A Design and Training Application for Deep Convolutional Neural Networks and Support Vector Machines Developed on MATLAB

Fusaomi Nagata; Kenta Tokuno; Hiroaki Ochi; Akimasa Otsuka; Takeshi Ikeda; Keigo Watanabe; Maki K. Habib

doi:10.1007/978-981-13-8323-6_3

A Design and Training Application for Deep Convolutional Neural Networks and Support Vector Machines Developed on MATLAB

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

Research output: Chapter in Book/Report/Conference proceeding › Chapter

1 Citation (Scopus)

Abstract

This paper presents a user-friendly design application development environment based on MATLAB that facilitates two applications using convolutional neural networks (CNNs) and support vector machines (SVMs). Firstly, an application of deep CNN (DCNN) for visual inspection is developed and is trained using a large number of images to inspect undesirable defects such as crack, burr, protrusion, chipping, spot and fracture phenomena which appear in the manufacturing process of resin molded articles. The DCNN is named sssNet. Then, two kinds of SVMs are respectively incorporated with two trained DCNNs, i.e., our designed sssNet and well-known AlexNet, to classify test images with high recognition rate into accept as OK or reject as NG categories, in which compressed features obtained from the DCNNs are used as inputs for the SVMs. The usability and operability of the developed design and training application for DCNNs and SVMs are demonstrated and evaluated through training and classification experiments.

Original language	English
Title of host publication	Lecture Notes in Mechanical Engineering
Publisher	Pleiades Publishing
Pages	27-33
Number of pages	7
DOIs	https://doi.org/10.1007/978-981-13-8323-6_3
Publication status	Published - Jan 1 2020

Publication series

Name	Lecture Notes in Mechanical Engineering
ISSN (Print)	2195-4356
ISSN (Electronic)	2195-4364

Keywords

Convolutional neural network (CNN)
Design tool for CNN and SVM
MATLAB
Support vector machine (SVM)
Visual inspection

ASJC Scopus subject areas

Automotive Engineering
Aerospace Engineering
Mechanical Engineering
Fluid Flow and Transfer Processes

Access to Document

10.1007/978-981-13-8323-6_3

Cite this

Nagata, F., Tokuno, K., Ochi, H., Otsuka, A., Ikeda, T., Watanabe, K., & Habib, M. K. (2020). A Design and Training Application for Deep Convolutional Neural Networks and Support Vector Machines Developed on MATLAB. In Lecture Notes in Mechanical Engineering (pp. 27-33). (Lecture Notes in Mechanical Engineering). Pleiades Publishing. https://doi.org/10.1007/978-981-13-8323-6_3

@inbook{73bfcb923ff443338da19463762ba4b2,

title = "A Design and Training Application for Deep Convolutional Neural Networks and Support Vector Machines Developed on MATLAB",

abstract = "This paper presents a user-friendly design application development environment based on MATLAB that facilitates two applications using convolutional neural networks (CNNs) and support vector machines (SVMs). Firstly, an application of deep CNN (DCNN) for visual inspection is developed and is trained using a large number of images to inspect undesirable defects such as crack, burr, protrusion, chipping, spot and fracture phenomena which appear in the manufacturing process of resin molded articles. The DCNN is named sssNet. Then, two kinds of SVMs are respectively incorporated with two trained DCNNs, i.e., our designed sssNet and well-known AlexNet, to classify test images with high recognition rate into accept as OK or reject as NG categories, in which compressed features obtained from the DCNNs are used as inputs for the SVMs. The usability and operability of the developed design and training application for DCNNs and SVMs are demonstrated and evaluated through training and classification experiments.",

keywords = "Convolutional neural network (CNN), Design tool for CNN and SVM, MATLAB, Support vector machine (SVM), Visual inspection",

author = "Fusaomi Nagata and Kenta Tokuno and Hiroaki Ochi and Akimasa Otsuka and Takeshi Ikeda and Keigo Watanabe and Habib, {Maki K.}",

year = "2020",

month = jan,

day = "1",

doi = "10.1007/978-981-13-8323-6_3",

language = "English",

series = "Lecture Notes in Mechanical Engineering",

publisher = "Pleiades Publishing",

pages = "27--33",

booktitle = "Lecture Notes in Mechanical Engineering",

}

TY - CHAP

T1 - A Design and Training Application for Deep Convolutional Neural Networks and Support Vector Machines Developed on MATLAB

AU - Nagata, Fusaomi

AU - Tokuno, Kenta

AU - Ochi, Hiroaki

AU - Otsuka, Akimasa

AU - Ikeda, Takeshi

AU - Watanabe, Keigo

AU - Habib, Maki K.

PY - 2020/1/1

Y1 - 2020/1/1

N2 - This paper presents a user-friendly design application development environment based on MATLAB that facilitates two applications using convolutional neural networks (CNNs) and support vector machines (SVMs). Firstly, an application of deep CNN (DCNN) for visual inspection is developed and is trained using a large number of images to inspect undesirable defects such as crack, burr, protrusion, chipping, spot and fracture phenomena which appear in the manufacturing process of resin molded articles. The DCNN is named sssNet. Then, two kinds of SVMs are respectively incorporated with two trained DCNNs, i.e., our designed sssNet and well-known AlexNet, to classify test images with high recognition rate into accept as OK or reject as NG categories, in which compressed features obtained from the DCNNs are used as inputs for the SVMs. The usability and operability of the developed design and training application for DCNNs and SVMs are demonstrated and evaluated through training and classification experiments.

AB - This paper presents a user-friendly design application development environment based on MATLAB that facilitates two applications using convolutional neural networks (CNNs) and support vector machines (SVMs). Firstly, an application of deep CNN (DCNN) for visual inspection is developed and is trained using a large number of images to inspect undesirable defects such as crack, burr, protrusion, chipping, spot and fracture phenomena which appear in the manufacturing process of resin molded articles. The DCNN is named sssNet. Then, two kinds of SVMs are respectively incorporated with two trained DCNNs, i.e., our designed sssNet and well-known AlexNet, to classify test images with high recognition rate into accept as OK or reject as NG categories, in which compressed features obtained from the DCNNs are used as inputs for the SVMs. The usability and operability of the developed design and training application for DCNNs and SVMs are demonstrated and evaluated through training and classification experiments.

KW - Convolutional neural network (CNN)

KW - Design tool for CNN and SVM

KW - MATLAB

KW - Support vector machine (SVM)

KW - Visual inspection

UR - http://www.scopus.com/inward/record.url?scp=85067662000&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067662000&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-8323-6_3

DO - 10.1007/978-981-13-8323-6_3

M3 - Chapter

AN - SCOPUS:85067662000

T3 - Lecture Notes in Mechanical Engineering

SP - 27

EP - 33

BT - Lecture Notes in Mechanical Engineering

PB - Pleiades Publishing

ER -

A Design and Training Application for Deep Convolutional Neural Networks and Support Vector Machines Developed on MATLAB

Abstract

Publication series

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this