Image texture classification and retrieval using self-organizing map

摘要

Nowadays there has been great interest in field of image texture classification and retrieval. The increasing use of digital images has increased the size of image database which resulted in the need to develop a system that will classify and retrieve the required image of interest efficiently and accurately. This paper presents an effective and accurate method to classify and retrieve image using Self-organizing maps (SOM). The proposed method employs two phases, in the first phase color histogram is used to extract the color features and then the extracted features are given to Self-organizing map for initial classification. In the second phase Gray level co-occurrence matrix (GLCM) is used to extract the texture information from all images in each class from initial classification and then again given to Self-organizing map for final classification. The experimental results show the efficiency of the proposed method.