Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks
This study aims to develop a novel automated computer vision algorithm for quality inspection of surfaces with complex patterns. The proposed algorithm is based on both an autoencoder (AE) and a fully convolutional neural network (FCN). The AE is adopted for the self-generation of templates from tes...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-08-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/11/17/7838 |
_version_ | 1797521714221940736 |
---|---|
author | Cheng-Wei Lei Li Zhang Tsung-Ming Tai Chen-Chieh Tsai Wen-Jyi Hwang Yun-Jie Jhang |
author_facet | Cheng-Wei Lei Li Zhang Tsung-Ming Tai Chen-Chieh Tsai Wen-Jyi Hwang Yun-Jie Jhang |
author_sort | Cheng-Wei Lei |
collection | DOAJ |
description | This study aims to develop a novel automated computer vision algorithm for quality inspection of surfaces with complex patterns. The proposed algorithm is based on both an autoencoder (AE) and a fully convolutional neural network (FCN). The AE is adopted for the self-generation of templates from test targets for defect detection. Because the templates are produced from the test targets, the position alignment issues for the matching operations between templates and test targets can be alleviated. The FCN is employed for the segmentation of a template into a number of coherent regions. Because the AE has the limitation that its capacities for the regeneration of each coherent region in the template may be different, the segmentation of the template by FCN is beneficial for allowing the inspection of each region to be independently carried out. In this way, more accurate detection results can be achieved. Experimental results reveal that the proposed algorithm has the advantages of simplicity for training data collection, high accuracy for defect detection, and high flexibility for online inspection. The proposed algorithm is therefore an effective alternative for the automated inspection in smart factories with a growing demand for the reliability for high quality production. |
first_indexed | 2024-03-10T08:16:21Z |
format | Article |
id | doaj.art-44ca5701faca4cfb8e56266fb926d06d |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-10T08:16:21Z |
publishDate | 2021-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-44ca5701faca4cfb8e56266fb926d06d2023-11-22T10:17:15ZengMDPI AGApplied Sciences2076-34172021-08-011117783810.3390/app11177838Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural NetworksCheng-Wei Lei0Li Zhang1Tsung-Ming Tai2Chen-Chieh Tsai3Wen-Jyi Hwang4Yun-Jie Jhang5Department of Computer Science and Information Engineering, National Taiwan Normal University, Taipei 117, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan Normal University, Taipei 117, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan Normal University, Taipei 117, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan Normal University, Taipei 117, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan Normal University, Taipei 117, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan Normal University, Taipei 117, TaiwanThis study aims to develop a novel automated computer vision algorithm for quality inspection of surfaces with complex patterns. The proposed algorithm is based on both an autoencoder (AE) and a fully convolutional neural network (FCN). The AE is adopted for the self-generation of templates from test targets for defect detection. Because the templates are produced from the test targets, the position alignment issues for the matching operations between templates and test targets can be alleviated. The FCN is employed for the segmentation of a template into a number of coherent regions. Because the AE has the limitation that its capacities for the regeneration of each coherent region in the template may be different, the segmentation of the template by FCN is beneficial for allowing the inspection of each region to be independently carried out. In this way, more accurate detection results can be achieved. Experimental results reveal that the proposed algorithm has the advantages of simplicity for training data collection, high accuracy for defect detection, and high flexibility for online inspection. The proposed algorithm is therefore an effective alternative for the automated inspection in smart factories with a growing demand for the reliability for high quality production.https://www.mdpi.com/2076-3417/11/17/7838surface inspectiondefect detectionartificial intelligenceautoencoderconvolutional neural networks |
spellingShingle | Cheng-Wei Lei Li Zhang Tsung-Ming Tai Chen-Chieh Tsai Wen-Jyi Hwang Yun-Jie Jhang Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks Applied Sciences surface inspection defect detection artificial intelligence autoencoder convolutional neural networks |
title | Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks |
title_full | Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks |
title_fullStr | Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks |
title_full_unstemmed | Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks |
title_short | Automated Surface Defect Inspection Based on Autoencoders and Fully Convolutional Neural Networks |
title_sort | automated surface defect inspection based on autoencoders and fully convolutional neural networks |
topic | surface inspection defect detection artificial intelligence autoencoder convolutional neural networks |
url | https://www.mdpi.com/2076-3417/11/17/7838 |
work_keys_str_mv | AT chengweilei automatedsurfacedefectinspectionbasedonautoencodersandfullyconvolutionalneuralnetworks AT lizhang automatedsurfacedefectinspectionbasedonautoencodersandfullyconvolutionalneuralnetworks AT tsungmingtai automatedsurfacedefectinspectionbasedonautoencodersandfullyconvolutionalneuralnetworks AT chenchiehtsai automatedsurfacedefectinspectionbasedonautoencodersandfullyconvolutionalneuralnetworks AT wenjyihwang automatedsurfacedefectinspectionbasedonautoencodersandfullyconvolutionalneuralnetworks AT yunjiejhang automatedsurfacedefectinspectionbasedonautoencodersandfullyconvolutionalneuralnetworks |