ICCEA '10 Proceedings of the 2010 Second International Conference on Computer Engineering and Applications - Volume 01
IEEE Transactions on Image Processing
| Hi-index | 0.00 |
The availability of huge multimedia databases and the development of information highways have urged many researchers for developing effective methods of retrieval based on their content. The traditional way of searching the available huge collections of multimedia data was by keyword indexing or simply by browsing, where by the user's main interest lies in the maximum retrieval of similar data. Digital image databases however, opened the way to content-based searching and retrieval. A lot of research has been done in retrieving the content based on image features like color, texture, and shape. In this paper an attempt is made to design a methodology for an efficient image retrieval system by extracting low level and high level features from images through relevance feedback. In order to reduce the computational complexity and to achieve efficiency, a two phase approach is adapted. In the first phase color segmentation and GLCM of second order statistics for texture are performed. The second phase takes the feedback obtained from phase1 and involves the usage of wavelets combined with PCA for a refined search and subsequent retrieval of similar images.