Battery screen printing quality detecting method based on block template matching

Abstract

The invention discloses a battery screen printing quality detecting method based on block template matching. Original images of batteries are acquired to build a template data set; images of the batteries are corrected; the corrected images of the batteries are tailored; detecting areas are extracted to reduce information redundancy; battery screen printing images are cut from battery backgrounds;a battery screen printing position is determined; block matching is performed on images to be detected by using block template images in the template data set to judge the matching degree between images of the batteries to be detected and templates; block sub images of the batteries to be detected are compared with built template sub images one by one; difference results are analyzed to obtain battery screen printing defect information; aberration quantitative analysis is performed on each sub image to obtain aberration conditions of each part in global images; and meanwhile, average processing is performed to observe global aberration conditions to finish the battery screen printing quality detection. The method needs no manual intervention, and can realize quick high-precision battery screen printing quality detection.

Description

technical field [0001] The invention belongs to the technical field of machine vision automatic surface detection, and in particular relates to a battery silk screen quality detection method based on block template matching. Background technique [0002] The screen printing of the test battery is an important step in the battery assembly process. Different types of batteries have differences in character types (Chinese characters, Korean, English, numbers, etc.), character formats, illustration content, and barcode formats. At present, the screen printing of battery cells is affected by fixtures, equipment, personnel and other related factors to cause printing defects. The types of screen printing defects are mainly divided into: silk screen defect / skew / fuzzy / ghosting / dirty / chromatic aberration, screen printing and barcode information do not match, and screen printing position is wrong. Comply with specifications etc. At present, battery manufacturers mainly rely on artific...

AI Technical Summary

Problems solved by technology

However, most of the current research on silk screen defect detection algorithms is focused on the recognition rate and stability of the algorithm, and the real-time detection is taken as a secondary consideration, resulting in many excellent methods that can only be run in the experimental environment or offline testing

However, some advanced automated screen printing testing equipment in foreign countries are expensive, making it difficult for many domestic companies to accept

Images