融合知识蒸馏与记忆机制的无监督工业缺陷检测  

Unsupervised industrial defect detection by integrating knowledge distillation and memory mechanism

作　　者：刘兵[1,2] 史伟峰[1,2] 刘明明 周勇 刘鹏[3] Liu Bing;Shi Weifeng;Liu Mingming;Zhou Yong;Liu Peng(School of Computer Science and Technology,China University of Mining and Technology,Xuzhou 221116,China;Mine Digitization Engineering Research Center of the Ministry of Education,Xuzhou 221116,China;China University of Mining and Technology Internet of Things Center,Xuzhou 221116,China)

机构地区：[1]中国矿业大学计算机科学与技术学院,徐州221116 [2]矿山数字化教育部工程研究中心,徐州221116 [3]中国矿业大学物联网中心,徐州221116

出　　处：《中国图象图形学报》2025年第3期660-671,共12页Journal of Image and Graphics

基　　金：国家自然科学基金项目(62276266)。

摘　　要：目的基于深度学习的工业缺陷检测方法可以降低传统人工质检的成本,提升检测的准确性与效率,因而在智能制造中扮演重要角色。针对无监督工业缺陷检测中存在的过检测和逻辑缺陷检测失效等问题,提出一种融合知识蒸馏与记忆机制的无监督工业缺陷检测模型。方法使用显著性检测网络和柏林噪声合成缺陷图像,提升合成图像与真实缺陷图像的分布一致性,缓解传统模型的过检测问题;同时,对传统无监督工业缺陷检测框架进行改进,引入平均记忆模块提取正常样本的原型特征,通过记忆引导提高模型对逻辑缺陷的检测性能。结果在工业缺陷检测基准数据集MVTec AD(MVTec anomaly detection dataset)上的实验结果表明,针对晶体管逻辑缺陷检测难题,在像素级接受者操作特征曲线下面积(area under the receiver operating characteristic curve,AUROC)指标上本文方法相比于基线模型提升了9.1%;针对各类缺陷检测场景,在更具挑战性的平均准确率(average precision,AP)指标上提升了2.5%。针对更具挑战性的Breakfast box数据集中的逻辑缺陷问题,本文方法在图像级AUROC指标上相较于基线模型提升了11.5%。同时,在像素级AUROC指标上,本文方法相较于基线模型提升了4.0%。结论本文不受传统缺陷合成方法的限制,能够有效缓解现有缺陷合成方法引起的过检测问题;引入平均记忆模块不仅可以减小内存开销,而且无需设计复杂的检索算法,节省了检索内存库所耗费的时间;将所提出的缺陷合成方法与记忆机制进行有机结合,能够准确检测出不同种类的工业缺陷。Objective From airplane wings to chip grains,industrial products are ubiquitous in modern society.Industrial defect detection,which aims to identify appearance defects in various industrial products,is a crucial technology for ensuring product quality and maintaining stable production.Previous defect detection methods rely on manual screening,which is costly,inefficient,and often inadequate for large-scale quality inspection needs.In recent years,the continuous emer gence of new technologies in industrial imaging,computer vision,and deep learning has notably advanced vision-based industrial defect detection,making it an effective solution for inspecting product appearance quality.However,several types of industrial defects are found in the actual scene,and a lack of sufficient samples poses challenges for existing unsupervised industrial defect detection methods,These methods often struggle to effectively detect local normal logic defects,such as when a normal target appears in the wrong position or is missing altogether.This difficulty arises from the lack of prior knowledge regarding normal samples during the testing phase,which can lead to defective parts being incorrectly identified as normal.Additionally,deep neural networks possess strong generalization capability,but existing methods often misidentify interference factors on the background of the image as defects,leading to issues of over-detection.To address the challenges of logic defect detection failures and over-detection in unsupervised industrial defect detection,a new unsupervised industrial defect detection model is proposed.Method First,a saliency detection network and Berlin noise are used to synthesize defect images,enhancing the distribution consistency between synthesized and real defect images while alleviating the over-detection problem in traditional models.Second,the proposed model comprises a teacher-student branch and a memory branch.The teacher-student branch trains the student network by distilling knowledge and synthesizing defect im

关 键 词：缺陷检测 知识蒸馏 记忆机制 缺陷合成 显著性检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]