Phase Congruency Detects Corners and Edges

There are many applications such as stereo matching, motion tracking and image registration that require so called ‘corners’ to be detected across image sequences in a reliable manner. The Harris corner detector is widely used for this purpose. However, the response from the Harris operator, and other corner operators, varies considerably with image contrast. This makes the setting of thresholds that are appropriate for extended image sequences difficult, if not impossible. This paper describes a new corner and edge detector developed from the phase congruency model of feature detection. The new operator uses the principal moments of the phase congruency information to determine corner and edge information. The resulting corner and edge operator is highly localized and has responses that are invariant to image contrast. This results in reliable feature detection under varying illumination conditions with fixed thresholds. An additional feature of the operator is that the corner map is a strict subset of the edge map. This facilitates the cooperative use of corner and edge information.