Computer Vision Framework Research Articles

Artificial Intelligence Vision Methods for Robotic Harvesting of Edible Flowers.

Edible flowers, with their increasing demand in the market, face a challenge in labor-intensive hand-picking practices, hindering their attractiveness for growers. This study explores the application of artificial intelligence vision for robotic harvesting, focusing on the fundamental elements: detection, pose estimation, and plucking point estimation. The objective was to assess the adaptability of this technology across various species and varieties of edible flowers. The developed computer vision framework utilizes YOLOv5 for 2D flower detection and leverages the zero-shot capabilities of the Segmentation Anything Model for extracting points of interest from a 3D point cloud, facilitating 3D space flower localization. Additionally, we provide a pose estimation method, a key factor in plucking point identification. The plucking point is determined through a linear regression correlating flower diameter with the height of the plucking point. The results showed effective 2D detection. Further, the zero-shot and standard machine learning techniques employed achieved promising 3D localization, pose estimation, and plucking point estimation.

A computer-vision based framework for virtual 3D garment reconstruction

AbstractExisting 3D garment reconstruction methods are difficult to implement for online fashion design and e-commerce or special applications. This paper proposes a novel computer-vision framework for 3D garment digital reconstruction, which aims to reconstruct high-quality and realistic virtual 3D garments with fabric mechanic properties for 3D virtual try-on. The new segmentation scheme is proposed to separate the 3D garment point clouds from background points, which is suitable for 3D human shapes and is adaptive for different 3D garment models in different scenes. The new Statistical Outlier Removal algorithm and the learning-based method PointCleanNet are combined to remove noise and outliers in 3D garment point clouds, which provides high-fidelity and high-quality 3D garment point clouds. The 3D garment meshes are then reconstructed from their corresponding point clouds with a modified rolling ball algorithm. Finally, the meshes are improved and converted into physics-based virtual try-on 3D garments with fabric mechanic properties added, which enables the assessment of different body shapes with varied sizes for the same reconstructed 3D garment. Comparison experiments demonstrate that our framework achieves high-quality and realistic 3D garment reconstruction and accurate 3D virtual try-on from 2D garment images. We also demonstrate the proposed framework on a large range of various garments to show this approach has a great potential for garment future technology, such as online garment shopping, garment design and manufacturing.

Computer vision and statistical insights into cycling near miss dynamics

Across the globe, many transport bodies are advocating for increased cycling due to its health and environmental benefits. Yet, the real and perceived dangers of urban cycling remain obstacles. While serious injuries and fatalities in cycling are infrequent, “near misses”-events where a person on a bike is forced to avoid a potential crash or is unsettled by a close vehicle-are more prevalent. To understand these occurrences, researchers have turned to naturalistic studies, attaching various sensors like video cameras to bikes or cyclists. This sensor data holds the potential to unravel the risks cyclists face. Still, the sheer amount of video data often demands manual processing, limiting the scope of such studies. In this paper, we unveil a cutting-edge computer vision framework tailored for automated near-miss video analysis and for detecting various associated risk factors. Additionally, the framework can understand the statistical significance of various risk factors, providing a comprehensive understanding of the issues faced by cyclists. We shed light on the pronounced effects of factors like glare, vehicle and pedestrian presence, examining their roles in near misses through Granger causality with varied time lags. This framework enables the automated detection of multiple factors and understanding their significant weight, thus enhancing the efficiency and scope of naturalistic cycling studies. As future work, this research opens the possibility of integrating this AI framework into edge sensors through embedded AI, enabling real-time analysis.

