Workpiece edge extraction method based on improved Canny operator

Abstract

The invention relates to a workpiece edge extraction method based on an improved Canny operator, which comprises the steps of performing convolution processing on an original image from the zero point; weighting the convolution result to obtain a weighted grayscale map; calculating the gradient direction and amplitude of the weighted grayscale map; performing non-maximum suppression on the gradient; determining image edges by applying high and low thresholds; and extracting and connecting the image edges. According to the method, the improved Canny operator makes up a problem that it is difficult for the classical Canny operator to give consideration to both image edges and noise in a filtering stage due to selection of Gauss filtering parameters and a problem of the adaptivity of double threshold setting. Experimental results show that the improved Canny operator better balance the relation between the noise and the edge information and has a better extraction effect for workpiece edge features.

Description

technical field [0001] The invention relates to the field of visual positioning of industrial robots, in particular to a workpiece edge extraction method based on an improved Canny operator. Background technique [0002] The edge of the image refers to the place where the gray value has a sudden change in the image, and it is the dividing line between the workpiece to be tested and the background pattern. Edge acquisition is an important link in the robot vision system. The edge of the image contains a large amount of information of the workpiece to be tested, and is one of the most basic features of the image. The effect of edge extraction is directly related to the fitting, classification and recognition processes in subsequent image processing. In engineering applications, the edge of the image is also often used as the recognition basis for people to detect workpiece defects, define the size of the workpiece, and locate the workpiece. Therefore, it is very important f...

AI Technical Summary

Problems solved by technology

However, the Canny operator still has some shortcomings. Since the classic Canny operator uses a Gaussian filter to smooth the image, the selection of its smoothing parameter σ has a decisive impact on the filtering effect.

In addition, although the classic Canny operator uses high and low thresholds to extract edges, it is more flexible than the single threshold method, but there are still common problems of threshold segmentation.

Images