An Efficient Multi-Object Tracking and Counting Framework Using Video Streaming in Urban Vehicular Environments

摘要

Object counting is an active research area that gained more attention in the last few years. Since deep learning methods outperformed all other object detection algorithms, the design of efficient object counting algorithms became more realistic and achievable. Numerous algorithms targeting various challenges associated with object counting have been introduced. In a smart transportation system, vehicle counting plays a crucial role as it helps in creating autonomous systems, and better planning for roads. In this paper, we present an efficient object counting system and assess its performance using a dataset of 20 different videos. The proposed system leverage an efficient object detector, and object tracker to perform the counting. This paper combines different approaches to count objects by tracking them, but performs the tracking operation efficiently. Therefore, the proposed systems achieve high accuracy values with low processing time.