Moving vehicle detection based on optical flow estimation of edge
Yanfeng Chen, Qingxiang Wu · 2015
This paper proposed a moving vehicle detection algorithm based on optical flow estimation on an edge image. Using the Canny operator, the image edge is obtained and refined. Then a set of feature points is extracted from the edge image. The pyramid model of Lucas-Kanade optical flow is used to calculate the optical flow information of the feature point set. A new algorithm, which is called the weighted Kmeans optical flow clustering algorithm, is proposed to cluster feature points and used to identify vehicle pattern of the feature point set on the optical flow so that the moving vehicle can be efficiently extracted from the complicated dynamic background. In this paper the vehicle videos, which are captured by a camera on a moving car, are used as the test data set. The experimental results show that this algorithm can effectively detect the moving vehicles in the videos from the camera on the moving car.