Advanced Deep Learning Techniques for Indoor-Outdoor Scene Recognition Integrating CNN and Edge Detection for Enhanced Classification Accuracy in Dynamic Environments

Recognizing indoor-outdoor scenes is an vital portion of computer vision that has an impact on numerous zones, counting driverless driving, virtual reality, and natural following. This article talks approximately a more progressed profound learning framework that employments Convolutional Neural Systems (CNNs) and edge acknowledgment strategies to create classification more precise in settings that alter over time. Distinctive lighting, surfaces, and structure highlights can make it difficult for conventional scene acknowledgment strategies to tell the contrast between complex scenes. We recommend a way to solve these issues that employments the leading parts of both CNNs and edge acknowledgment to urge exact pictures of geometric shapes. The system begins by altering the photographs it gets by utilizing edge acknowledgment strategies, like Canny and Sobel, to discover critical structure edges and lower the clamor. The edge-enhanced pictures are at that point sent to a CNN engineering that was made to recognize scenes. The CNN show is learned on a huge dataset that incorporates a wide run of indoor and open air settings. This makes beyond any doubt that it works well in all sorts of circumstances. A modern layer integration strategy is utilized to connect the CNN's learned highlights with the edge-based highlights. This makes a difference demonstrate tell the contrast between complex scene points of interest superior. It comes about of our tests appear that our combined strategy makes classification much more exact than conventional CNN models. The edge acknowledgment portion truly makes strides the model's capacity to choose up on little changes in picture structures, which leads to more exact forecasts. Moreover, the framework's capacity to adjust to changing situations has been demonstrated by a parcel of testing in a part of distinctive circumstances. Combining profound learning with edge discovery might make scene acknowledgment way better, and the proposed strategy shows how this may work. This may lead to more advanced and viable computer vision frameworks in real-world circumstances that are complicated.

Airside Surveillance by Computer Vision in Low-Visibility and Low-Fidelity Environment

Low visibility at airports can significantly impact airside capacity, leading to ground delays and runway/taxiway incursions. Digital tower technology, enabled by live camera feeds, leverages computer vision to enhance airside surveillance and operational efficiency. However, technical challenges in digital camera systems can introduce low-fidelity transmission effects such as blurring, pixelation, or JPEG compression. Additionally, adverse weather conditions like rain and fog can further reduce visibility for tower controllers, whether from digital video or out-of-tower views. This paper proposes a computer vision framework and deep learning algorithms to detect and track aircraft in low-visibility (due to bad weather) and low-fidelity (due to technical issues) environments to enhance visibility using digital video input. The framework employs a convolutional neural network for aircraft detection and Kalman filters for tracking, especially in low-visibility conditions. Performance enhancements come from pre- and postprocessing algorithms like object filtering, corrupted image detection, and image enhancement. It proves effective on an airport video dataset from Houston Airport, enhancing visibility under adverse weather conditions.

Innovative cloud quantification: deep learning classification and finite-sector clustering for ground-based all-sky imaging

Abstract. Accurate cloud quantification is essential in climate change research. In this work, we construct an automated computer vision framework by synergistically incorporating deep neural networks and finite-sector clustering to achieve robust whole-sky image-based cloud classification, adaptive segmentation and recognition under intricate illumination dynamics. A bespoke YOLOv8 (You Only Look Once 8) architecture attains over 95 % categorical precision across four archetypal cloud varieties curated from extensive annual observations (2020) at a Tibetan highland station. Tailor-made segmentation strategies adapted to distinct cloud configurations, allied with illumination-invariant image enhancement algorithms, effectively eliminate solar interference and substantially boost quantitative performance even in illumination-adverse analysis scenarios. Compared with the traditional threshold analysis method, the cloud quantification accuracy calculated within the framework of this paper is significantly improved. Collectively, the methodological innovations provide an advanced solution to markedly escalate cloud quantification precision levels imperative for climate change research while offering a paradigm for cloud analytics transferable to various meteorological stations.

A Computer Vision Framework for Structural Analysis of Hand-Drawn Engineering Sketches.

