engleski

An algorithm for Vehicle Distance and Speed Estimation

In this paper, an algorithm for vehicle distance and speed estimation is proposed. In particular, the estimation is performed using video frames obtained from the cameras, located on the front and rear of the car. Moreover, the vehicle detection is performed using a Viola-Jones algorithm, while a Median flow is used for detected vehicle tracking. The speed of the vehicle is calculated using a distance of the detected vehicle and a change in the detected bounding box width between two consecutive frames. The proposed algorithm performance is tested on different car brands, different weather conditions, and different car distances. The testing results of the proposed method show promising results.

Rain, Vehicle detection, Neural networks, Estimation, Lighting, Cameras, Real-time systems

nije evidentirano