Color image edge detection method based on quaternion theory and canny algorithm

Abstract

The invention discloses a color image edge detection method based on a quaternion theory and a canny algorithm, and belongs to the field of image processing and computer vision. The method comprises the steps that 1) an original color image is represented by a quaternion; 2) smooth filtering is carried out on the color image; 3) a color quaternion edge characteristic matrix is built and used for analyzing the edge characteristics and the structural characteristics of the color image; 4) non-maximum value restraining is carried out on the gradient magnitude of the image obtained at the step 3); 5) dual threshold method detection and edge connection are carried out. The algebraic operation and the spatial characteristic of the quaternion are utilized, all the pixel points of the color image are represented by the pure quaternion, the color quaternion edge characteristic matrix is built and used for analyzing the edge characteristics and the structural characteristics of the color image, edge neighbourhood brightness hopping and color hopping are fused, color information missing in a traditional algorithm is avoided, the introduction of noise is reduced, and locating precision and anti-noise performance are improved.

Description

technical field [0001] The invention belongs to the field of image processing and computer vision, and in particular relates to a quaternion, an algebraic tool, combined with its theory and the traditional grayscale Canny algorithm, which has good anti-noise performance and high positioning accuracy. A method for edge detection in color images. Background technique [0002] The edge of the image is defined as a discontinuous point in the image function, which contains a large amount of image feature information, and is the key to distinguish between objects and backgrounds, regions of interest and surrounding information. The traditional edge detection algorithm is mainly to grayscale the image to be processed and then process it, so that the edge of the image to be obtained is regarded as a collection of pixels with sudden changes in the gray value in the neighborhood. Traditional classic algorithms include Roberts operator, Prewitt operator, Sobel operator, Canny operator...

AI Technical Summary

Problems solved by technology

[0004] The existing grayscale algorithm for color image edge detection has many defects, such as literature 1 (Li Haiyang, Wen Yongge. An improved color image edge detection method [J]. Journal of Mianyang Normal University, 2013, 32(8 )) mentioned in the Roberts algorithm is very sensitive to noise in edge detection, document 2 (Wang Xuan. Research and implementation of liquid level detection system based on image edge extraction [D]. Hunan University, 2010) and document 3 (Jin Pengfei. An Improved Sobel Image Edge Detection Algorithm[J][J]. Applied Optics, 2008, 29(4):625-628) Although the prewitt algorithm and the sobel algorithm mentioned can filter out part of the noise, the anti-noise performance is still Very poor, entraining more noise edges, and the edges are thicker, document 4 (Huang Jianling, Zheng Xuemei. An improved image edge extraction algorithm based on Canny operator[J][J]. Computer Engineering and Application, 2008,44( 25): 170-172), the canny algorithm mentioned in has a better effect of filtering noise, but it often contains false edges, and its robustness to color image edge detection is not good. In addition, these classic grayscale algorithms have common problems, That is, reprocessing after grayscale will inevitably cause the loss of edges, mainly focusing on the loss of color jump edges, and the problem of low positioning accuracy

Although the quaternion rotation color edge detection method mentioned in Document 5 (Xing Yan. Quaternion and its application in graphics and image processing [D]. Hefei University of Technology, 2009) well preserves the neighborhood edge color information, but the positioning accuracy can be improved, and the anti-noise performance is not very ideal

Images