Structural engineers are often required to draw two-dimensional engineering sketches for quick structural analysis, either by hand calculation or using analysis software. However, calculation by hand is slow and error-prone, and the manual conversion of a hand-drawn sketch into a virtual model is tedious and time-consuming. This paper presents a complete and autonomous framework for converting a hand-drawn engineering sketch into an analyzed structural model using a camera and computer vision. In this framework, a computer vision object detection stage initially extracts information about the raw features in the image of the beam diagram. Next, a computer vision number-reading model transcribes any handwritten numerals appearing in the image. Then, feature association models are applied to characterize the relationships among the detected features in order to build a comprehensive structural model. Finally, the structural model generated is analyzed using OpenSees. In the system presented, the object detection model achieves a mean average precision of 99.1%, the number-reading model achieves an accuracy of 99.0%, and the models in the feature association stage achieve accuracies ranging from 95.1% to 99.5%. Overall, the tool analyzes 45.0% of images entirely correctly and the remaining 55.0% of images partially correctly. The proposed framework holds promise for other types of structural sketches, such as trusses and frames. Moreover, it can be a valuable tool for structural engineers that is capable of improving the efficiency, safety, and sustainability of future construction projects.

A comprehensive end-to-end computer vision framework for restoration and recognition of low-quality engineering drawings

The digitization of engineering drawings is crucial for efficient reuse, distribution, and archiving. Existing computer vision approaches for digitizing engineering drawings typically assume the input drawings have high quality. However, in reality, engineering drawings are often blurred and distorted due to improper scanning, storage, and transmission, which may jeopardize the effectiveness of existing approaches. This paper focuses on restoring and recognizing low-quality engineering drawings, where an end-to-end framework is proposed to improve the quality of the drawings and identify the graphical symbols on them. The framework uses K-means clustering to classify different engineering drawing patches into simple and complex texture patches based on their gray level co-occurrence matrix statistics. Computer vision operations and a modified Enhanced Super-Resolution Generative Adversarial Network (ESRGAN) model are then used to improve the quality of the two types of patches, respectively. A modified Faster Region-based Convolutional Neural Network (Faster R-CNN) model is used to recognize the quality-enhanced graphical symbols. Additionally, a multi-stage task-driven collaborative learning strategy is proposed to train the modified ESRGAN and Faster R-CNN models to improve the resolution of engineering drawings in the direction that facilitates graphical symbol recognition, rather than human visual perception. A synthetic data generation method is also proposed to construct quality-degraded samples for training the framework. Experiments on real-world electrical diagrams show that the proposed framework achieves an accuracy of 98.98% and a recall of 99.33%, demonstrating its superiority over previous approaches. Moreover, the framework is integrated into a widely-used power system software application to showcase its practicality. The reference codes and data can be found at https://github.com/Lattle-y/AI-recognition-for-lq-ed.git Future work will focus on improving the generalizability of the proposed framework to different quality degradation scenarios and extrapolating the application to different engineering domains.

HivNet: Studying in Depth the Morphology of HIV-1 Virion Using Deep Learning

Purpose: Human Immunodeficiency Virus (HIV) continues to be a disease that kills thousands of individuals each year. The HIV infection is incurable. However, HIV infection has turned into a treatable chronic health condition because of improved access to efficient HIV prevention, diagnosis, treatment, and care. Transmission Electron Microscopy’s (TEM) ability to directly visualize virus particles and distinguish ultrastructure morphology at the nanometer scale, makes it useful in HIV-1 research where it is used for assessing the actions of inhibitors that obstruct the maturation and morphogenesis phases of the virus lifecycle. Hence with its use, the disease's serious stage can be avoided by receiving an early diagnosis.
 Materials and Methods: Through the dedicated use of computer vision frameworks and machine learning techniques, we have developed an optimized low-computational-cost 8-layer Convolutional Neural Network (CNN) backbone capable of classifying HIV-1 virions at various stages of maturity and morphogenesis. The dataset including TEM images of HIV-1 viral life cycle phases is analysed and augmented through various techniques to make the framework robust in real-time. The CNN layers then extract pertinent disease traits from TEM images and utilise them to provide diagnostic predictions.
 Results: It was discovered that the framework performed with an accuracy of 99.76% on the training set, 85.83% on the validation set, and 91.33% on the test set, after being trained on a wide range of micrographs which comprised of different experimental samples and magnifications.
 Conclusion: The suggested network's performance was compared to that of other state-of-the-art networks, and it was discovered that the proposed model was undisputed for classifying TEM images of unseen HIV-1 virion and required less time to train and tweak its weights. The framework can operate more effectively than machine learning algorithms that consume a lot of resources and can be deployed with limited computation and memory resource requirements.

