Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (11): 3200-3205.DOI: 10.11772/j.issn.1001-9081.2020121974

• Artificial intelligence • Previous Articles     Next Articles

Surface defect detection method based on auto-encoding and knowledge distillation

Taiheng LIU, Zhaoshui HE()   

  1. School of Automation,Guangdong University of Technology,Guangzhou Guangdong 510006,China
  • Received:2020-12-15 Revised:2021-07-28 Accepted:2021-08-03 Online:2021-05-11 Published:2021-11-10
  • Contact: Zhaoshui HE
  • About author:LIU Taiheng, born in 1993, Ph. D. candidate. His research interests include defect detection,deep learning,recommendation system
    HE Zhaoshui,born in 1978,Ph. D.,professor. His research interests include intelligent information processing,machine learning.
  • Supported by:
    the National Natural Science Foundation of China(61773127);the Leading Talents in Scientific and Technological Innovation of the “Ten Thousand Talent Program” of China in 2018, the Key Project of Joint Foundation of the Guangdong Basic and Applied Basic Research Foundation(2019B1515120036);the Natural Science Foundation of Guangdong Province(2018A030313306);the Guangzhou Science and Technology Foundation(201802010037);the Research and Development Program of Key Areas of Guangdong Province(2019B010147001)

基于自编码和知识蒸馏的表面缺陷检测方法

刘太亨, 何昭水()   

  1. 广东工业大学 自动化学院,广州 510006
  • 通讯作者: 何昭水
  • 作者简介:刘太亨(1993—),男,广东肇庆人,博士研究生,主要研究方向:缺陷检测、深度学习、推荐系统
    何昭水(1978—),男,湖南郴州 人,教授,博士,主要研究方向:智能信息处理、机器学习。
  • 基金资助:
    国家自然科学基金资助项目(61773127);2018年度国家“万人计划”科技创新领军人才;广东省基础与应用基础研究基金联合基金重点项目(2019B1515120036);广东省自然科学基金资助项目(2018A030313306);广州科学技术基金资助项目(201802010037);广东省重点领域研发计划项目(2019B010147001)

Abstract:

The traditional surface defect detection methods can only detect obvious defect contours with high contrast or low noise. In order to solve the problem, a surface defect detection method based on auto-encoding and knowledge distillation was proposed to accurately locate and classify the defects that appeared in the input images captured from the actual industrial environment. Firstly, a new Cascaded Auto-Encoder (CAE) architecture was designed to segment and locate defects, whose purpose was to convert the input original image into the CAE-based prediction mask. Secondly, the threshold module was used to binarize the prediction results, thereby obtaining the accurate defect contour. Then, the defect area extracted and cropped by the defect area detector was regarded as the input of the next module. Finally, the defect areas of the CAE segmentation results were classified by knowledge distillation. Experimental results show that, compared with other surface defect detection methods, the proposed method has the best comprehensive performance, and its average accuracy of defect detection is 97.00%. The proposed method can effectively segment the smaller defects with blurred edges, and meet the engineering requirements for real-time segmentation and detection of item surface defects.

Key words: automated surface inspection, Auto-Encoder (AE), knowledge distillation, defect detection, image processing

摘要:

针对传统的表面缺陷检测方法只能对具有高对比度或低噪声的明显缺陷轮廓进行检测的问题,提出了一种基于自编码和知识蒸馏的表面缺陷检测方法来准确定位和分类从实际工业环境捕获的输入图像中出现的缺陷。首先,设计了一种级联自动编码器(CAE)架构用于分割和定位缺陷,其目的是将输入的原始图像转换为基于CAE的预测蒙版;其次,利用阈值模块对预测结果进行二值化以获得准确的缺陷轮廓;然后,把缺陷区域检测器提取并裁剪出来的缺陷区域视为下一个模块的输入;最后,将CAE分割结果的缺陷区域通过知识蒸馏进行类别分类。实验结果表明,与其他几种表面缺陷检测方法相比,所提出的方法综合性能最好,其缺陷检测平均准确率为97.00%。该方法能够有效地对较小的、边缘不清晰的缺陷进行分割,满足对物品表面缺陷实时分割检测的工程要求。

关键词: 自动表面检测, 自编码器, 知识蒸馏, 缺陷检测, 图像处理

CLC Number: