Design tool of deep convolutional neural network for visual inspection

Fusaomi Nagata, Kenta Tokuno, Akimasa Otsuka, Takeshi Ikeda, Hiroaki Ochi, Hisami Tamano, Hitoshi Nakamura, Keigo Watanabe, Maki K. Habib

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this paper, a design tool for deep convolutional neural network (DCNN) is considered and developed. As a test trial, a DCNN designed by using the tool is applied to visual inspection system of resin molded articles. The defects to be inspected are crack, burr, protrusion and chipping phenomena that occur in the manufacturing process of resin molded articles. An image generator is also developed to systematically generate many similar images for training. Similar images are easily produced by rotating, translating, scaling and transforming an original image. The designed DCNN is trained using the produced images and is evaluated through classification experiments. The usefulness of the proposed design tool has been confirmed through the test trial.

Original languageEnglish
Title of host publicationData Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings
PublisherSpringer Verlag
Pages604-613
Number of pages10
ISBN (Print)9783319938028
DOIs
Publication statusPublished - Jan 1 2018
Event3rd International Conference on Data Mining and Big Data, DMBD 2018 held in conjunction with the 9th International Conference on Swarm Intelligence, ICSI 2018 - Shanghai, China
Duration: Jun 17 2018Jun 22 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10943 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other3rd International Conference on Data Mining and Big Data, DMBD 2018 held in conjunction with the 9th International Conference on Swarm Intelligence, ICSI 2018
CountryChina
CityShanghai
Period6/17/186/22/18

Fingerprint

Inspection
Neural Networks
Neural networks
Resins
Cracks
Defects
Rotating
Crack
Manufacturing
Vision
Design
Scaling
Generator
Experiments
Experiment

Keywords

  • DCNN design tool
  • Deep convolutional neural network (DCNN)
  • MATLAB
  • Training image generator
  • Visual inspection system

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Nagata, F., Tokuno, K., Otsuka, A., Ikeda, T., Ochi, H., Tamano, H., ... Habib, M. K. (2018). Design tool of deep convolutional neural network for visual inspection. In Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings (pp. 604-613). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10943 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-93803-5_57

Design tool of deep convolutional neural network for visual inspection. / Nagata, Fusaomi; Tokuno, Kenta; Otsuka, Akimasa; Ikeda, Takeshi; Ochi, Hiroaki; Tamano, Hisami; Nakamura, Hitoshi; Watanabe, Keigo; Habib, Maki K.

Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings. Springer Verlag, 2018. p. 604-613 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10943 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nagata, F, Tokuno, K, Otsuka, A, Ikeda, T, Ochi, H, Tamano, H, Nakamura, H, Watanabe, K & Habib, MK 2018, Design tool of deep convolutional neural network for visual inspection. in Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10943 LNCS, Springer Verlag, pp. 604-613, 3rd International Conference on Data Mining and Big Data, DMBD 2018 held in conjunction with the 9th International Conference on Swarm Intelligence, ICSI 2018, Shanghai, China, 6/17/18. https://doi.org/10.1007/978-3-319-93803-5_57
Nagata F, Tokuno K, Otsuka A, Ikeda T, Ochi H, Tamano H et al. Design tool of deep convolutional neural network for visual inspection. In Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings. Springer Verlag. 2018. p. 604-613. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-93803-5_57
Nagata, Fusaomi ; Tokuno, Kenta ; Otsuka, Akimasa ; Ikeda, Takeshi ; Ochi, Hiroaki ; Tamano, Hisami ; Nakamura, Hitoshi ; Watanabe, Keigo ; Habib, Maki K. / Design tool of deep convolutional neural network for visual inspection. Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings. Springer Verlag, 2018. pp. 604-613 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{c84fab2951e748159e425177d7c087fe,
title = "Design tool of deep convolutional neural network for visual inspection",
abstract = "In this paper, a design tool for deep convolutional neural network (DCNN) is considered and developed. As a test trial, a DCNN designed by using the tool is applied to visual inspection system of resin molded articles. The defects to be inspected are crack, burr, protrusion and chipping phenomena that occur in the manufacturing process of resin molded articles. An image generator is also developed to systematically generate many similar images for training. Similar images are easily produced by rotating, translating, scaling and transforming an original image. The designed DCNN is trained using the produced images and is evaluated through classification experiments. The usefulness of the proposed design tool has been confirmed through the test trial.",
keywords = "DCNN design tool, Deep convolutional neural network (DCNN), MATLAB, Training image generator, Visual inspection system",
author = "Fusaomi Nagata and Kenta Tokuno and Akimasa Otsuka and Takeshi Ikeda and Hiroaki Ochi and Hisami Tamano and Hitoshi Nakamura and Keigo Watanabe and Habib, {Maki K.}",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-319-93803-5_57",
language = "English",
isbn = "9783319938028",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "604--613",
booktitle = "Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings",

}

TY - GEN

T1 - Design tool of deep convolutional neural network for visual inspection

AU - Nagata, Fusaomi

AU - Tokuno, Kenta

AU - Otsuka, Akimasa

AU - Ikeda, Takeshi

AU - Ochi, Hiroaki

AU - Tamano, Hisami

AU - Nakamura, Hitoshi

AU - Watanabe, Keigo

AU - Habib, Maki K.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In this paper, a design tool for deep convolutional neural network (DCNN) is considered and developed. As a test trial, a DCNN designed by using the tool is applied to visual inspection system of resin molded articles. The defects to be inspected are crack, burr, protrusion and chipping phenomena that occur in the manufacturing process of resin molded articles. An image generator is also developed to systematically generate many similar images for training. Similar images are easily produced by rotating, translating, scaling and transforming an original image. The designed DCNN is trained using the produced images and is evaluated through classification experiments. The usefulness of the proposed design tool has been confirmed through the test trial.

AB - In this paper, a design tool for deep convolutional neural network (DCNN) is considered and developed. As a test trial, a DCNN designed by using the tool is applied to visual inspection system of resin molded articles. The defects to be inspected are crack, burr, protrusion and chipping phenomena that occur in the manufacturing process of resin molded articles. An image generator is also developed to systematically generate many similar images for training. Similar images are easily produced by rotating, translating, scaling and transforming an original image. The designed DCNN is trained using the produced images and is evaluated through classification experiments. The usefulness of the proposed design tool has been confirmed through the test trial.

KW - DCNN design tool

KW - Deep convolutional neural network (DCNN)

KW - MATLAB

KW - Training image generator

KW - Visual inspection system

UR - http://www.scopus.com/inward/record.url?scp=85049007628&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049007628&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-93803-5_57

DO - 10.1007/978-3-319-93803-5_57

M3 - Conference contribution

AN - SCOPUS:85049007628

SN - 9783319938028

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 604

EP - 613

BT - Data Mining and Big Data - 3rd International Conference, DMBD 2018, Proceedings

PB - Springer Verlag

ER -