Feature Point Detection by Combining Advantages of Intensity-based Approach and Edge-based Approach

In this paper, a novel corner detection method is presented to stably extract geometrically important corners. Intensity-based corner detectors such as the Harris co ner can detect corners in noisy environments but has inaccurate co rner position and misses the corners of obtuse angles. Edge-based cor ner detectors such as Curvature Scale Space can detect structural corn ers but show unstable corner detection due to incomplete edge de t ction in noisy environments. The proposed image-based direct curva ture estimation can overcome limitations in both inaccurate structu ral corner detection of the Harris corner detector (intensity-based) and the unstable corner detection of Curvature Scale Space caused by incomp lete edge detection. Various experimental results validate th e robustness of the proposed method. Keywords—Feature, intensity, contour, hybrid.