PCB Board Defect Detection Technology And Application Based On Image

Nowadays,the electronics industry is playing an increasingly important role in the development of national economy.As an highly information integration for various electronic components,printed circuit boards(PCB)are widely used in various fields of the electronics industry.The continuous development of the economy leads to the constantly improvement of electronic technology,and light,convenient electronic technology has become the trend.Simultaneously high density and high integration become the trend in development of PCB,which brings a very great challenge for the quality inspection of traditional PCB industry.The traditional manual testing method has the problems of slow speed,long time and easy to leak,and so on.It can not adapt to the rapid development of technology and craftwork.How to realize accurate and efficient PCB automatic defect detection has always been a great concern problem in the field of electronics industry.At the same time,individuals and minor enterprises has an increasingly high demand in detecting PCB defects.To achieve low cost,high precision is the primary consideration.Therefore,it is of great significance to study how to improve the accuracy of PCB defect detection by image processing technology of low cost.The PCB defect detection process based on image includes image pre-treatment,image registration,image segmentation and image recognition,in which image pre-treatment includes image enhancement,image smoothing and image sharpening operations.According to the process and the requirement of test,the three main work and research contents are as follows:1.Compare the algorithm to get the appropriate algorithm.In order to obtain the ideal target image after image registration and image segmentation,this thesis analyzes the algorithm based on the existing algorithm,and finally selects the gray scale transformation,the adaptive filtering and the gradient operator as the processing algorithm of pre-treatment process,and select the maximum inter-class variance threshold as the image segmentation operation algorithm.The analysis shows that the obtained algorithm is beneficial to the registration and recognition of PCB images.2.Improve RHT to raise registration accuracy.In order to solve the problem of large amount of computation and time consuming in traditional algorithms,this thesis presents a new algorithm for PCB plate Mark localization based on principal component analysis(PCA)and segmented RHT.Firstly,the image is binarized by adding the original color PCB image,and the canny operator is used to extract the edge of the image.Then,the intersection and small line in the image are removed,and the remaining lines are marked,And then use the segmented RHT analysis to get the number of rounds and related parameters.Finally,combined with the above obtained circular parameters to get the desired circle Mark using least squares fitting.The experimental results indicates that the algorithm can raise the recognition accuracy and positioning accuracy of the circle Mark effectively compared to the traditional template matching and random Hough transform.3.Design system to identify defects.Using MATLAB image tool library and the principle of realization and improvement of the above algorithm to make a software of PCB defect detection by the function of the MATLAB GUI.Firstly,the thesis analyzes the processing flow and defect recognition method of PCB defect detection software,and then shows the function of the software system.Finally,the software is tested by detecting the short circuit and the recognition of the burr defect..The experimental results show that the software system can accurately detect the short circuit,circuit breakage,defect and burr of PCB board.