UAV-Based Computer Vision System for Orchard Apple Tree Detection and Health Assessment

Accurate and efficient orchard tree inventories are essential for acquiring up-to-date information, which is necessary for effective treatments and crop insurance purposes. Surveying orchard trees, including tasks such as counting, locating, and assessing health status, plays a vital role in predicting production volumes and facilitating orchard management. However, traditional manual inventories are known to be labor-intensive, expensive, and prone to errors. Motivated by recent advancements in UAV imagery and computer vision methods, we propose a UAV-based computer vision framework for individual tree detection and health assessment. Our proposed approach follows a two-stage process. Firstly, we propose a tree detection model by employing a hard negative mining strategy using RGB UAV images. Subsequently, we address the health classification problem by leveraging multi-band imagery-derived vegetation indices. The proposed framework achieves an F1-score of 86.24% for tree detection and an overall accuracy of 97.52% for tree health assessment. Our study demonstrates the robustness of the proposed framework in accurately assessing orchard tree health from UAV images. Moreover, the proposed approach holds potential for application in various other plantation settings, enabling plant detection and health assessment using UAV imagery.

Forecasting passenger flows and headway at train level for a public transport line: Focus on atypical situations

This paper proposes a computer vision framework based on deep learning approaches for the real-time prediction of passenger loads and train headways in a metro line in an urban transit network. The short-term prediction problem is formulated as an image completion task. The train journeys on the metro line are represented as images, with the pixels denoting the train data, including the departure location, time, and load. Metro line applicative constraints, e.g., irregular time sampling of trains and univariate space, are considered in the images. Several deep learning-based architectures are investigated, including two new architectures based on transformers. Finally, an in-depth analysis and comparison of the models is performed based on a real test case of Paris metro line 9, for which a large database was collected over three years. This allowed us to include numerous instances with atypical transport system performance indicators (e.g., strikes, lockdowns, and disruptions) and to design a methodology (based on a latent space representation of the dataset) to identify and label interesting test cases. We evaluate and demonstrate the robustness of the proposed approaches numerically and study their limits for passenger loads and headways in atypical scenarios.

Distinguishing enchondroma from low-grade chondrosarcoma using a multi-stage deep learning model for patient radiographs and histopathology.

