期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2016
卷号：7
期号：4
页码：164-167
语种：English
出版社：Ayushmaan Technologies
摘要：A Content-base Video retrieval, which provides convenient ways to retrieve Video from large Video database. While many previous Video retrieval techniques do not look at regions in an Video, region based Video retrieval techniques have been gaining attention recently. Traditional method for retrieval has failed to retrieve multimedia data like Video, audio, video etc. CBIR is very active research area which is used retrieves Videos based on Video visual features. The present method uses a new technique based on wavelet transformations to create feature vector of Video, characterizing texture feature of the Video is constructed. For quick retrieval statically detail of Video like mean of approximation, standard deviation and second moment is utilized for shape in Video. Our method derives feature vector for each Video characterizing the texture feature of sub Video from only three iterations of wavelet transforms. The K Means Clustering Algorithm is used to cluster the group of Videos based on feature vector of Videos by considering the minimum Euclidean distance. LIRe (Lucene Image Retrieval) is a light weight open source Java library for content based image retrieval. It provides common and state of the art global image features and offers means for indexing and retrieval. Due to the fact that it is based on a light weight embedded text search engine, it can be integrated easily in applications without relying on a database server.
