A rock fabric classification method based on the grey level co-occurrence matrix and the Gaussian mixture model

Accurate classification of the rock fabric plays a crucial role in revealing the heterogeneity of the reservoir at different scales. This paper proposes an image-based rock fabric classification method using grey level co-occurrence matrix (GLCM) properties and Gaussian Mixture Model (GMM) as texture descriptors and classifier, respectively. The proposed method is successfully used to classify the images with heterogeneous pore structures and the pictures of outcrops with different sedimentary beddings without preparing the training dataset. According to our results, the classification performance decreases along with the increase of the number of fabric types and the decrease of the structure contrast among different rock types. • GLCM is used to extract the texture features of the rock fabric. • GMM is applied to classify the different rock fabrics. • GLCM-GMM is an effective way for multiphase rock fabric classification.