e23500 Background: Enchondroma and low-grade (grade 1) chondrosarcoma are cartilaginous tumors of the peripheral bone. Enchondroma is benign, requiring only routine clinical monitoring. Chondrosarcoma is malignant, demanding prompt and aggressive treatment to prevent metastasis. Despite drastically different treatment protocols, these tumors are histologically and radiographically similar, making them difficult to tell apart. Even specialized treatment, years of experience, and access to both imaging modalities, pathologists and radiologists have reported low reliability in distinguishing between these two tumors. Better diagnostic and grading strategies are needed for these lesions. In this study, we propose a two-stage computer vision framework for distinguishing enchondroma from low-grade chondrosarcoma using deep learning. Methods: Our team compiled a ground truth dataset of radiographic and histological images from the annotated records of Dr. Henry Jaffe, a pioneering authority in musculoskeletal pathology. From his records, 55 cases of low-grade chondrosarcoma and 19 cases of enchondroma containing both radiology and histology imaging were identified and included. Radiographs and histology samples corresponding to the same individual were paired together to create 290 paired images of enchondroma and 318 paired images of low-grade chondrosarcoma (pair = 1 histology + 1 radiograph). These image pairings were randomly split into training (~80%) or validation (~20%) groups, excepting minimal reallocations to prevent leakage between training and validation data. All images were preprocessed and augmented for improved model training. Image pairings (one histology image, one radiograph) corresponding to the same patient case were passed through separate VGG16 models pre-trained on ImageNet weights. These models produced two feature vectors as outputs, which were concatenated and subsequently passed through a second-stage binary classification network to predict tumor identity. Results: After 17 epochs, the model recorded training loss and accuracy of 0.6723 and 0.7583, and validation loss and accuracy of 0.7362 and 0.7373, respectively – an improvement upon diagnostic accuracies reported in previous studies. Conclusions: We have created a two-stage deep learning model that predicts whether patient radiographs and histology show enchondroma or low-grade chondrosarcoma. The trained model’s validation accuracy of 73.7% improves upon past reported methods. This suggests that computer vision methods that utilize multiple imaging modalities might augment physicians' capability to discriminate between these two tumor types.

Accurate delineation of individual tree crowns in tropical forests from aerial RGB imagery using Mask R‐CNN

AbstractTropical forests are a major component of the global carbon cycle and home to two‐thirds of terrestrial species. Upper‐canopy trees store the majority of forest carbon and can be vulnerable to drought events and storms. Monitoring their growth and mortality is essential to understanding forest resilience to climate change, but in the context of forest carbon storage, large trees are underrepresented in traditional field surveys, so estimates are poorly constrained. Aerial photographs provide spectral and textural information to discriminate between tree crowns in diverse, complex tropical canopies, potentially opening the door to landscape monitoring of large trees. Here we describe a new deep convolutional neural network method, Detectree2, which builds on the Mask R‐CNN computer vision framework to recognize the irregular edges of individual tree crowns from airborne RGB imagery. We trained and evaluated this model with 3797 manually delineated tree crowns at three sites in Malaysian Borneo and one site in French Guiana. As an example application, we combined the delineations with repeat lidar surveys (taken between 3 and 6 years apart) of the four sites to estimate the growth and mortality of upper‐canopy trees. Detectree2 delineated 65 000 upper‐canopy trees across 14 km2 of aerial images. The skill of the automatic method in delineating unseen test trees was good (F1 score = 0.64) and for the tallest category of trees was excellent (F1 score = 0.74). As predicted from previous field studies, we found that growth rate declined with tree height and tall trees had higher mortality rates than intermediate‐size trees. Our approach demonstrates that deep learning methods can automatically segment trees in widely accessible RGB imagery. This tool (provided as an open‐source Python package) has many potential applications in forest ecology and conservation, from estimating carbon stocks to monitoring forest phenology and restoration.Python package available to install at https://github.com/PatBall1/Detectree2.

A computer vision framework for quantification of feather growth patterns.

Feather growth patterns are important anatomical phenotypes for investigating the underlying genomic regulation of skin and epidermal appendage development. However, characterization of feather growth patterns previously relied on manual examination and visual inspection, which are both subjective and practically prohibitive for large sample sizes. Here, we report a new high-throughput technique to quantify the location and spatial extent of reversed feathers that comprise head crests in domestic pigeons. Phenotypic variation in pigeon feather growth patterns were rendered by computed tomography (CT) scans as point clouds. We then developed machine learning based, feature extraction techniques to isolate the feathers, and map the growth patterns on the skin in a quantitative, automated, and non-invasive way. Results from five test animals were in excellent agreement with "ground truth" results obtained via visual inspection, which demonstrates the viability of this method for quantification of feather growth patterns. Our findings underscore the potential and increasingly indispensable role of modern computer vision and machine learning techniques at the interface of organismal biology and genetics.

