Research On On-line Detection Technology Of Metal Plate Printing Quality

Florescent coating is printed on metal plates which are used to be road signs and vehicle signs to attractive attentions.These surface with florescent coatings has characteristics such as colorful and strong light reflection.Printing surfaces are required high qualities so that color spots,incorrect location,uneven printing and pattern defects are not allowed.At present,the printing of florescent coating for metal plates has been produced automatically,but the printing qualities’ monitoring still manual.This detection method not only labor-intensive and inefficient,but also seriously affect the health of workers by those toxic and harmful substances in the coating.The inspection quality is easy to be affected by the subjective factors of the workers,so it is difficult to ensure the unity of the inspection standards;after working for a long time,the workers are very prone to visual fatigue,and are prone to missed inspection and wrong inspection.Therefore,it is of great theoretical and practical significance to study the on-line automatic detection technology of sheet metal printing quality.In this paper,a set of on-line inspection system for the quality of metal plate printing based on machine vision technology is set up,and the key technologies are studied systematically.The main research contents are as follows:Considering the high reflective characteristics,inspection requirements and production environment of the motor vehicle sign surface,an automatic detection device for the printing quality of metal plate based on machine vision was set up.The system includes a product transfer module,an image acquisition module,an image processing module,and a defective product capture Modules and defective product storage modules,where the image acquisition module includes industrial cameras,lenses,light sources,etc.;the image processing module is an image processing software system integrated on the industrial computer;the defective product grasping module and defective product storage module mainly include a six-axis manipulator,Single axis robot and storage trolley.The light source setting method of the motor vehicle sign printing surface with mirror characteristics and retro-reflective characteristics is studied,and the selection methods of light sources,industrial cameras,lenses,etc.are given to realize the online acquisition of high signal-to-noise ratio images.Aiming at the problem of being susceptible to noise during image sharpness evaluation,a high-frequency coefficient sharpness evaluation algorithm based on twodimensional discrete wavelet transform is proposed.The influence of the number of wavelet transform layers on the evaluation of image sharpness is studied.The experimental comparison with the gradient square function method,variance function method and FISH algorithm verifies the superiority of the proposed algorithm.The on-line identification technology of typical printing defects of motor vehicle signage is studied.Color space conversion technology,chroma analysis technology,sub-pixel edge extraction technology,template matching technology,morphology processing technology,pixel grayscale rotation interpolation technology,etc.are studied to identify the defects such as color spots,uneven printing,misprinted patterns,and incorrect positioning.Based on software development platforms such as C # and Halcon,a system for testing the quality of motor vehicle sign printing has been developed,and online quality testing has been performed using the developed system under actual producing conditions.The results show that the developed automobile plate printing quality detection system can meet the actual production needs.