Vehicle Speed Estimation in Accident Prone Areas using Image Processing

Abstract: In this paper, I present a method to determine the vehicle speed in accident prone areas using the video frames captured from the camera fixed on the road. One frame is taken as the reference frame and then two different frames are taken to calculate the absolute difference between those two frames with the reference frame. The absolute difference gives us the motion in the frame. Then we perform the thresholding and morphological operations to calculate the vehicle mask. The mask is used as a basis to calculate the centroid in the two images. The difference between the centroid of the two images gives us the displacement of the vehicle with respect to the two frames through which we calculate the velocity. Experimental results show that the proposed method exhibits good and consistent performance.

Keywords: Frame Differencing, Thresholding Operation, Morphological Operation, Centroid, Radar.