A Novel Feature Weighted Clustering Algorithm Based on Rough Sets for Shot Boundary Detection

Shot boundary detection as the crucial step attracts much more research interests in recent years. To partition news video into shots, many metrics were constructed to measure the similarity among video frames based on all the available video features. However, too many features will reduce the efficiency of the shot boundary detection. Therefore, it is necessary to perform feature reduction before shot boundary detection. For this purpose, the classification method based on clustering algorithm of Variable Precision Rough-Fuzzy Sets and Variable Precision Rough Sets for feature reduction and feature weighting is proposed. According to the particularity of news scenes, shot transition can be divided into three types: cut transition, gradual transition and no transition. The efficiency of the proposed method is extensively tested on UCI data sets and more than 3 h of news programs and 96.2% recall with 96.3% precision have been achieved.