期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2016
卷号：5
期号：5
页码：1427-1432
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Texture classification is one of the traditional challenges in the field of pattern recognition. In this paper, we propose a technique to classify textures using two very powerful tools: Modular PCA (m-PCA) and Local Binary Patterns (LBP). Being an advanced version of conventional PCA, m-PCA focuses more on various portions of a texture, which results into enhanced recognition and improved robustness. On the other hand, LBP is a well-known texture operator, which has been used for object segmentation. The proposed algorithm combines these two feature extraction approaches into one. We get highly competitive results which prove the effectiveness of the proposed methodology.
