Video scene mutation change detection combined with SIFT algorithm

Video scene change detection has a very important role for video annotation and semantic search. This paper proposes a scene mutation change detection algorithm combined with SIFT( Scale Invariant Feature Transformation) feature point extraction. Firstly, the feature points of two adjacent video frames are extracted respectively using SIFT algorithm and the number of them is counted respectively. Then image matching of the two adjacent frames of the video is performed and the number of matching feature points is counted. Finally, the ratio between the number of matching feature points of the current frame and the number of matching feature points of its previous frame is calculated, so as to judge the scene change by this ratio. The average scene mutation change detection rate in the experimental results can reach 95.79%. The proposed algorithm can judge scene change during image matching. Therefore, the algorithm can not only be applied widely, but also guarantee the accuracy of scene change detection. Experimental results show the effectiveness of the proposed algorithm.