PERFORMANCE EVALUATION OF CORNER DETECTORS: A SURVEY

An image may contains number of important features, such as closed-boundary regions, edges, contours, line intersections, Corners, etc. Among them corner is one of the important feature, which can be identified by the change of intensity gradient in at least two-directions. Corner detectors have many applications in computer vision, object tracking or recognition. The corner detectors are classified into three clauses: contour based, intensity based and parametric based model. The performances of corner detector are presented in terms of consistency, accuracy, matching score, information rate, ground truth, visual inspection. The objective of this paper is to provide a comprehensive study of corner detection methods and their evaluation.