Sub-pixel rapid recognition method of batch group circle vector based on region movement

Abstract

The invention provides a batch group circle vector sub-pixel rapid identification method based on area movement, which performs binarization preprocessing on the collected group circle images, and sequentially establishes data acquisition windows in each connected area formed by connected segmentation, Move the data acquisition window in the connected area to perform multiple data collection and circle fitting, and determine the optimal circle recognition position in the corresponding connected area on the basis of the circle fitting results and the recognition accuracy, and perform accurate circle recognition at the optimal circle recognition position. For circle fitting, the invention has low requirements for images, is convenient and fast to process, is suitable for rapid and precise detection and analysis of grouped circular workpieces on production lines, and can meet the needs of rapid and precise detection and analysis of a large number of dense circle elements in fine circuit board processing, and It also has good promotion and application value for other similar image processing needs, and has a very good promotion and application prospect.

Description

technical field [0001] The invention relates to an image recognition method, in particular to a batch group circle vector sub-pixel recognition method. Background technique [0002] In the precision design and volume manufacturing process of tiny printed circuit boards, the manufacturing accuracy of the relevant geometric elements is very important. The detection of printed circuit boards is currently mostly based on video image analysis. [0003] Hough transform is currently the most commonly used algorithm in the field of group circle automatic image recognition. The basic principle of Hough transform is to use the duality of point and line to transform the given curve in the original image space into a point in the parameter space according to the expression form of the curve. In this way, the problem of detecting a given curve in the original image is transformed into clustering the pixels in the parameter space that have a certain relationship with the image space, so...

AI Technical Summary

Problems solved by technology

[0013] 1. In the process of circle detection, the parameters are increased from two parameters of a straight line, that is, intercept and slope, to three, that is, the coordinates of the center of the circle and the radius. It is a one-to-many mapping, so the amount of calculation is large;

[0014] 2. It takes a lot of memory space, takes a long time, and has poor real-time performance;

[0015] 3. Images in reality are generally disturbed by external noise, and the signal-to-noise ratio is low. At this time, the performance of the conventional Hough transform will drop sharply. When searching for the maximum value of the parameter space, it is difficult to determine the appropriate threshold, often appearing " "Virtual peak" and "missing detection" problems

Images