Binary feature descriptor for omnidirectional images processing
Abstract
An omnidirectional image has a 360? view around a viewpoint and which could be applied in a variety of fields, such as autonomous navigation, surveillance systems, virtual reality and remote monitoring, is presented. Many techniques of digital image processing rely on local descriptors to characterize the scene information around interest points (or keypoints) in the omnidirectional images. Despite the fact that a lot of advancements have been made in the area of keypoint descriptors over the last years, the literature on omnidirectional image processing for the most part still focuses on float-point descriptors, such as SIFT and SURF, and largely neglects more recent descriptors, such as the BRISK descriptor. In this paper we try to bridge this gap and assess the usefulness of the BRISK descriptor for the task of omnidirectional images processing. This work is a direct comparison of the BRISK and SURF descriptors within a simple verification framework, in order to see the improvement that can be achieved. Our results indicate that the BRISK descriptor is a viable alternative to the SURF descriptor for omnidirectional images processing and due to its binary nature is particularly useful for real-time localization techniques and autonomus navigation which has the visual sensor, including omnidirectional camera.