Shot Boundary Detection Based on Histogram of Mismatching-Pixel Count of FMB

A universal cut-detection algorithm is proposed based on the histogram of mismatching-pixel count of forward matching block. The proposed algorithm uses motion compensated pixel difference to improve the level of tolerance to camera and video object motion and uses histogram of mismatching-pixel count to decrease the level of sensitiveness to noise. The experimental results confirmed that the proposed algorithm was very effective and robust, and it could significantly improve the cut-detection precision compared with conventional pixel-based statistical algorithm and grey histogram algorithm