The Design And Implementation Of Printed Defect Detection System

Various printing defects,such as missing prints,spots and scratches,may occur in high-speed printing production.The traditional manual inspection method has great limitation,which brings a lot of disadvantageous factors to the printing quality control.With the continuous development of image processing technology,the complexity of images is getting higher and higher,and the requirements for printing speed are also getting higher and higher.The traditional production mode of printing quality control has been unable to meet the actual needs of current printing production.In order to improve the efficiency of printing production,an automatic method is urgently needed to be applied in printing production.This system combines printing production with machine vision to ensure product quality.Firstly,under the appropriate light source system,the printed image is obtained by CCD camera to generate the data source to be processed.Secondly,GPU is used to preprocess the data source and carry out image matching.In image preprocessing,the detection ability of information should be enhanced as much as possible and the subsequent operation should be maximized.In image matching,the multi-template matching method of improved similarity algorithm is used to extract defects.Among them,the improved similarity algorithm has neither multiplication nor normalization,so it requires less computation and has good real-time performance.The multi-template matching method can effectively improve the inspection accuracy and maintain a high inspection speed.Then,shape defect classification algorithm and color defect classification algorithm are used to classify defects.Finally,the defects are evaluated in the form of a table,and the fault warning is carried out according to the running state and customer requirements.The system combines high-performance technology with image processing technology to ensure efficient production.In the field of machine vision,especially in the field of image processing,GPU plays an important role.The parallel computing power of GPU can simplify and develop tasks related to image processing.This system uses GPU as the underlying hardware and CUDA to provide parallel data processing,which can identify defects at a high speed and at the same time has a low energy consumption.Based on computer vision technology,the system provides an effective solution to the problems of slow speed and low accuracy in the defect detection of printed matter in the market.For the growth of the entire printing industry played a catalytic role.