Edge-relevant Structure Feature Detection Using Caputo-Fabrizio Fractional-order Gaussian Derivatives

Edge-relevant structure features (ERSFs), e.g., object edges, boundaries and contours, junctions, etc. play an important role in the low and middle level image processing task, such as image segmentation, as well as in higher-level computer vision tasks, such as scene analysis and vision understanding. Commonly-used ERSF detection methods employ the integer-order differentiation-based methods, which are noise-sensitive and have less selectivity of edge feature. Hence, they are difficult to effectively extract object boundaries especially in natural images with rich fractal-like structures. We presented a highly selective and noise-robust ERSF detection approach based on the fractional-order Gaussian derivatives (FoGDs) by using the Caputo-Fabrizio fractional definition. The proposed ERSF detector is constructed based on a concept of robust contour selection and inflexion point localization, whose detection mask can be designed with the close-form FoGD. Theoretical analysis and experimental results show that the proposed operator is capable of extracting ERSFs in natural images. It is especially capable of detecting object edges and junctions from seriously noise- contaminated image.