Image edge detection method and system

Abstract

The embodiment of the invention provides an image edge detection method and system. The method includes the steps of obtaining a gradient magnitude value of each pixel point in an original image according to a roberts operator detection algorithm and obtaining a gradient image, wherein roberts operators adopt a 3*3 operator template; conducting non-maximum value inhibition treatment on the gradient image and obtaining a candidate edge image, wherein pixel points serving as candidate edge points are marked in the candidate edge image; according to a canny operator detection algorithm, conducting edge detection on the candidate edge image to obtain an edge image. According to the embodiment, through the roberts operators adopting the 3*3 operator template, the accurate gradient magnitude values are figured out, and a great condition is created for applying the canny operator detection algorithm; by adopting canny operators for screening the pixel points, serving as the candidate edge points, of the border of the candidate edge image, the calculation amount of the canny operator detection algorithm can be decreased, the calculation precision can be improved, and meanwhile the smoothness and the advantages of extremum inhibition are preserved.

Description

technical field [0001] The present invention relates to the technical field of image processing, and more specifically, to an image edge detection method and system. Background technique [0002] In the image, the edge is the place where the local intensity of the image varies most obviously, and it mainly exists between the target and the target, the target and the background, and the region and the region. Edges indicate the end of one characteristic region and the beginning of another. The internal features or properties of the areas separated by the edges are consistent, but the internal features or properties of different areas are different. [0003] Roberts operator, also known as Roberts operator, is the simplest operator, which is an operator that uses local difference operators to find edges. The traditional Roberts operator is a 2x2 operator template, which uses the difference between two adjacent pixels in the diagonal direction to detect edges. Judging from t...

AI Technical Summary

Problems solved by technology

However, there is a disadvantage that the local pixel change gradient constructed by the Roberts operator of the traditional 2*2 operator template cannot be directly combined with the traditional canny operator of order 3*3

[0005] The edge detection of the image in the complex background of the taxi is easily affected by external factors such as environmental noise and changes in lighting conditions. Therefore, the traditional edge detection method has the disadvantages of weak anti-noise performance and discontinuous edges, resulting in poor detection results. , and requires manual selection of parameters, the algorithm has low flexibility

Images