Geometry-based Completed Local Binary Pattern for Texture Image Classification
Wei Wang, Qiqi Kou, Shu Zhou, Ke Luo, Lifeng Zhang · 2020
In view of the fact that the accuracy of texture image classification is easily affected by changes in illumination and rotation, based on the analysis of geometric curvatures information of the image microscopic geometric surface and the completed local binary pattern (CLBP), this paper proposed a new descriptor, named as Geometry-based Completed Local Binary Pattern (GCLBP). Inspired by the continuous rotation invariance and illumination robustness of the geometric curvature information, principal curvatures (PCs) of all pixels are first calculated and then used to represent the gradient magnitude information of the image, which are further exploited to replace the original gradient magnitude information in CLBP. To further improve the accuracy of texture classification, a cross-scale joint coding strategy is exploited to form the final GCLBP. The experimental results on two standard texture databases demonstrate that the GCLBP algorithm proposed in this paper is not only far superior to the original CLBP in terms of classification recognition accuracy and dimensionality of feature vector, but also better than most existing advanced texture classification methods.