Classification of Texture Using Gray Level Co-occurrence Matrix and Self-Organizing Map

Nowadays there has been great increase in use of digital images as a part of information exchange and storage in various fields like medical, science, entertainment, education and research. Because of the huge collection of digital images in different areas there is a need for efficient and accurate classification and retrieval system for image. This paper presents an improved method for image texture classification and retrieval using gray level co-occurrence matrix (GLCM) and Self-organizing maps (SOM). The gray level cooccurrence matrix represents how often different combinations of pixel values or gray levels co-occur in an image. The texture information is extracted from image using gray level co-occurrence matrix and processed. This information is then given to the self organizing map for the classification. The proposed approach is tested on the KTH-TIPS database and the experimental results shows that the proposed method is more accurate, useful and effective in image retrieval.