Image edge extracting method based on color block point features

Abstract

The invention discloses an image edge extracting method based on color block point features, wherein the method includes steps of pretreating a color image, transforming and storing to be a grey image; setting a pixel threshold value; traversing the grey image by a gliding window, judging the conditions of the pixel point in the gliding window, finding out and marking all internal color block points; traversing the pixel points of the grey image by the gliding window one by one, and judging if the pixel point within the neighbor domain scale where the internal color block point locates is marked, and marking the unmarked pixel points as a color block side point; traversing the pixel point one by one again by the gliding window, and judging if the pixel point of every color block side point in the neighbor domain scale is marked, and marking the unmarked pixel point as the boundary side point; marking the rest unmarked pixel point as the internal boundary point; connecting all boundary side points in the grey picture to be the color block and extracting the color block as the image edge. The edge extracted by the method is a complete continuous edge, and the threshold value is flexible to control, simple and high-efficient.

Description

technical field [0001] The invention relates to an image edge extraction method based on color block point features, and belongs to the technical field of image edge extraction. Background technique [0002] Image edge is an important part of image recognition and image analysis, and the edge information of various parts in the image is often an important feature of pattern recognition. When identifying a specific object, it generally starts with edge features, which can better reflect the shape features of the object compared to the internal texture and color information of the target. [0003] An image edge detection method based on a differential operator is a commonly used edge detection algorithm in the prior art, that is, a filter template is used to perform weighted calculations on an image. At present, there are many common methods, such as Sobel operator, Canny operator, and Prewitt operator, but the traditional differential operator method often has disadvantages ...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art, provide an image edge extraction method based on color block point features, and solve the problem that the existing image edge extraction method has poor flexibility and the continuity of the extracted edge is not complete , the problem of being unable to achieve an efficient and fast extraction process

Images