Research On Defects Inspection Technology For Development Board Surface Based On Machine Vision

As the core part of modern electronic products,the development board has been widely used in various fields.The components of the development board in the production process are missing,and surface defects such as chip soldering defects will directly affect its normal function.The traditional development board detection method relies on labor.It not only has slow detection speed and labor intensity,but also can cause large detection errors due to subjective factors and visual fatigue.It is difficult to meet the inspection requirements in modern production lines,but machine vision technology The surface defect detection applied to the development board can effectively solve this problem,and has the advantages of high recognition accuracy,high efficiency,and no contact with the detected object.First of all,this paper aims at the basic characteristics of the development board,combines the relevant technologies of machine vision detection to design the overall plan of the development board defect detection system,the main hardware structure of the system is designed in detail,the parameter performance and selection basis of each hardware part are studied,and the analysis Software design process.Second,in order to obtain better image quality,in addition to the perfection of hardware facilities,this paper theoretically studies the pinhole model imaging of the camera,analyzes the causes of image distortion,and successfully implements image distortion correction.In addition,the image preprocessing method is also studied,and finally a series of preprocessing operations such as graying,image enhancement,denoising,and mathematical morphology transformation are designed.Finally,the defect detection algorithm is studied.Based on the characteristics of the development board,a gray-based template matching and edge-based geometric matching are combined to match the image matching mode.This method is applied to the detection of missing components of the development board and achieved good results.In the detection of development defects of the development board chip welding,the threshold segmentation method is used to extract the welding area,and the existence of welding defects is judged by analyzing their characteristic parameters.In order to further perfect the machine vision inspection system,this paper expands the database interface of the system.It facilitates the systematic management of test results.Through experiments,it is shown that the hardware configuration and software design of this system can successfully identify the common defects of the development board.It is a high-efficiency and high-precision detection method and has certain feasibility and practicability.