Recent developments in the use of the co-occurrence matrix for texture recognition

Abstract

We present an overview of several developments in the use of the gray level co-occurrence matrix method for image texture analysis and classification. We first introduce the theory and notation of gray level co-occurrence matrices (GLCM), then we outline several developments including the use of a multi-scale GLCM matrix, the use of an associated discrimination matrix, and finally several adaptive methods for the creation of highly discriminatory features for classification problems.