Multi-Scale Motion Attention Fusion Algorithm for Video Moving Target Detection
Long Liu, Fan Jiang · Journal of Physics Conference Series · 2018
Motion Target detection is a key technology in video analysis, this paper propose a target detection algorithm based on multi-scale motion attention analysis, which provides a new method for motion target detection under global motion scene; Firstly, the noise of motion vector field is removed by filter, and according to the mechanism of visual attention, spatial-temporal motion attention model is built, then the trust degree of motion vector is suggested on the basis of validity analysis of motion vector, and decision fusion of multi-scale motion attention is carried by D-S theory to detect region of motion target; The test result of different video shows the algorithm is effectively for target detection under global motion scene, overcome the defect of traditional algorithm and has better robustness than other algorithms.