首页|面向工业缺陷分类的交互式易混淆缺陷分离方法研究

面向工业缺陷分类的交互式易混淆缺陷分离方法研究

扫码查看
在工业生产中会根据严重程度对缺陷做不同处理,所以需要对缺陷进行分类。但是,实际生产中经常因为存在一些易混淆缺陷而导致分类精度不够,使得在生产实践中只能对所有缺陷进行保守处理,带来很大人力成本和经济代价。为解决该问题,提出一种交互式易混淆缺陷分离方法,将少量易混淆缺陷从其他缺陷中分离出来,从而保证剩下的绝大部分缺陷的分类结果能被直接使用。首先,将训练数据中的易混淆缺陷挑选出来作为一个或多个新缺陷类别,称之为虚缺陷,从而使得训练所得网络能区分虚缺陷和其它类缺陷。其中,还设计了一套可视化界面辅助用户交互地挑选易混淆缺陷以构建虚类别。使用实际工业现场的CMOS缺陷数据进行有效性验证,结果表明所提方法能快速分类出易混淆缺陷,并保证剩余缺陷的分类精度满足工业应用要求。
An interactive separation method for confusable defects in industrial defect classification
In industrial production,defects are treated differently based on their severity,so it is nec-essary to classify defects.However,in actual production,the classification accuracy is often insufficient due to the presence of few easily confused defects,which requires conservative treatment of all defects in production practice,resulting in significant human and economic costs.To solve this problem,this pa-per proposes a method for interactive separating easily confused defects.This method separates few easi-ly confused defects from other defects,ensuring that the classification results of the remaining majority of defects can be directly used.This method selects easily confused defects from the training data as one or more new defect categories,called virtual defects,so that the trained network can distinguish be-tween virtual defects and other defects.This paper designs a visual interface to assist users in interac-tively selecting easily confused defects to construct virtual categories.CMOS defect data from actual in-dustrial sites are adopted for effectiveness verification,and the results show that the proposed method can quickly classify few confusing defects and ensure that the classification accuracy of remaining defects meets the requirements of industrial applications.

surface defect classificationconfusable defectdeep learningvisual analysis

罗月童、李超、周波、张延孔

展开 >

合肥工业大学计算机与信息学院,安徽 合肥 230601

表面缺陷分类 易混淆缺陷 深度学习 可视分析

国家自然科学基金国家重点基础研究发展计划安徽省科技攻关计划

616021462017YFB14022001604d0802009

2024

计算机工程与科学
国防科学技术大学计算机学院

计算机工程与科学

CSTPCD北大核心
影响因子:0.787
ISSN:1007-130X
年,卷(期):2024.46(3)
  • 16