Improved global optical flow estimation with mean shift algorithm for target detection
Zhonghua Wang, Chunyong Li, Qingping Liu · International Journal of Wireless and Mobile Computing · 2015
In view of the problem that the global optical flow algorithm cannot acquire accurate motion parameter estimation at a low-gradient value, an improved method has been presented in order to enhance the self-adaptive ability. The proposed method is divided into three steps: firstly, the Gaussian filter is used to restrain the background noise; secondly, the optical basic constraint weighted function is modified to acquire targets area; finally, the morphological filtering and mean shift segmentation algorithm is adopted to reduce the false alarm rate. Compared with other algorithms, the experiment results show that the proposed method can improve the self-adaptive ability and the detection rate.