Texture based matching of medical images

Abstract

Texture is one of the crucial primitives in human vision Containing important information about the structural arrangement of surfaces and their relationship to the surrounding environment In the past many years texture based methods are developed for extracting features & matching of medical images. Medical images have become the recent key investigation tools for medical diagnosis and treatment planning. Due to the advent of digital imaging the need of data storage and retrieval of medical images increased rapidly. Although medical images contain useful information, their use is limited to access and availability. Therefore, storage solutions to facilitate cataloguing, browsing, and retrieval of medical images are required. In this paper a system for effective texture based matching of and medical images is presented. Firstly various levels of texture features of database image are extracted and saved. Features of query image are then compared with corresponding features of database images and the difference between the two is then used to compute cost function which implies maximum matching category for query image.