Analysis of GLCM Parameters for Textures Classification on UMD Database Images

Texture analysis is one of the most important techniques that have been used in image processing for many purposes, including image classification. The texture determines the region of a given gray level image, and reflects its relevant information. Several methods of analysis have been invented and developed to deal with texture in recent years, and each one has its own method of extracting features from the texture. These methods can be divided into two main approaches: statistical methods and processing methods. Gray Level Co-occurrence Matrix (GLCM) is the most popular statistical method used to get features from the texture. In addition to GLCM, a number of equations of Haralick characteristics will be used to calculate values used as discriminate features among different images in this study. There are many parameters of GLCM that should be taken into consideration to increase the discrimination between images belonging to different classes. In this study, we aim to evaluate GLCM parameters. For three decades now, GLCM is popular method used for texture analysis. Neural network which is one of supervised methods will also be used as a classifier. And finally, the database for this study will be images prepared from UMD (University of Maryland database).