Content based video retrieval using entropy, edge detection, black and white color features

Traditional video retrieval methods fail to meet technical challenges due to large and rapid growth of multimedia data, demanding effective retrieval systems. In this paper we propose integration of entropy and black and white points on edge, features of video key frames for developing proposed video retrieval systems. First, video feature database is created using entropy feature extracted from key video frames of video database. Same feature is extracted from video frame query. We then extract the edge and black and white points on edge from database frames and query frame. Finally similarity measure is applied to retrieve the best matching frames and corresponding videos are presented as output. The experimental results show that feature integration is effective.