CVRRSS‐CHD: Computer vision‐related roadside surveillance system using compound hierarchical‐deep models. Issue 11 (15th September 2020)

• Record Type: Journal Article
• Authors: Mao, Jia; Hong, Dou; Wang, Xi; Hsu, Ching‐Hsien; Shanthini, A.
• Is Part Of: IET intelligent transport systems. Volume 14:Issue 11(2020)
• Page Start: 1353
• Page End: 1362
• Publication Date: 2020-09-15
• Subjects: Internet of Things -- decision making -- road traffic -- Big Data -- object detection -- cloud computing -- traffic engineering computing -- computer vision -- surveillance -- data analysis -- image matching -- smart cities -- intelligent transportation systems -- image denoising -- fuzzy set theory; CVRRSS‐CHD -- compound hierarchical‐deep models -- Cloud Computing -- smart city measures -- smart city initiatives -- Big Data Analytics -- busy roadside activities -- intelligent transportation system -- city side region -- traffic misbehaviour problems -- effective computer vision‐related roadside surveillance system -- road traffic images -- IoT device -- noise reduction techniques -- geometric pattern matching algorithm -- gap‐related decision -- fuzzy‐based decision‐making system -- Internet of Things -- unwanted traffic reduction -- vehicle object -- uncertainty problems -- object detected data; Intelligent transportation systems -- Periodicals; Electronics in transportation -- Periodicals; 388.31205
• DOI: 10.1049/iet-its.2019.0834
• Languages: English
• ISSNs: 1751-956X

Abstract

Abstract : Recent years, Big Data, Cloud Computing and the advancement of the Internet of Things (IoT) played a major role in making smart city measures feasible. During this smart city, development, busy roadside activities and appropriate parking are considered as one of the major issues in the intelligent transportation system. Especially, in the city side region, the roadside activities are creating traffic misbehaviour problems which lead to various surveillance issues. So, in this study, the focus on the effective computer vision‐related roadside surveillance system is created to reduce the unwanted traffic and misbehaviour issues. Initially, road traffic images are collected with the help of the IoT device, which is processed by noise reduction techniques to eliminate the noise. After that, the vehicle object is identified in terms of geometric pattern matching algorithm as named as compound hierarchical‐deep models. Here, the geometric matching process is used to solve the uncertainty problems during the prediction of the vehicle in roadside activities. From the object detected data, roadside activities, such as vehicle position, occupancy, gap‐related decision, have been handled with the help of a fuzzy‐based decision‐making system. Furthermore, the efficiency of the system has been evaluated using respective case studies and experimental analysis.