This research paper aims at devising efficient ways of performing vehicle detection using Deep learning techniques. Vehicle detection is most commonly used for traffic surveillance, at toll plazas, etc. In this paper, the efficient execution of detecting vehicles will be discussed using various YOLO architectures. This research provides us with detailed information about the models that can efficiently work on the Jetson nano kit while providing maximum accuracy. It also provides the researchers and developers with practical and comprehensive deep-learningbased solutions for detecting vehicles providing information about the process from development to deployment. In this paper, we have considered multiple YOLO algorithms such as v3, v3 tiny, v4, and v4 tiny. Thus, providing an analysis of all algorithms and their efficiencies. After comparing the results we have analyzed that YOLO v4 provides the best efficiency in terms of accuracy which is 95.41%.
Development of Real Time AI-Based Vehicle Detection System for Traffic Surveillance
17.05.2023
3067812 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
High‐accuracy vehicle lamp detection for real‐time night‐time traffic surveillance
Wiley | 2020
|High-accuracy vehicle lamp detection for real-time night-time traffic surveillance
IET | 2021
|A real-time computer vision system for vehicle tracking and traffic surveillance
Online Contents | 1998
|