Enhanced Video Retrieval and Classification of Video Database Using Multiple Frames Based on Texture Information

Abstract — Content Based Video Retrieval (CBVR) has been increasingly used to describe the process of retrieving desired videos from a large collection on the basis of features that are extracted from the videos. The extracted features are used to index, classify and retrieve desired and relevant videos while filtering out undesired ones. Videos can be represented by their audio, texts, faces and objects in their frames. An individual video possesses unique motion features, color histograms, motion histograms, text features, audio features, features extracted from faces and objects existing in its frames. Videos containing useful information and occupying significant space in the databases are under-utilized unless there exist CBVR systems capable of retrieving desired videos by sharply selecting relevant while filtering out undesired videos. Results have shown performance improvement when features suitable to particular types of videos are utilized wisely. Various combinations of these features can also be used to achieve desired performance. Many researchers have an opinion that result is poor when images are used as a query for video retrieval. Here, instead of using a single image or key frames, multiple frames of the video clip being searched are used. Technique used for CBVR system shown in this paper yields significantly acceptable and higher retrieval results. The system is implemented using MATLAB. Performance of the system is assessed using a database containing 1000 video clips of 20 different categories with each category having 50 clips. The performance is shown using two different types of texture features- features extracted using Gabor filter and Kekre’s Fast Codeboo