计算机技术与发展2017，Vol.27Issue(12)：85-88,4.DOI:10.3969/j.issn.1673-629X.2017.12.019
不同对比度纹理结构特征的多粒度分解及描述
Multi-granularity Decomposition and Description of Texture Features with Different Contrast
摘要
Abstract
Texture is a kind of common visual phenomena,which plays a very important role in image processing,3D reconstruction and pattern recognition. Due to the multi-granularity characteristic of texture,structural primitives based on single size are not enough to de-scribe texture features accurately. However,based on different size structural elements for iterative morphological operations in the mathe-matical morphology,the structure information with different granularity can be obtained. On this basis,a multi-granularity decomposition of the texture structure based on morphological opening operation is put forward,and a multi-granularity description method of texture features with different overall contrasts and the same structure is proposed. In order to verify its effectiveness,with 20 texture images from Brodatz database and MIT VisTex database as experimental data, the proposed method is made a comparison with texture description method based on gray level co-occurrence matrix and based on Gabor filtering. The experiment shows that it is better than the other two methods.关键词
纹理结构特征/数学形态学/开运算/多粒度分解/多粒度描述Key words
texture structure feature/mathematical morphology/opening operation/multi-granularity decomposition/multi-granularity description分类
信息技术与安全科学引用本文复制引用
冯霞,周勃,崔卫红..不同对比度纹理结构特征的多粒度分解及描述[J].计算机技术与发展,2017,27(12):85-88,4.基金项目
国家"973"重点基础研究发展计划项目(2012CB719903) （2012CB719903）
