WSEAS Transactions on Signal Processing
Print ISSN: 1790-5052, E-ISSN: 2224-3488
Volume 11, 2015
Video Cut Detection Method Based on a 2D Luminance Histogram Using an Appropriate Threshold and a Post Processing
Authors: , , ,
Abstract: Segmenting a video is a fundamental component in video content based and automatic video analysis. Video cut detection is considered as a mainstay for video retrieval, especially for the keyframe extraction and the video summarization. In this paper, we present an algorithm for partitionning a video into shots. This algorithm abide to the two following steps. First, we calculate the difference in the luminance histogram between successive frames, which we compare against a predefined threshold to detect the cuts. A statistical study of the dissimilarity measures allows us to calculate that threshold in such a way that there are no missed shot. Second, we use a post processing to eliminate the false detections. Experimental results are presented to demonstrate the good performance of the proposed detector with high values of the evalution criteria.
Search Articles
Pages: 99-106
WSEAS Transactions on Signal Processing, ISSN / E-ISSN: 1790-5052 / 2224-3488, Volume 11, 2015, Art. #12