dc.contributor.author | Huang, Zihao | |
dc.contributor.author | Xiao, Hong | |
dc.contributor.author | Zhang, Rongyue | |
dc.contributor.author | Wang, Hao | |
dc.contributor.author | Zhang, Cheng | |
dc.contributor.author | Shi, Xiucong | |
dc.date.accessioned | 2019-12-03T12:23:58Z | |
dc.date.available | 2019-12-03T12:23:58Z | |
dc.date.created | 2019-12-02T15:09:42Z | |
dc.date.issued | 2019 | |
dc.identifier.isbn | 978-1-7281-2980-8 | |
dc.identifier.uri | http://hdl.handle.net/11250/2631505 | |
dc.description.abstract | The traditional defect detection algorithms based on image registration, image contrast and other image processing algorithms are only limited to a single defect. Though deep-learning-based object detection algorithms can be used to detect a variety of different defects, the state-of-the-art deep-learning-based object detection algorithms still have low detection accuracy on small size defects. Basing on Cascade R-CNN in this paper, a new multi-scale feature extraction method-the Multi-Scale Feature Pair-is proposed and is used to establish a defect detection model for metal can products of an enterprise. Experimental results show that the accuracy (AP@0.5) of our improved model is 6.1% higher than Cascade R-CNN. | nb_NO |
dc.language.iso | eng | nb_NO |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | nb_NO |
dc.relation.ispartof | 2019 IEEE International Congress on Cybermatics | |
dc.title | Multi-Scale Feature Pair Based R-CNN Method for Defect Detection | nb_NO |
dc.type | Chapter | nb_NO |
dc.description.version | acceptedVersion | nb_NO |
dc.source.pagenumber | 46-51 | nb_NO |
dc.identifier.doi | https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00031 | |
dc.identifier.cristin | 1755558 | |
dc.description.localcode | © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. | nb_NO |
cristin.unitcode | 194,63,10,0 | |
cristin.unitname | Institutt for datateknologi og informatikk | |
cristin.ispublished | true | |
cristin.fulltext | postprint | |