CH-SIFT: A local kernel color histogram SIFT based descriptor

This paper presents a novel method for extracting distinctive invariant features from color images that can be used to perform reliable matching between different views of an object or scene. The features are invariant to image scale and rotation, and also to illumination changes because of using a color kernel histogram descriptor incorporating with a classic SIFT descriptor. The method is compared with SIFT, SURF and CSIFT and is shown that the accuracy of proposed method is better than others by testing all of them on the ALOI database.