Font Size: a A A

Research On Key Technologies Of Two-color Printing Machine Printing Quality Control System

Posted on:2017-03-29Degree:MasterType:Thesis
Country:ChinaCandidate:H T KanFull Text:PDF
GTID:2351330512976503Subject:Mechanical Manufacturing and Automation
Abstract/Summary:PDF Full Text Request
With the improvement of living standard,people put forward higher request to the quality of the printing.The printing quality control system of double color metal sheet printing machine is a key factor to determine the quality of printing products.Due to the drawback of printing quality control system in current printing industry,the technology of image processing and pattern recognition are adopted in this paper.The main research contents of this paper are summarized as follows:According to the control requirements of the quality of sheet metal printing,a set of image acquisition device will be developed based on machine vision.Combined with the size and precision of image acquisition for color printing,the CCD camera and image acquisition cards are chosen.According to the different characteristics of different color space,select the appropriate color space f1f2f3.Designed and produced a standard sample of printed dots images that contains 9 kinds of color pattern information.On this basis,Anisotropic weighted fuzzy C-means clustering?AWFCM?is proposed,which is used to collect the data samples of each color pattern.Finally,an algorithm based on competitive learning is used to capture the image information.In view of the complexity and diversity of color pattern of color printing dot image,Firstly,the color pattern of dot is identified by using the coarse divider which is constructed by the extreme learning machine?ELM?,then use the neural network group to build the fine division do further classification,and the pixel color is separately identified by each sub network.Finally,the integration of the output of each sub network is carried out by the fuzzy integral,so as to further improve the classification accuracyThis paper presents an improved algorithm by PID neural network on the printing machine ink key opening adjustment.And the printing quality control system is validated by using MATLAB simulation.The simulation results show that the algorithm has good stability and convergence speed,and will meet the requirements of the control system.At last,the printing quality control system was developed by using C#,which realized the basic function of printing quality control.
Keywords/Search Tags:Dot Image, Image Processing, Pattern Recognition, Color Pattern, Printing Quality Control
PDF Full Text Request
Related items