CBMIR: Content based medical image retrieval system using texture and intensity for dental images

Abstract

Image retrieval systems attempt to search through a database to find images that are perceptually similar to a query image. This work aims to develop an efficient visual-Content-based technique to search, browse and retrieve relevant images from large-scale of medical image collections Features play a vital role during the image retrieval. The various features that can be extracted are texture, color, intensity, shape, resolution, global and local features etc. In this work, we concentrate on the specific medical domain. The features such as color may not prove to be a very efficient method because the medical domain largely deals with the gray scale images. The features explored in this work are intensity, texture. The first step is to extract the texture feature and the intensity feature from the given input image. Then the both features are combined to form the single feature vector of the image by using the fusion method. The resulting image is compared to the images in the database. The N top most similar images are then retrieved from the database. © 2012 Springer-Verlag.