基于机器视觉的胶体面积检测判定系统的研究与实现  

Research and Implementation of a Colloid Area Detection and Determination System Based on Machine Vision

作　　者：戴白杨 李立军[1] 刘天成[1] 张天禹 赵芮延 DAI Baiyang;LI Lijun;LIU Tiancheng;ZHANG Tianyu;ZHAO Ruiyan(Advanced Technology&Materials Co.,Ltd.,Beijing 100081,China)

机构地区：[1]安泰科技股份有限公司,北京100081

出　　处：《自动化应用》2024年第13期52-55,共4页Automation Application

摘　　要：为解决纳米晶汽车级共模电感因点胶质量不足而导致铁芯和护盒粘连不够牢固,人工单个直接称重劳动强度大、效率低、容易误判的问题,创新性提出一种基于机器视觉检测胶体面积大小,间接判断胶体质量是否合格的判定系统。该系统通过机器视觉系统采集胶体面积图像,运用VisionBank软件开发系统经图像预处理、特征定位、掩膜处理、斑块面积计算等算法完成对胶体面积的自动检测。研究表明,胶体的面积和胶体质量存在一定的比例关系,通过对比分析面积和质量的关系得出胶体面积的判断标准。此判定系统检测速度约0.8 s/个,检测准确度在95%以上,极大地提高了纳米晶汽车共模电感产品品质。In order to solve the problem that the iron core and protective box are not firmly adhered due to the insufficient dispensing quality of nanocrystalline automobile grade common mode inductor,the manual single direct weighing has high labor intensity,low efficiency and is prone to misjudgment,a judgment system based on machine vision is proposed to detect the large and small colloid area,and then judge whether the colloid quality is qualified.The colloidal area image is collected through the machine vision system,and the visionbank software development system is used to complete the automatic detection of colloidal area through image preprocessing,feature positioning,mask processing,plaque area calculation and other algorithms.The data research shows that there is a certain proportional relationship between the area of colloid and the quality of colloid.Through the comparative analysis of the relationship between area and quality,the judgment standard of colloid area is obtained.The detection speed of this judgment system is about 0.8 s/per unit,and the detection accuracy is more than 95%,which greatly improves the product quality of nanocrystalline automobile common mode inductor.

关 键 词：机器视觉 胶体面积 纳米晶铁芯 共模电感 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]