TY - GEN
T1 - Design Application of Deep Convolutional Neural Network for Vision-Based Defect Inspection
AU - Nagata, Fusaomi
AU - Tokuno, Kenta
AU - Watanabe, Keigo
AU - Habib, Maki K.
N1 - Funding Information:
VI. CONCLUSIONS In this paper, a user-friendly DCNN design application is implemented and tested. The application does not require the technical knowledge about programming languages such as Python, C++, Visual Studio, etc. Three types of DCNNs for two, five and six classifications are experimentally designed and trained using the application to inspect small defects such as crack, burr, protrusion, chipping and spot, so that the usefulness and effectiveness of the developed application are confirmed. An additional training method to cope with not well trained samples is shown and evaluated, so that the classification ability can be efficiently and pinpointedly improved to a desired level of categorization accuracy. The proposed DCNN design application is planned to be applied to actual physical inspection processes. ACKNOWLEDGMENT This work was supported by JSPS KAKENHI Grant Number 16K06203.
PY - 2019/1/16
Y1 - 2019/1/16
N2 - In this decade, deep convolutional neural network called DCNN has been attracting attention due to its high ability of image Recognition and other applications. In this paper, a design application of DCNN is considered and developed for vision-based defect inspection. As a trial test, three kinds of DCNNs are designed, implemented and tested to inspect small defects, such as, crack, burr, protrusion, chipping and spot phenomena seen in the manufacturing process of resin molded articles. An image generator is also implemented to systematically generate range of relevant deformed version of similar images for training. The designed DCNNs are trained using the generated images and then evaluated through classification experiments. The usefulness of the proposed DCNN design application is demonstrated and discussed.
AB - In this decade, deep convolutional neural network called DCNN has been attracting attention due to its high ability of image Recognition and other applications. In this paper, a design application of DCNN is considered and developed for vision-based defect inspection. As a trial test, three kinds of DCNNs are designed, implemented and tested to inspect small defects, such as, crack, burr, protrusion, chipping and spot phenomena seen in the manufacturing process of resin molded articles. An image generator is also implemented to systematically generate range of relevant deformed version of similar images for training. The designed DCNNs are trained using the generated images and then evaluated through classification experiments. The usefulness of the proposed DCNN design application is demonstrated and discussed.
KW - additional training
KW - deep convolutional neural network
KW - design application
KW - similar image generator
KW - vision-based defect inspection
UR - http://www.scopus.com/inward/record.url?scp=85062231721&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85062231721&partnerID=8YFLogxK
U2 - 10.1109/SMC.2018.00295
DO - 10.1109/SMC.2018.00295
M3 - Conference contribution
AN - SCOPUS:85062231721
T3 - Proceedings - 2018 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2018
SP - 1705
EP - 1710
BT - Proceedings - 2018 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2018
Y2 - 7 October 2018 through 10 October 2018
ER -
