期刊名称：International Journal of Advanced Research In Computer Science and Software Engineering
印刷版ISSN：2277-6451
电子版ISSN：2277-128X
出版年度：2013
卷号：3
期号：2
出版社：S.S. Mishra
摘要：The digital image is used to identify and recognize a person, object, location a nd various things. To find specific digital images from large database of images has become an area of wide interest nowadays. CBIR (Content based image retrieval) means that searching, browsing and retrieval the image using the actual contents of the image like visual features of an image such as color, shape, texture and spatial layout rather than the metadata such as keywords, tags and descriptions associated with the image. To improve existing CBIR performance, it is very important to find effective and efficient image decomposition, feature extraction and image matching mechanisms. This research aims to improve the performance of CBIR using wavelet decomposition by haar wavelet. After that features are extracted using f-norm theory. K-mean clustering is used to form the cluster of images and similarity matching is done using f-norm theory. The use of progressive retrieval strategy is to provide balance between computational complexity and retrieval accuracy. In this paper we compared the retrieval perfo rmance of proposed content based image ret rieval system with the exiting technique of wavelet histogram. The proposed research produces better results than wavelet histogram
关键词：f-norm; haar wavelet; image decomposition; progressive retrieval strategy; feature extraction