Cooperative multi-camera vehicle tracking and traffic surveillance with edge artificial intelligence and representation learning

Traffic surveillance cameras are the eyes of the Intelligent Transportation Systems (ITS). However, they are currently isolated and can only extract information from each of their fixed views. To track vehicles across multiple cameras and help public agencies collect link travel time and speed information, an Edge-empowered Cooperative Multi-camera Sensing (ECoMS) System is proposed. ECoMS system presents a novel algorithmic and edge-server cooperative system construct to push edge computing and multi-camera re-identification workflow serving for traffic sensing based on Internet of Things (IoT) architecture. On the algorithm side, ECoMS system proposes a featherlight edge-based computer vision framework for vehicle detection, tracking, and features selection process in a real-time manner. Then, by only sending the objects’ representations to the server, the high-bandwidth data transmission and the heavy post-processing system can be abandoned. Furthermore, a hierarchical clip-based deep vehicle re-identification framework is proposed and integrated into the ECoMS system, and significantly outperforms other state-of-the-art methods by 4%–8% on Rank-1 accuracy. Finally, to balance the accuracy level of different camera pairs, a collaborative cross-camera traffic information estimation framework based on kernel density estimation with kernel smoother is implemented, which can get the precise link and region traffic information together with distributions (less than 1.01 KL distance). By maximizing the cooperation of the computational resources, orchestrating the data transmission and integrating road network graph features in the system, the ECoMS can precisely model the network-scale traffic information in a flexible, cost-effective, and scalable workflow. To the author’s best knowledge, ECoMS is the first multi-camera vehicle tracking and traffic monitoring system based on cooperative IoT architecture.

Computer vision-based evaluation of dimensional accuracy for MEAM in new product development

Evaluation of dimensional accuracy for parts is a key component of quality assurance for material extrusion based additive manufacturing (MEAM) in new product development. Computer vision-based frameworks are currently the most suitable methods for MEAM dimensional accuracy evaluation due to their non-contact measurement characteristic, rapidity and high accuracy. However, the previous computer vision-based frameworks in MEAM had the limitations of high camera cost and strict installation requirements. To address this issue, a cheap and accurate computer vision-based framework is proposed for MEAM based on a single mobile phone camera without strict installation requirements. First, camera is calibrated to reduce the influence of lens distortion. Then, dimensions are measured through projective transformation and scaling. To find the best calibration pattern, three calibration pattern named checkerboard, symmetric circle and asymmetric circle are investigated. The proposed framework is validated experimentally through printing tests of a cube and a cylinder fabricated by a fused deposition modeling (FDM) 3D printer. The result shows that camera calibration using asymmetric circular grid pattern has the highest accuracy and the proposed computer vision framework is validated as the maximum relative error is less than 1%.

Towards an automatized and objective assessment of data from visual inspections of building envelopes

The renovation planning process is filled with uncertainties and subjective decisions. These make the decisions upon what and when to renovate a complex and ambiguous problem. Selection of renovation measures related to building envelope are often far from optimal as decisions are usually made based on visual inspections. These are manned and thus prone to subjective assessment and the knowhow of individual inspectors. Furthermore, objective criteria which could indicate non-structural failures are often missing. The objective based planning process allowing the estimation of the current damage status of the building envelope by only using non-destructive measurements is still in its infancy. The first step requires establishing reliable and objective based data collection. These could be efficiently collected by Unmanned Aerial Vehicles (UAV) with subsequent image recognition algorithms allowing the identification of imperfections and store the position and extent of such deviations into the building’s digital assessment database. Such tools do not exist. The aim of this study is to investigate the current objectivization possibilities in the domain of building inspections. The first part provides a literature review describing how an autonomous UAV survey of a building envelope may be planned and what computer vision techniques may be used for automatic damage recognition and classification. Subsequently, an objective detection model based on the YOLO-tiny (You Only Look Once) computer vision framework is employed in a case study investigating a building envelope of historical Tjolöholm castle in Sweden. This study contributes to developing a methodology for an objective based visual inspection process.

