Rough Sets Based Video Mining Preprocessing Algorithm in Compressed Domain

A critical and fundamental task in video mining is data preprocessing, in this paper, aimed to overcome limitations of redundant data for video mining, the paper propose a video mining preprocessing algorithm based on Rough Sets. Firstly, the representative data of video sequences is extracted in compressed domain. Secondly, the Information System Table is constructed by extracted representative data. Finally, the Core of Information System Table is achieved by making use of the attributes reduction theory of RS. As our experimental results indicate, the algorithm can get effective and scientific data to complete video mining such as key frame extraction and shot segmentation and other operations. Compared to existing techniques, our proposed algorithm enjoys following advantages. (1) only a subset of frames need to be considered during video mining. (2) The limitations of requirements for a huge amount of memory and CPU resource are overcome.