期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2011
卷号：11
期号：9
页码：65-72
出版社：International Journal of Computer Science and Network Security
摘要：The theme of the work presented in the paper is novel texture pattern based image retrieval techniques using ternary image maps and two different non-sinusoidal orthogonal Walsh wavelet transforms namely Global Walsh Wavelet Transform and Local Walsh Wavelet Transform. Different texture patterns namely ��64-pattern�� and ��256-pattern�� are generated using both the Walsh wavelet transform matrices giving rise to global and local texture patterns. The generated texture patterns are then compared with the ternary image maps to generate the feature vector based on structural matching as the matching number of ones, zeros, minus ones per Walsh wavelet texture pattern. Here total 4 variations of the proposed novel image retrieval methods using texture patterns are considered with two different texture patterns and two different ways to generate these texture patterns (Global and Local). The proposed texture content based image retrieval (CBIR) techniques are tested on the image database with help of 55 queries (randomly selected 5 from each of 11 image categories) fired on image database. The performance comparison of texture pattern based CBIR methods is done with help of precision-recall crossover points.
关键词：CBIR; Walsh Transform; Wavelet Transform; Texture; Patterns; Ternary Image Maps; Global & Local Textures
