Layered filtering technique for content based video retrieval

Content-based video searching and retrieval is an upcoming and challenging issue. Efficient techniques for the analysis and retrieval of videos have become very important with the proliferation of digital video content. This paper outlines a technique to retrieve the matching videos from the video database, given a query clip. The key frames which are extracted at fixed time intervals from the videos are hashed as feature vectors. Layered filtering technique has been implemented on the features - corners, Adjacent Pixel Intensity Difference, color distribution and edges in the aforementioned order. Histogram based comparison is employed for each feature in order to retrieve the similar videos. The performance analysis on the obtained experimental results demonstrates that the proposed method is effective and efficient.