Font Size: a A A

The Design Of Visual Inspection System For Bottle Printing

Posted on:2019-11-23Degree:MasterType:Thesis
Country:ChinaCandidate:H F ZhuFull Text:PDF
GTID:2371330545457423Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Glass bottles need to be fired,printed,packaged,and inspected prior to surface printing.In the glass bottle printing process,due to the complex printing process,high-speed printing links and low-precision mechanical systems,a variety of appearance defects are inevitable in the printing process,such as background shift,missing printing,printing ink is not Even appearance defects.These defective defective tasting bottles must be tested for defects and rejected.Due to the defects of poor repeatability,inconsistent consistency of inspection results,low accuracy,high false positive detection rate,and high subjectivity,traditional manual inspections are difficult to meet the high-speed,high-precision,high-quality and safe production requirements of industrial production lines.Therefore,it is very important social and economic significance to study the machine vision inspection technology to replace the traditional human eye inspection.The main work and innovations of this thesis are as follows:1.Explained the status quo of surface defect detection in wine bottle printing,and combined with the actual needs of the industrial site,designed the technical solution of the entire visual system.In this set of equipment,a set of self-rotating mechanism with a clamping device was specially designed,and a printing image of one week of the bottle mouth was obtained through this mechanism,so that the full-scale detection of the printing surface of the wine bottle was realized.At the same time,the electrical control module was designed and a suitable lighting scheme was designed after studying the transparent characteristics of the bottle.2.In the preprocessing algorithm for printed images,image enhancement is performed by gradation transformation and morphological processing.The image denoising is performed by comparing several filtering algorithms.In the segmentation and extraction of defects,a shape-based template matching method is adopted,and a Gauss pyramid acceleration strategy is applied to speed up matching.In the feature analysis of defects,image difference and morphological processing were applied to analyze the details of the defects.Based on the above,the defect detection of the bottle surface printing was completed.3.Developed a visual inspection prototype and developed a database system to save all image and defect data.The visual inspection prototype is tested on the industrial site and the actual speed can reach 1 800 per hour.Through the practice on the industrial site,the stability and reliability of the bottle defect visual inspection equipment have been proved.Fully meet the testing requirements of high precision,high speed and high real-time performance of industrial production lines.
Keywords/Search Tags:machine vision, the external wine bottle printing, external assessment, image difference
PDF Full Text Request
Related items