Object Detection and Tracking Algorithm Study for Football Sports Videos
Chao Chen · 2022 14th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA) · 2022
To further improve the accuracy and robustness of intelligent algorithms in moving target tracking such as football sport, this paper proposes an object detection and tracking strategy based on CamShift and Kalman filtering. First, we adopt CamShift to track players and applies background weighted histogram to improve CamShift algorithm since the tracking defect is easily interrupted by surrounding environment. Then the combination between CamShift and Kalman filter is used to process players’ occlusion and the paper puts forward occlusion factors based on search windows and the method through occlusion factor to provide weights for the results of Kalman and CamShift. The experiment results show that algorithm is effective for football target tracking and resistant to target occlusion.