Video Object Tracking Method Based on Normalized Cross-correlation Matching

Abstract

Combing with specific temporal information of video, this paper proposes a kind of video object tracking method based on normalized cross-correlation matching by using the high precision characteristics of normalized cross-correlation image matching. Firstly, extract video background from the temporal information of video. Then, acquire the region of moving object using background subtraction. Lastly, carry out related matching and updating towards the extracted moving object by means of normalized cross-correlation. Experimental result shows that the adaptability of our method is strong, which can well solve the tracking problems when tracking objects have scale transform. It also has good anti-interference ability and robustness, and can track moving objects accurately under the condition of noise interference, lens dithering and background mutation.