SCALE INVARIANT FEATURES EVALUATION USING IMAGE-BASED MOMENTS

In this paper two new criteria are proposed for performance evaluation of point features extracted using SIFT detector. The criteria measure the quality of image features for visual servoing applications. The first one is derived from the repeatability rate criterion and the second one is a new approach to analyse the spreading degree of image features using Mahalanobis distance. New feature descriptors are introduced based on image moments and results of real experiments are analysed.