Motion target detection algorithm based on monocular vision
Yuxi Feng, Qizhou Wu, He Guotian · 2017
In this paper, presents an improved motion obstacle detection algorithm based on monocular vision by contrast analysis of optical flow method, frame difference method and background difference method. Firstly, the image is collected by the monocular camera, preprocessed, and then processed by an improved motion obstacle detection algorithm. Finally, the image is processed to solve the void problem and get the complete motion obstacle information. The experimental results show that the method proposed in this paper requires less computational complexity and simpler parameter design than the traditional detection algorithm, and the motion detection algorithm proposed in this paper can satisfy the real-time and real-time Accuracy requirements.