Research On Automatic Generation Method Of Component PCB Footprint Library Based On Machine Vision

Electronic components are the most basic units on the Printed Circuit Board(PCB),while the PCB take on the important responsibility of the electronic components assembly,effectively completing the signal connection between electronic components and control module,and realizing the electrical function control.Its development and design efficiency affects the exploitation period and prime cost of electronic products directly.The primary task of designing printed circuit board is to draw a footprint library that can accurately match the electronic components.It is directly related to whether the electronic components can be soldered with high quality and precision.The packaging type of electronic components has the division of SMD and plug-in.At present,the industry has been able to realize the automa tic generation of the package library for SMD components.But the production method for plug-in components uses manual calculation,which exist shortcomings such as complicated calculation,low efficiency and error prone.The last step is to mount electronic components after the completion of printed circuit board physical forming.If it is found that the material and the hole position of components on PCB are not connected to each other until the final mounting step,it will not only prolong the development cycle of electronic products,but also increase the cost of enterprises and reduce the competitive advantage due to improper packaging and manufacturing.Therefore,it is an inevitable trend to explore an automatically generate plug-in components package production method and to shorten the enterprise's research and development cycle,reducing the cost of enterprise research and development.Based on machine vision and centralized the plug-in components,a method that can automatically generate PCB footprint library for plug-in components is realized.The parameter information needed for packaging production includes the peripheral dimension information of the component,the position information of each pin and the diameter information of the pin.The system use intelligent machine vision technology to algorithmically process the specific picture data in the component data manual,and then extracts the profile information and size information necessary for the production of the package library,and use PCB design software to automatically produce and generate the plug-in component package library.The main tasks completed include: it will also summarize and analyze a variety of plug-in components' data manuals.According to the plug-in's shape connection characteristics,pin characteristics and label size characteristics,the top view in the components data manual is determined as the first choice for intercepting pictures.The peripheral contours are clear and the size of the digital label is complete.And it intercepted a large number of plug-in components of the top view to save the name and as the original image of the system.In the application of graphics processing technology,it carries out pre-processing on the original image,mainly including gray-scale transformation and enhancement,and achieved the image filtering and smoothing.What's more,it use Gaussian,mean and median three filtering methods for comparison and peak signal-to-noise ratio and other denoising parameters to determine the median filtering as the best form of smooth noise reduction processing.The global and local threshold segmentation was compared in the image segmentation process,and the research found that the results of local threshold image segmentation are more suitable for extracting the contours of foreground objects in complex backgrounds.The application effects of the three edge detection operators are compared,and denoising parameters are introduced to verify the canny edge detection operator as the mai n contour extraction algorithm.Using regional morphological calculations,the outermost circumscribed rectangle of the component and the smallest circumscribed rectangle representing the pin are finally extracted.The obtained peripheral contour information,pin positioning information and pin diameter information are written into the document and saved.The PCB design software is imported to complete the package drawing,and lots of experimental data are used to verify the accuracy of the system.This solution is based on Microsoft Visual Studio as the basic development tool and.NET Framework as the framework to realize an automatic footprint library generation system for human-machine interaction plug-in components with a visual interface.Compared with the traditional manual method of footprint library,the automatic generation system of plug-in components takes less time and has a low error rate.It verified the feasibility of the project,which has high engineering value.