Two-value image connected domain marking algorithm based on AOI bullet apparent defect detecting system

Abstract

The invention discloses a two-value image connected domain marking algorithm based on an AOI bullet apparent defect detecting system. The two-value image connected domain marking algorithm comprises the following steps: (1) utilizing an AOI system to acquire original grayness image of the surface of a bullet; (2) performing binaryzation division on the original grayness image to obtain a two-value image of the surface of the bullet; (3) scanning the two-value image from left to right and from top to bottom, and performing initial marking; (4) establishing an equivalent table according to the marking; (5) performing secondary scanning on the image with the marking, performing corresponding replacement on the equivalent satisfying the conditions, and eliminating connected domain marking conflict. The two-value image connected domain marking algorithm changes the marking mode in the conventional algorithm, performs overall scanning and marking on the image, establishes the equivalent table, and finally performs direct overall matching and replacing during the secondary scanning, so that the operation efficiency is improved, the condition that a plurality of marking values exist in the same connected domain is greatly improved in speed.

Description

[0001] technical field [0002] The invention relates to an AOI-based bullet apparent defect detection system, in particular to a binary image connected domain marking algorithm of the AOI-based bullet apparent defect detection system. Background technique [0003] AOI (Automatic Optic Inspection), also known as automatic optical inspection, is based on motion machine vision as a basic technology, as a method to improve the shortcomings of traditional human inspection using optical instruments, and to improve the accuracy and speed of optical image inspection systems. technology. [0004] The inventor studied the preliminary image of the bullet surface obtained by the AOI system in the bullet apparent defect detection, and then preprocessed the image to obtain a binarized image. Before the final feature extraction and recognition, the binarized Images are marked with connected domains to facilitate and accurately extract features. [0005] The so-called connected domain ma...

AI Technical Summary

Problems solved by technology

[0015] As for the apparent defect detection of bullets, there is a relatively big problem that often occurs in the same connected domain, and different tag values ​​appear. At this time, it will be very time-consuming to scan the entire graph again, which will affect the speed of the entire algorithm. , thereby reducing the overall efficiency of bullet appearance detection

Images