CASIA OpenIR  > 精密感知与控制研究中心  > 精密感知与控制
Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks
Tao Xian; Zhang Dapeng; Ma Wenzhi; Liu Xilong; Xu De; Xian Tao
Source PublicationAPPLIED SCIENCES-BASEL
ISSN2076-3417
2018-09-01
Volume8Issue:9Pages:15
Corresponding AuthorTao, Xian(taoxian2013@ia.ac.cn)
AbstractAutomatic metallic surface defect inspection has received increased attention in relation to the quality control of industrial products. Metallic defect detection is usually performed against complex industrial scenarios, presenting an interesting but challenging problem. Traditional methods are based on image processing or shallow machine learning techniques, but these can only detect defects under specific detection conditions, such as obvious defect contours with strong contrast and low noise, at certain scales, or under specific illumination conditions. This paper discusses the automatic detection of metallic defects with a twofold procedure that accurately localizes and classifies defects appearing in input images captured from real industrial environments. A novel cascaded autoencoder (CASAE) architecture is designed for segmenting and localizing defects. The cascading network transforms the input defect image into a pixel-wise prediction mask based on semantic segmentation. The defect regions of segmented results are classified into their specific classes via a compact convolutional neural network (CNN). Metallic defects under various conditions can be successfully detected using an industrial dataset. The experimental results demonstrate that this method meets the robustness and accuracy requirements for metallic defect detection. Meanwhile, it can also be extended to other detection applications.
Keywordmetallic surface autoencoder convolutional neural network defect detection
DOI10.3390/app8091575
WOS KeywordFAULT-DIAGNOSIS ; INSPECTION ; CLASSIFICATION ; IMAGES ; MODEL
Indexed BySCI
Language英语
Funding ProjectScience Challenge Project[TZ2018006-0204-02] ; National Natural Science Foundation of China[61703399] ; National Natural Science Foundation of China[61503376] ; National Natural Science Foundation of China[61673383]
Funding OrganizationScience Challenge Project ; National Natural Science Foundation of China
WOS Research AreaChemistry ; Materials Science ; Physics
WOS SubjectChemistry, Multidisciplinary ; Materials Science, Multidisciplinary ; Physics, Applied
WOS IDWOS:000445760200164
PublisherMDPI
Citation statistics
Cited Times:7[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/21696
Collection精密感知与控制研究中心_精密感知与控制
Corresponding AuthorXian Tao
AffiliationResearch Center of Precision Sensing and Control, Institute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Tao Xian,Zhang Dapeng,Ma Wenzhi,et al. Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks[J]. APPLIED SCIENCES-BASEL,2018,8(9):15.
APA Tao Xian,Zhang Dapeng,Ma Wenzhi,Liu Xilong,Xu De,&Xian Tao.(2018).Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks.APPLIED SCIENCES-BASEL,8(9),15.
MLA Tao Xian,et al."Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks".APPLIED SCIENCES-BASEL 8.9(2018):15.
Files in This Item: Download All
File Name/Size DocType Version Access License
applsci-08-01575.pdf(3978KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Tao Xian]'s Articles
[Zhang Dapeng]'s Articles
[Ma Wenzhi]'s Articles
Baidu academic
Similar articles in Baidu academic
[Tao Xian]'s Articles
[Zhang Dapeng]'s Articles
[Ma Wenzhi]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Tao Xian]'s Articles
[Zhang Dapeng]'s Articles
[Ma Wenzhi]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: applsci-08-01575.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.