Based on Image Matching and Tracking of Moving Objects
Jian Xiong Wang, Yu Lan Wang, Bin Hong, Hong Ling Zhang · Advanced materials research · 2012
According to the traditional gray-scale value matching algorithm and its defect of high time complexity, this paper presents a matching algorithm based background subtraction. First we find moving target in the background by subtraction, and then match the template image to the target image of motion, which greatly reduces the amount of calculation and improves the matching speed. We effectively predict the target’s location in the next frame by five-point linear prediction method. Simulation shows that the algorithm is better than the traditional gray-scale matching algorithm and it has strong robustness.