University of Isfahan Computational Intelligence in Electrical Engineering 2821-0689 15 1 2024 03 20 Fast and flexible detection of vehicles in a sequence of images by deep networks Fast and flexible detection of vehicles in a sequence of images by deep networks 57 72 28013 10.22108/isee.2023.136806.1615 FA Nafiseh Zarei Dept. of Electrical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran Payman Moallem Dept. of Electrical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran Mohammad Reza Shams Dept. of Computer Engineering, Shahreza Campus, University of Isfahan, Shahreza, Iran Rasoul Asgarian Dehcordi Power Department, Faculty of Electrical Engineering, Shahrood University of Technology, Shahrood, Iran Journal Article 2023 02 13 Surveillance cameras can provide more accurate traffic information compared to GPS and infrared radiation sensors. With the intelligent processing of image information provided by them, the analysis of traffic problems is done more precisely. Meanwhile, the speed of car positioning is of particular importance because, after positioning the car, other processes are needed to make decisions that require time management. The purpose of the present study is to propose an algorithm based on deep networks that solves this challenge. In the presented algorithm, a fast and accurate detector network is designed based on multiple receptive fields, segmentation, and differential images, which increases the accuracy of the detector by producing semantic feature maps and filtering them. Also, it increases detector speed by reducing the parameters. Each vehicle's maneuver is decided based on the time and location information of that vehicle and the vehicles nearby. Then, according to the type of maneuver, the predictive network is selected in one of the modes of lane keeping, left turn or right turn. The networks designed in the proposed algorithm are complementary to each other. The performance of the proposed algorithm is demonstrated by experiments on the Highway and UA-DETRAC datasets. Surveillance cameras can provide more accurate traffic information compared to GPS and infrared radiation sensors. With the intelligent processing of image information provided by them, the analysis of traffic problems is done more precisely. Meanwhile, the speed of car positioning is of particular importance because, after positioning the car, other processes are needed to make decisions that require time management. The purpose of the present study is to propose an algorithm based on deep networks that solves this challenge. In the presented algorithm, a fast and accurate detector network is designed based on multiple receptive fields, segmentation, and differential images, which increases the accuracy of the detector by producing semantic feature maps and filtering them. Also, it increases detector speed by reducing the parameters. Each vehicle's maneuver is decided based on the time and location information of that vehicle and the vehicles nearby. Then, according to the type of maneuver, the predictive network is selected in one of the modes of lane keeping, left turn or right turn. The networks designed in the proposed algorithm are complementary to each other. The performance of the proposed algorithm is demonstrated by experiments on the Highway and UA-DETRAC datasets. https://isee.ui.ac.ir/article_28013_3cce8868befe17344b850ccdc0331b4a.pdf