에지 영상을 이용한 MPEG 비디오의 점진적 장면전환 검출 = Dissolve Scene Change Detection using Edge Image for MPEG

다국어 초록 (Multilingual Abstract)

In this paper, a new algorithm for dissolve scene change detection in MPEG frame sequences is proposed. The proposed algorithm is based on the fact that most of dissolve curves can be characterized by a variance distribution of edge information in the frame sequences. Average edge frame sequences are obtained by performing "sobel" edge detection twice. Features are extracted by comparing a variance with those of local blocks in the average edge frames. Those features are further processed by the opening operation to obtain a smoothing variance curve. The lowest variance in the local frame sequences is chosen as a dissolve detection point. Experimental results show that the proposed algorithm retrieves up to 18 dissolves in total 20 dissolves of some tested videos.