Seamless Fundus image stitching using WLD to improve field of view

Digital Fundus images are used for the manifestation of many ocular and systemic diseases but the main problem with fundus cameras is that their captured images have a field of view limited to a maximum of 45 degrees. In order to increase the field view image stitching of the fundus images is required. This paper presents a new technique for stitching multiple fundus images of a single patient.

The proposed method uses weber local descriptor for features extraction and presents a novel technique to achieve seamless blending. The paper also compares the proposed algorithm with the other state-of the art techniques like SURF, SIFT and ASIFT. A locally gathered dataset of 15 patients having 56 images with a minimum of 3 images per patient is used to evaluate the proposed system. The results show the significance of proposed system.