Printed circuit board defect detection and classification method, device and computer storage medium

Abstract

The present invention provides a printed circuit board defect detection and classification method, device and computer storage medium. The method includes the following steps: S1, using the watershed algorithm to divide the color image gradient into several image blocks for the standard image, and recording the image blocks in the standard Coordinate position information in the image; S2, complete the final color image segmentation; S3, fill each segmented color image block according to the coordinate position information on the standard image; S4, obtain the image of the product to be tested, and adapt the standard module In S5, compare and analyze the matched image of the product to be tested and the standard modular image to obtain defect differences, and then classify according to the coordinate positioning information of the defect image block. The present invention can automatically correct the product image acquired visually, and perform defect detection according to the segmented image blocks, and then perform corresponding detection according to the positions of the image blocks, so as to facilitate later classification and maintenance.

Description

technical field [0001] The invention belongs to the technical field of visual detection of electronic products, and relates to a defect detection and classification method of a printed circuit board, a device and a computer storage medium. Background technique [0002] The printed circuit board is composed of multiple electronic components and multiple functional modules fixed on a rectangular substrate. The traditional quality inspection of printed circuit boards is mainly based on manual spot checks. Sampling is carried out in each process to check whether the process meets the requirements. Considering that the human eye is prone to fatigue, false detection and missed detection often occur, and some SMD components have high printing density, and there are often pressure lines or misprints, which are difficult to observe with the naked eye, and generally need to be enlarged on a microscope Observation reduces the sampling efficiency. [0003] Although there is currently a...

AI Technical Summary

Problems solved by technology

Considering that the human eye is prone to fatigue, false detection and missed detection often occur, and some SMD components have high printing density, and there are often pressure lines or misprints, which are difficult to observe with the naked eye, and generally need to be enlarged on a microscope Observation reduces sampling efficiency

[0003] Although there is currently an automated detection system for defect detection of printed circuit boards, the existing detection methods generally only detect whether there are defects in the circuit board and divide them into qualified products and unqualified products. However, we do not know the defects of intelligent detection. Which patch position of the printed circuit board has defects; and because the products to be tested on the detection line conveying line are placed unevenly, the angle of view is inclined, which is easy to cause errors in product detection, which puts forward a series of tests for intelligent detection.

Images