Region based texture descriptor for content based medical image retrieval using second order moments

The ever-increasing popularity of the use of larger-volume image database in various applications, it is an imperative to build an efficient retrieval system to browse through the entire database. Our approach relies on image feature that exploit texture features using second order statistical values such as gray-level co-occurrence matrix (GLCM), this feature extraction process is as follows: the image is divided into equal sized blocks and the average intensity is computed on the pixels in each block. These values are stored for image matching and similarity measure are based on Euclidean distance, City block of absolute value metric and Murkowski distance. Through the image retrieval experiment, We tested different images database images and measured Recall rate and Error rate as a performance measure which indicate that the use of proposed Texture features is an efficient retrieval technique which has obvious advantage and gives higher recall rate as compared to the histogram technique.