a special issue of the journal "Electronics" that focuses on recent advances in computer vision. The introduction explains how computer vision has transformed various industries and daily life by enabling machines to interpret and understand visual information. It also highlights the challenges that still exist in the field, such as model robustness and interpretability. The future of computer vision is discussed, including the development of multimodal models and advancements in areas like self-supervised learning. The special issue includes 10 papers that cover a range of topics, including stereo matching, low-light image enhancement, automated test grading, image segmentation, virtual clothing design, large-scale learning, camera pose estimation, few-shot segmentation, and image-to-audio conversion. The papers present novel studies, approaches, and reviews that contribute to the advancement of computer vision. The document concludes by emphasizing the importance of computer vision in addressing various challenges and the need for ongoing research and interdisciplinary collaboration to tackle complex real-world problems. [Extracted from the article]

The introduction of YOLOv9, the latest version of the you only look once (YOLO) series, has led to its widespread adoption across various scenarios. This paper is the first to apply the YOLOv9 algorithm model to the fracture detection task as computer‐assisted diagnosis to help radiologists and surgeons to interpret X‐ray images. Specifically, this paper trained the model on the GRAZPEDWRI‐DX dataset and extended the training set using data augmentation techniques to improve the model performance. Experimental results demonstrate that compared to the mAP 50–95 of the current state‐of‐the‐art model, the YOLOv9 model increased the value from 42.16% to 43.73%, with an improvement of 3.7%. The implementation code is publicly available at https://github.com/RuiyangJu/YOLOv9‐Fracture‐Detection. [ABSTRACT FROM AUTHOR]

Image processing methods that take place in different areas of life from identification to traffic flow control are starting to be used on mobile devices. Hence, usage of optical mark recognition systems (OMRS) on mobile devices is also not widespread as it is for image processing applications. A mobile OMRS is developed using image processing methods. In this developed system, image enhancement, edge detection, and tilt-shifting operations were performed on the optical answer paper images that are obtained using mobile devices. Within the scope of this study, it is aimed to achieve the maximum success of image processing operations on a mobile device that has limited hardware resources. The evaluation of optical answer papers that are obtained using mobile devices is performed using the image processing methods and algorithms developed within the scope of this study. The developed system is tested using different type of optical answer papers. According to the test results, the developed system provides a flexible and efficient use and partly eliminates previous work constraints. The success rate of the developed system is determined to be 93.54% using the test results obtained. © 2018 SPIE and IS&T [DOI: 10.1117/1.JEI.27.3.033007] [ABSTRACT FROM AUTHOR]

Transformer model architectures have recently received great interest in natural language, machine translation, and computer vision, where attention mechanisms are their building blocks. However, the attention mechanism is expensive because of its intensive matrix computations and complicated data flow. The existing hardware architecture has some disadvantages for the computing structure of attention, such as in- flexibility and low efficiency. Most of the existing papers accelerate attention by reducing the amount of computation through various pruning algorithms, which will affect the results to a certain extent with different sparsity. This paper proposes the hardware accelerator for the multi-head attention (MHA) on field-programmable gate arrays (FPGAs) with reconfigurable architecture, efficient systolic array, and hardware-friendly radix-2 softmax. We propose a novel method called Four inputs Processing Element (FPE) to double the computation rate of the data-aware systolic array (SA) and make it efficient and load balance. Especially, the computation framework is well designed to ensure the utilization of SA efficiently. Our design is evaluated on a Xilinx Alveo U250 card, and the proposed architecture achieves 51.3x, 17.3x improvement in latency, and 54.4x, 17.9x energy savings compared to CPU and GPU. [ABSTRACT FROM AUTHOR]

