An efficient interest point detection technique for real-time applications

Feature selection is a key step for image registration process. The success of feature selection has a fundamental effect on image matching. Corner points are considered as interest points as they are formed from two or more edges and edges usually define the boundary between two different objects or parts of the same objects. Corners determine the contours characteristics of the target image and the number of corners is far smaller than the number of image pixels, thus corner can be a good feature for image registration. Detection of corners for image registration is difficult task for an airborne real-time platform. Airborne platform suffers with many constraints like processing power, qualified hardware, real time constraints etc. In this paper, we propose an efficient interest point detection technique to meet airborne real-time constraints. It requires sorting in local neighborhood in spatial domain. Performance is evaluated quantitatively using SNR criteria in neighborhood and random location. The results are further validated against image translation and rotation.