Automatic Screw Detection and Tool Recommendation System for Robotic Disassembly

Abstract Disassembly is an essential process for the recovery of end-of-life (EOL) electronics in remanufacturing sites. Nevertheless, the process remains labor-intensive due to EOL electronics’ high degree of uncertainty and complexity. The robotic technology can assist in improving disassembly efficiency; however, the characteristics of EOL electronics pose difficulties for robot operation, such as removing small components. For such tasks, detecting small objects is critical for robotic disassembly systems. Screws are widely used as fasteners in ordinary electronic products while having small sizes and varying shapes in a scene. To enable robotic systems to disassemble screws, the location information and the required tools need to be predicted. This paper proposes a computer vision framework for detecting screws and recommending related tools for disassembly. First, a YOLOv4 algorithm is used to detect screw targets in EOL electronic devices and a screw image extraction mechanism is executed based on the position coordinates predicted by YOLOv4. Second, after obtaining the screw images, the EfficientNetv2 algorithm is applied for screw shape classification. In addition to proposing a framework for automatic small-object detection, we explore how to modify the object detection algorithm to improve its performance and discuss the sensitivity of tool recommendations to the detection predictions. A case study of three different types of screws in EOL electronics is used to evaluate the performance of the proposed framework.

Monocular Camera Viewpoint-Invariant Vehicular Traffic Segmentation and Classification Utilizing Small Datasets.

The work presented here develops a computer vision framework that is view angle independent for vehicle segmentation and classification from roadway traffic systems installed by the Virginia Department of Transportation (VDOT). An automated technique for extracting a region of interest is discussed to speed up the processing. The VDOT traffic videos are analyzed for vehicle segmentation using an improved robust low-rank matrix decomposition technique. It presents a new and effective thresholding method that improves segmentation accuracy and simultaneously speeds up the segmentation processing. Size and shape physical descriptors from morphological properties and textural features from the Histogram of Oriented Gradients (HOG) are extracted from the segmented traffic. Furthermore, a multi-class support vector machine classifier is employed to categorize different traffic vehicle types, including passenger cars, passenger trucks, motorcycles, buses, and small and large utility trucks. It handles multiple vehicle detections through an iterative k-means clustering over-segmentation process. The proposed algorithm reduced the processed data by an average of 40%. Compared to recent techniques, it showed an average improvement of 15% in segmentation accuracy, and it is 55% faster than the compared segmentation techniques on average. Moreover, a comparative analysis of 23 different deep learning architectures is presented. The resulting algorithm outperformed the compared deep learning algorithms for the quality of vehicle classification accuracy. Furthermore, the timing analysis showed that it could operate in real-time scenarios.

Computer vision framework for crack detection of civil infrastructure—A review

Civil infrastructure (e.g., buildings, roads, underground tunnels) could lose its expected physical and functional conditions after years of operation. Timely and accurate inspection and assessment of such infrastructures are essential to ensure safety and serviceability, e.g., by preventing unsafe working conditions and hazards. Cracks, which are one of the most common distress, can indicate severe structural integrity issues that threaten the safety of the structure and people in the environment. As such, accurate, fast, and automatic detection of cracks on structure surfaces is a major issue for a variety of civil engineering applications. Due to advances in hardware data acquisition systems, significant progress has been made in the automatic detection and quantification of cracks in recent decades. This paper provides a comprehensive review of the research progress and prospects in computer vision frameworks for crack detection of civil infrastructures from multiple materials, including asphalt, concrete, and metal-like materials. The review encompasses major components of typical frameworks, i.e., data acquisition techniques, publicly available datasets, detection algorithms, and evaluation metrics. In particular, we provide a taxonomy of detection algorithms with a detailed discussion of the advantages, limitations, and application scenarios of the methods in each category, as well as the relationships between methods of different categories. We also discuss unsolved issues and key challenges in crack detection that could drive future research directions.