Corner Detection via Two-Layer Threshold Method
This paper presents a two-layer threshold method for the corner detection. The method is inspired by the classical Susan corner detection model, and the improvement is two-fold. One is the choice of the self-adaptive threshold, which can be used to detect the possible corner areas. However, since a corner has an arbitrary orientation and sometimes is disturbed by noises, we provide a series of rotate coordinate systems, the second improvement, to enhance the accuracy of our method. Experiment results show that the improved algorithm is robust and is of high efficiency.
KeywordsCorner detection Self-adaptive threshold Susan model
This work was supported by National Natural Science Foundation No. 61170327, Organization Department Program of Beijing for the Talents No.2010D005002000008, Beijing Natural Science Foundation No.1082007, Scientific Research Common Program of Beijing No. KM200410009010, and PHR(IHLB) No.PHR201008199.
- 6.Harris C, Steven MA (1988) Combined corner and edge detector. In: Proceeding of 4th Alvey vision conference, Manchester, pp 189–192Google Scholar