Research Journal of Information Technology1815-74322151-7959Science International10.3923/rjit.2009.70.78Xiang-WeiL. Zhan-MingL. Ming-XinZ. Ya-lLnZ. Wei-YiW. 2200912Based on rough sets (RS), a novel rapid shot boundary
detection algorithm was proposed in discrete cosine transform (DCT) compressed
domain. First, DCT coefficients and DC coefficients are extracted from
video sequences, so an information system is constructed with DC coefficient.
Second, information system is reduced by ruduction theory of RS, the representation
of the video frame is obtained by reduced DC coefficients. Finally, by
introducing subdividing theory of RS, the frames of video are segmented
objectively. The experimental results show that the algorithm can achieve
higher performance. Compared to conventional algorithm, the algorithm
enjoys many advantages. Firstly, only a subset of frames needs to be considered
during analysis, allowing the reduction of the computational complexity,
so the algorithm can avoid the expensive computations in decompression
processes. Secondly, the relativity of segmentation of video shot becomes
more scientific than earlier methods. Its robustness and effectiveness
are validated by experiments with various kinds of video sequences.]]>Money, A.G. and H. Agius,2007Lei, B. and L.Q. Xu,2006 Lo, C.C. and S.J. Wang, 2001Yi, H., D. Rajan and L.T. Chia,2006Ioannis, T., A. Cavallaro and L.Q. Xu,2008Sze, K.W., K.M. Lam and G. Qiu,2005Kim, K., T.H. Chalidabhongse and D. Harwood,2005Chen, L.H., Y.C. Lai and H.Y. Mark-Liao,2008Lee, M.H., H.W. Yoo and D.S. Jang,2006De Bruyne, S., D.V. Deursen, J.D. Cock, W.D. Neve, P. Lambert and R. van de Walle,2008Bhandarkar, S.M. and S.R. Chandrasekaran,2004Money, A.G. and H. Agius,2007