Edge-Based Vehicle Tracking and Speed Estimation in Intelligent Transportation

Edge-Based Vehicle Tracking and Speed Estimation in Intelligent Transportation

Sumithra S1, Dr. Ramya R S2, Dr. Deepak G3
1(PG Scholar, Computer Science Department, Dayananda Sagar College of Engineering, Bengaluru) 2(Associate Professor, Computer Science Department, Dayananda Sagar College of Engineering, Bengaluru) 3(Assistant Professor, Computer Science Department, Dayananda Sagar College of Engineering, Bengaluru)

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - An intelligent transportation system plays an important role in traffic management we have an innovative approach for real-time vehicle tracking and speed estimation in an Intelligent Transportation System (ITS) using YOLOv5 and Deep SORT techniques. The proposed solution is designed to be implemented on an edge device, specifically the Jetson AGX Orin. The combination of YOLOv5, a state-of-the-art object detection algorithm, and Deep SORT, a powerful tracking algorithm, enables efficient and accurate vehicle tracking Subsequently, the Deep SORT algorithm is applied to track the identified vehicles over consecutive frames, allowing for consistent and reliable tracking results. By conducting all computations at the edge using the Jetson AGX Orin, real-time processing is achieved, minimizing latency and enhancing the system's responsiveness. The use of edge computing not only reduces dependence on cloud resources but also ensures data privacy and security, making it an ideal choice for ITS applications

Key Words: traffic management, object detection, vehicle tracking, frames, speed detection, latency