Illumination Invariant Medical Image Retrieval Using Relative Vector

In this paper we design a medical image retrieval system that conations variety of type images for clinical student to learning or patient to understand his health condition. The image contains variety of type image, thus we consider global and local image features expect to describe variety of type images. We proposed a relative vector method for medical image content retrieval. The similarity metric based on the relative vector representation will immune in defective illumination. Based on the relative vector, we also can extract local and global feature. The experimental results show that the proposed approach can effectively remove the irrelevant results and improve the precision. Key-words: Medical image retrieval; Content based; Illumination Invariance;