With the advancement of image sensing technology, estimating 3Dhuman pose frommonocular video has become a hot research topic in computer vision. 3D human pose estimation is an essential prerequisite for subsequent action analysis and understanding. It empowers a wide spectrum of potential applications in various areas, such as intelligent transportation, human-computer interaction, and medical rehabilitation. Currently, some methods for 3D human pose estimation in monocular video employ temporal convolutional network (TCN) to extract inter-frame feature relationships, but the majority of them suffer from insufficient inter-frame feature relationship extractions. In this paper, we decompose the 3D joint location regression into the bone direction and length, we propose the TCG, a temporal convolutional network incorporating Gaussian error linear units (GELU), to solve bone direction. It enablesmore inter-frame features to be captured andmakes the utmost of the feature relationships between data. Furthermore, we adopt kinematic structural information to solve bone length enhancing the use of intra-frame joint features. Finally, we design a loss function for joint training of the bone direction estimation network with the bone length estimation network. The proposed method has extensively experimented on the public benchmark dataset Human3.6M. Both quantitative and qualitative experimental results showed that the proposed method can achieve more accurate 3D human pose estimations. [ABSTRACT FROM AUTHOR]

Semantic segmentation is a hot topic in computer vision and various deep learning networks are designed to achieve higher accuracy on that by fully exploring the capability of neural networks. This paper aims to address the issue and proposes the substructures with novelty for popular networks. Meanwhile, we present a cross‐channel structure, which simultaneously reduces parameter while the kernel size becomes larger. After that, to overcome the weakness of insufficient dataset which refers to satellite image data, we propose a feature attention mechanism with generative adversarial network to enhance the images' features. We show the recognition result on the satellite image dataset with a large picture. This paper evaluates substructures on the PASCAL VOC2012 dataset and improves the mIOU from 74.68% to 88.15%. [ABSTRACT FROM AUTHOR]

Due to the self-affine property of the grinding surface, the sample images with different roughness captured by the micron-scale camera exhibit certain similarities. This similarity affects the prediction accuracy of the deep learning model. In this paper, we propose an illumination method that can mitigate the impact of self-affinity using the two-scale fractal theory as a foundation. This is followed by the establishment of a machine vision detection method that integrates a neural network and correlation function. Initially, a neural network is employed to categorize and forecast the microscopic image of the workpiece surface, thereby determining its roughness category. Subsequently, the corresponding correlation function is determined in accordance with the established roughness category. Finally, the surface roughness of the workpiece was calculated based on the correlation function. The experimental results demonstrate that images obtained using this lighting method exhibit significantly enhanced accuracy in neural network classification. In comparison to traditional lighting methods, the accuracy of this method on the micrometer scale has been found to have significantly increased from approximately 50% to over 95%. Concurrently, the mean squared error (MSE) of the surface roughness calculated by the proposed method does not exceed 0.003, and the mean relative error (MRE) does not exceed 5%. The two-scale fractal geometry offers a novel approach to image processing and machine learning, with significant potential for advancement. [ABSTRACT FROM AUTHOR]

Human pose estimation is an important task in computer vision, which can provide key point detection of human body and obtain bone information. At present, human pose estimation is mainly utilized for detection of large targets, and there is no solution for detection of small targets. This paper proposes a multi-channel spatial information feature based human pose (MCSF-Pose) estimation algorithm to address the issue of medium and small targets inaccurate detection of human key points in scenarios involving occlusion and multiple poses. The MCSF-Pose network is a bottom-up regression network. Firstly, an UP-Focus module is designed to expand the feature information while reducing parameter computation during the up-sampling process. Then, the channel segmentation strategy is adopted to cut the features, and the feature information of multiple dimensions is retained through different convolutional groups, which reduces the parameter lightweight network model and makes up for the loss of the feature information associated with the depth of the network. Finally, the three-layer PANet structure is designed to reduce the complexity of the model. With the aid of the structure, it also to improve the detection accuracy and anti-interference ability of human key points. The experimental results indicate that the proposed algorithm outperforms YOLO-Pose and other human pose estimation algorithms on COCO2017 and MPII human pose datasets. [ABSTRACT FROM AUTHOR]

