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：8
出版社：S.S. Mishra
摘要：A content based image retrieval has became one of the most active research in the past few years. As previously, we were using the text-based approach where it founds very tedious and difficult task for solving the purpose of image retrieval. But the CBIR is the method where there are many methodologies are available and the task of image retrieval becomes effectively easier. In this, there are some effective methods for CBIR are discussed and the comparative study is made. However most of the proposed approaches emphasize on finding the best representation for different image features. There are the techniques called a'trous wavelet transform (AWT) and Julesz's texton elements are used to generate the texton image. Also the multi texton histogram (MTH) is one of the method for this task. It integrates the advantages of co-occurrence matrix and histogram by representing the attributes of co-occurrence matrix using histogram. Here, the user oriented mechanism for CBIR method based on an interactive genetic algorithm (IGA) is proposed. The color attributes like the mean value, standard deviation and image bitmap of a color image are used as a features for retrieval
关键词：CBIR; texton; texton detection; multi texton; IGA; population; crossover.
