An Enhanced Local Binary Pattern for Texture Classification
Huawei Tao, Rugang Wang, Li Zhao · 2016
In order to improve the recognition rate of texture classification, an enhanced Local binary pattern, called completed local binary pattern based on gray level and structural information (CLBP_GLSI), is proposed in this paper.We firstly proposed an structural texture operator called gray level and structural information (GLSI), which adopts the average gray level of image to make image converted into binary images, and binary images are encoded as border or interior pixels image by Border/Interior Pixel Classification (BIC).Secondly, by combing with CLBP_M, CLBP_S and GLSI in into joint or hybrid distributions, the CLBP_GLSI are obtained.Experimental results obtained from two databases show that CLBP_GLSI achieves better results than other texture features.