Automatic translation from signed to spoken languages is an interdisciplinary research domain on the intersection of computer vision, machine translation (MT), and linguistics. While the domain is growing in terms of popularity—the majority of scientific papers on sign language (SL) translation have been published in the past five years—research in this domain is performed mostly by computer scientists in isolation. This article presents an extensive and cross-domain overview of the work on SL translation. We first give a high level introduction to SL linguistics and MT to illustrate the requirements of automatic SL translation. Then, we present a systematic literature review of the state of the art in the domain. Finally, we outline important challenges for future research. We find that significant advances have been made on the shoulders of spoken language MT research. However, current approaches often lack linguistic motivation or are not adapted to the different characteristics of SLs. We explore challenges related to the representation of SL data, the collection of datasets and the evaluation of SL translation models. We advocate for interdisciplinary research and for grounding future research in linguistic analysis of SLs. Furthermore, the inclusion of deaf and hearing end users of SL translation applications in use case identification, data collection, and evaluation, is of utmost importance in the creation of useful SL translation models. [ABSTRACT FROM AUTHOR]

The rapid evolution of wireless communication technologies has underscored the critical role of antennas in ensuring seamless connectivity. Antenna defects, ranging from manufacturing imperfections to environmental wear, pose significant challenges to the reliability and performance of communication systems. This review paper navigates the landscape of antenna defect detection, emphasizing the need for a nuanced understanding of various defect types and the associated challenges in visual detection. This review paper serves as a valuable resource for researchers, engineers, and practitioners engaged in the design and maintenance of communication systems. The insights presented here pave the way for enhanced reliability in antenna systems through targeted defect detection measures. In this study, a comprehensive literature analysis on computer vision algorithms that are employed in end-of-line visual inspection of antenna parts is presented. The PRISMA principles will be followed throughout the review, and its goals are to provide a summary of recent research, identify relevant computer vision techniques, and evaluate how effective these techniques are in discovering defects during inspections. It contains articles from scholarly journals as well as papers presented at conferences up until June 2023. This research utilized search phrases that were relevant, and papers were chosen based on whether or not they met certain inclusion and exclusion criteria. In this study, several different computer vision approaches, such as feature extraction and defect classification, are broken down and analyzed. Additionally, their applicability and performance are discussed. The review highlights the significance of utilizing a wide variety of datasets and measurement criteria. The findings of this study add to the existing body of knowledge and point researchers in the direction of promising new areas of investigation, such as real-time inspection systems and multispectral imaging. This review, on its whole, offers a complete study of computer vision approaches for quality control in antenna parts. It does so by providing helpful insights and drawing attention to areas that require additional exploration. [ABSTRACT FROM AUTHOR]

One of the key tasks to ensure infrastructure safety is the periodic detection of bridge cracks. Since manual crack detection is subjective and inefficient, it is very important to develop an automatic crack recognition system by using machine vision. Inspired by the pyramid vision transformer (PVT) and the feature pyramid network (FPN) variants, a method combining PVT, residual transformer (REST), holistically nested edge detection (HED), and downstream detection tasks is proposed in this paper, which is named EVPTMFF (efficient visual pyramid transformer and multiple-feature fusion). Based on the PVT, the multiheaded attention module was replaced and the efficient attention module was adopted, which could process the data efficiently and flexibly. To improve the performance of EVPTMFF, the original perceptual field windows were changed. The adjacent windows were partially overlapped, which was more conducive to feature interaction and improves detection performance. To prove the generalization ability of the model, three different data sets related to bridges were collected and formed. We carried out experiments on these three data sets, and EVPTMFF showed good results. Especially for larger data sets, the performance advantage was more significant. Practical Applications: The crack detection model proposed in this paper presents a good detection effect under different illumination and interference. And the detection results, collected data, time, and other information are saved to the bridge crack detection software. This can help engineers quickly and accurately detect cracks on the bridge surface, as well as predict the development trend of cracks and possible safety issues. In practical application, the bridge crack detection system can help engineers find and solve the bridge crack problem in time and avoid the security risks and economic losses caused by cracks. At the same time, the efficiency and accuracy of bridge maintenance can be improved, and the maintenance cost and time can be reduced. The bridge crack detection system can be integrated with other hardware equipment and management systems to form a complete bridge management platform, which contributes to the traffic construction and social and economic development of the city. [ABSTRACT FROM AUTHOR]