Skip to main navigation Skip to main content
  • E-Submission

JKSPE : Journal of the Korean Society for Precision Engineering

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS
SPECIAL

심층학습 기법을 이용한 종이용기 자동 검사

Automated Inspection for Paper Cups Using Deep Learning

Journal of the Korean Society for Precision Engineering 2017;34(7):449-453.
Published online: July 1, 2017

1 ㈜ 파이벡스 계측시스템사업부

2 ㈜ 현진제업 기술연구소

1 Department of Instrumentation, Pibex Co., Ltd.

2 Technical Laboratory, Hyun Jin Co., Ltd.

#E-mail: chpark89@naver.com, TEL: +82-54-223-9943, FAX: +82-54-223-9943
• Received: April 12, 2017   • Revised: May 11, 2017   • Accepted: May 15, 2017

Copyright © The Korean Society for Precision Engineering

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 8 Views
  • 0 Download
  • 2 Crossref
  • 2 Scopus
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • Research and Evaluation on an Optical Automatic Detection System for the Defects of the Manufactured Paper Cups
    Ping Wang, Yang-Han Lee, Hsien-Wei Tseng, Cheng-Fu Yang
    Sensors.2023; 23(3): 1452.     CrossRef
  • Method and Installation for Efficient Automatic Defect Inspection of Manufactured Paper Bowls
    Shaoyong Yu, Yang-Han Lee, Cheng-Wen Chen, Peng Gao, Zhigang Xu, Shunyi Chen, Cheng-Fu Yang
    Photonics.2023; 10(6): 686.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Automated Inspection for Paper Cups Using Deep Learning
J. Korean Soc. Precis. Eng.. 2017;34(7):449-453.   Published online July 1, 2017
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
Automated Inspection for Paper Cups Using Deep Learning
J. Korean Soc. Precis. Eng.. 2017;34(7):449-453.   Published online July 1, 2017
Close

Figure

  • 0
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
Automated Inspection for Paper Cups Using Deep Learning
Image Image Image Image Image Image Image
Fig. 1 Optical instruments of paper cups inspection with area CCD camera and diffused LED light
Fig. 2 Inner image of paper cup (Left: General lens, Right: Fish-Eye lens)
Fig. 3 Defects of paper cups (Top left: Dirty, Top right: Bad curl, Bottom left: Pollution, Bottom right: Wrinkle)
Fig. 4 Inspection procedure of conventional system
Fig. 5 The proposed inspection procedure for paper cups
Fig. 6 VGG-Like model for paper cups inspection (Conv: Convolutional layer, Fc: Fully connected layer)
Fig. 7 Model loss graph (Train data and validation)
Automated Inspection for Paper Cups Using Deep Learning

Confusion matrix for test data (Accuracy = 96.5%)

Classes Predicted
0 1 2 3 4 Precision
Actual 0 172 14 2 0 2 90.53
1 6 196 1 0 2 95.61
2 0 0 198 0 0 100
3 4 0 0 187 0 97.91
4 3 1 0 0 212 98.15
Recall 92.97 92.89 98.51 100 98.15 96.50
Table 1 Confusion matrix for test data (Accuracy = 96.5%)