Deep Learning Based Vehicle Recognition Schemes For Intelligent Transportation Systems
Download Deep Learning Based Vehicle Recognition Schemes For Intelligent Transportation Systems PDF/ePub or read online books in Mobi eBooks. Click Download or Read Online button to get Deep Learning Based Vehicle Recognition Schemes For Intelligent Transportation Systems book now. This website allows unlimited access to, at the time of writing, more than 1.5 million titles, including hundreds of thousands of titles in various foreign languages.
Deep Learning-Based Vehicle Recognition Schemes for Intelligent Transportation Systems
With the increasing highlighted security concerns in Intelligent Transportation System (ITS), Vision-based Automated Vehicle Recognition (VAVR) has attracted considerable attention recently. A comprehensive VAVR system contains three components: Vehicle Detection (VD), Vehicle Make and Model Recognition (VMMR), and Vehicle Re-identification (VReID). These components perform coarse-to-fine recognition tasks in three steps. The VAVR system can be widely used in suspicious vehicle recognition, urban traffic monitoring, and automated driving system. Vehicle recognition is complicated due to the subtle visual differences between different vehicle models. Therefore, how to build a VAVR system that can fast and accurately recognize vehicle information has gained tremendous attention. In this work, by taking advantage of the emerging deep learning methods, which have powerful feature extraction and pattern learning abilities, we propose several models used for vehicle recognition. First, we propose a novel Recurrent Attention Unit (RAU) to expand the standard Convolutional Neural Network (CNN) architecture for VMMR. RAU learns to recognize the discriminative part of a vehicle on multiple scales and builds up a connection with the prominent information in a recurrent way. The proposed ResNet101-RAU achieves excellent recognition accuracy of 93.81% on the Stanford Cars dataset and 97.84% on the CompCars dataset. Second, to construct efficient vehicle recognition models, we simplify the structure of RAU and propose a Lightweight Recurrent Attention Unit (LRAU). The proposed LRAU extracts the discriminative part features by generating attention masks to locate the keypoints of a vehicle (e.g., logo, headlight). The attention mask is generated based on the feature maps received by the LRAU and the preceding attention state generated by the preceding LRAU. Then, by adding LRAUs to the standard CNN architectures, we construct three efficient VMMR models. Our models achieve the state-of-the-art results with 93.94% accuracy on the Stanford Cars dataset, 98.31% accuracy on the CompCars dataset, and 99.41% on the NTOU-MMR dataset. In addition, we construct a one-stage Vehicle Detection and Fine-grained Recognition (VDFG) model by combining our LRAU with the general object detection model. Results show the proposed VDFG model can achieve excellent performance with real-time processing speed. Third, to address the VReID task, we design the Compact Attention Unit (CAU). CAU has a compact structure, and it relies on a single attention map to extract the discriminative local features of a vehicle. We add two CAUs to the truncated ResNet to construct a small but efficient VReID model, ResNetT-CAU. Compared with the original ResNet, the model size of ResNetT-CAU is reduced by 60%. Extensive experiments on the VeRi and VehicleID dataset indicate the proposed ResNetT-CAU achieve the best re-identification results on both datasets. In summary, the experimental results on the challenging benchmark VMMR and VReID datasets indicate our models achieve the best VMMR and VReID performance, and our models have a small model size and fast image processing speed.
Deep Learning Based Solutions for Vehicular Adhoc Networks
This book provides a holistic and comprehensive approach to deep learning for vehicular ad hoc networks (VANETs), covering various aspects such as applications, agency involvement, and potential ethical and legal issues. It begins with discussions on how the transportation system has been converted into Intelligent Transportation System (ITS). The use of VANETs is increasing in the development of ITS to enhance road safety, traffic efficiency, and driver comfort. However, the dynamic nature of vehicular environments and the high mobility of vehicles pose significant challenges to designing and implementing VANETs and ensuring reliable and efficient communication. Deep learning, a subset of machine learning, has the potential to revolutionize vehicular ad hoc networks (VANETs) to enable various applications such as traffic management, collision avoidance, and infotainment. DL has demonstrated great potential in addressing various challenges involved in VANETs by leveraging its ability to learn from vast data and make accurate predictions. It reviews the state-of-the-art DL-based approaches for various applications in VANETs, including routing, congestion control, autonomous driving, and security. In addition, this book provides a comprehensive analysis of these approaches' advantages and limitations and discusses their future research directions. The study in this book shows that DL-based techniques can significantly improve the performance and reliability of VANETs. Still, in-depth research is required to address the challenges of deploying these methods in real-world scenarios. Finally, the book discusses the potential of DL-based VANETs in supporting other emerging technologies, such as autonomous driving and smart cities. It explores the simulation/emulation tools for practical exposure to the vehicular ad hoc network.
Internet of Vehicles and Computer Vision Solutions for Smart City Transformations
This book compiles recent research endeavors at the intersection of computer vision (CV) and deep learning for Internet of Vehicles (IoV) applications, which are pivotal in shaping the landscape of smart cities. These technologies play instrumental roles in enhancing various facets of urban life, encompassing safety, transportation, infrastructure management, and sustainability. The amalgamation of CV and deep learning within smart cities creates a powerful synergy that fosters safer, more efficient, and sustainable urban environments. By harnessing these cutting-edge technologies to drive data-driven decision-making, cities can elevate the quality of life for their inhabitants, mitigate environmental impact, and optimize overall urban functionality. Additionally, this compilation provides in-depth technical and scientific insights into various facets of artificial intelligence (AI) technologies, including forthcoming trends and innovations that are poised to transform smart cities. The book also extends its focus to other areas of smart city development. It explores the application of these technologies in the creation of smart parking solutions, discusses the role of surveillance for public safety, and examines how CV and IoV can be utilized for environmental monitoring. The book also delves into urban planning and infrastructure development, emphasizing the importance of a data-driven approach. It sheds light on the social impact of smart cities and the importance of citizen engagement and discusses issues of security and privacy in the context of smart cities. The book concludes with a look at future trends and challenges in the field of smart cities. Targeted at researchers, practitioners, engineers, and scientists, this book is geared toward those engaged in the development of advanced algorithms for future-forward smart city applications in computer vision, vehicular networking, communication technology, sensor devices, IoT communication, vehicular and on-road safety, data security, and services for IoV-related devices.