An Intelligent Application of Real Time Speed Estimation: An Enhanced Detection and Tracking Approach Towards Vehicle Speed Estimation
Source of Publication
Proceedings of the 2023 7th International Conference on Computing and Data Analysis
A major factor in road traffic fatalities and injuries globally is traffic speed. The need to develop efficient and effective methods that assist in boosting safety and monitoring vehicle speed on the roads is urgent and growing. In this work, we proposed an improved tracker and YOLOv4-based model for estimating vehicle speed. The model has been tuned to operate with monocular camera-based traffic management systems set up on roads with a medium-distance horizon. In order to determine the speed of the vehicles and reduce the usage of processing power and memory, we proposed the Speed and Fast blocks. The results of the experiments demonstrate that the proposed model outperformed the baseline model and yielded impressive results, with an accuracy that was 5.44% higher than the baseline model. The proposed model attained an MSE rate of 1.47 and a speed estimation accuracy rate of 98.53%.
Alsanabani, Ala; Abugabah, Ahed; and Jiao, Licheng, "An Intelligent Application of Real Time Speed Estimation: An Enhanced Detection and Tracking Approach Towards Vehicle Speed Estimation" (2023). All Works. 6273.
Indexed in Scopus