Edge Detection of Multi-scale Colour Image Based on Neighbourhood Feature

In order to improve general adaptive capability of algorithm, a new multi-scale color image edge detection algorithm based on neighborhood feature is proposed. Based on the theory of multi-scale, with the help of neighborhood characteristics and fuzzy dissimilarity function algorithm accurately extract the edge. Compared with traditional edge detection algorithm based on neighborhood characteristics of multi-scale color image edge detection algorithm maximum extent retains the color information, and to achieve the adaptive color image edge extraction. The experimental results show that the method take full advantage of color information, the color image edge extraction has good detection precision and accuracy.