A novel shot detection algorithm based on graph theory

Shot boundary detection has attracted much more research interesting in recent years, but many of the algorithms proposed in the literatures are not robust to flash lamp and motion of camera and object. This paper present a novel shot boundary detection algorithm based on graph theory. It can detect cut and gradual shot and distinguish them, and it is not intensive to all of those interference factors. At first the feature of color is extracted, the dissimilarity of video frames is defined. Then the video frames are divided into several different groups through performing graph-theoretical algorithm. According to cut and gradual changes have the different characters on the two successive frames belong to different groups, it detect cut and gradual shot. The efficiency of the proposed algorithm is extensively tested on movie, sports and other videos. The experiments results indicate the method had a high accurate rate in both cut shot detection and gradual shot detection.