Techniques Of High Reflecting Surface Defect Detection

The product reliability and quality revolution are the challenge of the manufacturing industry nowadays,and it is also the focus that every manufacturing field should pay attention to.Only by constantly improving product quality and production efficiency can we be in an invincible position in the rapidly developing manufacturing field.Product appearance is an important aspect of reflecting product quality However,some products with surface reflection characteristics are difficult to detect defects,such as irregular metal surfaces,or reflective coating surfaces.For these surface defects detection,due to the unreliability of machine vision equipment detection,most domestic companies still use manual detection methods.Manual detection is not only inefficient,but also leads to a large number of false detections and missed detections due to the visual fatigue of the inspectors and the different judgments on the defect standards.Based on image processing and machine vision technology,research on surface quality of tin-plated pins on electronic components mainly include light sources and defect recognition algorithms which for the irregular and highly reflectibe surface defects detetion,and proposes an approach to detect the defects of electronic elements.The device's highly reflective surface pins are a relatively versatile and viable detection solution and have been developed and designed to complete a complete set of machine vision inspection equipment.The main research work includes:1.A complete set of high-reflection surface defect detection solutions was designed.A new type of light source lighting scheme was proposed for the high reflection characteristics of the product surface,which greatly reduced the interference of thereflection characteristics with the product surface defect detection.At the same time,according to the requirements of product surface defect detection,a system with multiple viewing angles for simultaneous detection was designed.2.In-depth analysis of product surface defect type characteristics,designed a set of image processing algorithms.Using channel analysis,filter enhancement,image segmentation and other algorithms to achieve the detection of surface defects in high-reflective products.3.Based on software engineering methods,analysis of user needs,based on C#language designed a complete electronic components full appearance detection system software.Software includes camera control,image debugging,parameter setting,user management,data query and other modules.4.Construting a convolution neural network deep learning algorithm framework based on analysis of high-reflection surface defects.The algorithm framework can realize the defect classification of highly reflective surface defect samples and achieve high reflective surface products defects detection.Through the experiment and test of the surface defect detection of the tin lead pin of electronic components,the detection scheme proposed in this paper has good implementability.The proposed detection algorithm can better achieve the detection of high reflection surface defects.The results of application on the actual industrial production show that the plan meet the requirements of majority electronic components pin defect detection,and the detection rate is 97.9%...