An improved approach of scene change detection in archived films

Scene change detection is an essential step for archived films restoration. But the image quality of archived films often degrades gradually which makes false detection during shot segmentation. This paper focuses on some problems about the scene change detection in archived films. For abrupt change shot detection, the difference between histograms of block-bases adjacent frames and across frames based on the brightness information is used, the similar histogram problem is treated and the error caused by flicker is eliminated. Then the gradual change of shots are detected based on the fact that the variances of inter-frame are usually monotonically increasing in fade-in sequence and monotonically decreasing in fade-out sequence,. Combining with the twin-comparison approach makes the algorithm more robust. Experimental results show that these methods have better performance in shot segmentation for archived films.