Vehicle forward Collision Warning and Speed Estimation System Using Dashboard Camera

Tan, Chee Wang (2021) Vehicle forward Collision Warning and Speed Estimation System Using Dashboard Camera. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text TAN CHEE WANG_Full text.pdf Restricted to Registered users only Download (26MB)

Abstract

The increase of accidents leads to the improvement of the safety features of vehicles for preventing accidents. However, the safety features applied are still limited because most of the safety features are related to distance of two vehicles, and speed of the vehicle in front. For the present case, there are many developments in distance estimation, but only few developments in speed estimation, especially there is not speed estimation applied in a single board computer. Hence, the main purpose of this project is to propose a method for distance and speed estimation which can be applied in Forward Collision Warning System. Besides, the system is necessarily able to implement into the single board computer platform while able to satisfy the real time operation. The key concept of the method is to estimate the distance and speed based on the width size of the vehicle. The proposed methodology is able to achieve 82.37% of accuracy on distance and speed estimation, and 80.18% of accuracy on alerting the driver. The experiment results show that the proposed method is able to alert the driver to realize the Forward CollisionWarning System at a relative speed difference within 30km/h by using the distance and speed estimation.