Image Retrieval Framework for Medical Images using Improved SVM

Abstract: Extraction of certain features from the images is quite difficult to conclude as a result, which can be further used for some kind of specific purpose. In medical field, there is a huge amount of digital images that are related with different diseases. Since, it is a field that demands keen observation of this digital data and responsible work from the concern staff; the chances of error are also very high. This project is an attempt to assistant the medical staff, especially in case of brain tumor to draw a more accurate conclusion on the basis of pattern matching of the captured image with the pre-designed training set. The primary objective of this project is to collaborate the latest available technologies with the knowledge set of the experts to work in a more accurate way. It also aims to serve the man-kind and assist the surgeons in their noble work.

Keywords: Content Based Image Retrieval (CBIR), Query by image content (QBIC), Content-based visual information retrieval (CBVIR), Feature Extraction, Low Level Feature extraction.