Automatic Fabric Defect Detection Method Using PRAN-Net
Fabric defect detection is very important in the textile quality process. Current deep learning algorithms are not effective in detecting tiny and extreme aspect ratio fabric defects. In this paper, we proposed a strong detection method, Priori Anchor Convolutional Neural Network (PRAN-Net), for fab...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-11-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/10/23/8434 |
_version_ | 1797546610568200192 |
---|---|
author | Peiran Peng Ying Wang Can Hao Zhizhong Zhu Tong Liu Weihu Zhou |
author_facet | Peiran Peng Ying Wang Can Hao Zhizhong Zhu Tong Liu Weihu Zhou |
author_sort | Peiran Peng |
collection | DOAJ |
description | Fabric defect detection is very important in the textile quality process. Current deep learning algorithms are not effective in detecting tiny and extreme aspect ratio fabric defects. In this paper, we proposed a strong detection method, Priori Anchor Convolutional Neural Network (PRAN-Net), for fabric defect detection to improve the detection and location accuracy of fabric defects and decrease the inspection time. First, we used Feature Pyramid Network (FPN) by selected multi-scale feature maps to reserve more detailed information of tiny defects. Secondly, we proposed a trick to generate sparse priori anchors based on fabric defects ground truth boxes instead of fixed anchors to locate extreme defects more accurately and efficiently. Finally, a classification network is used to classify and refine the position of the fabric defects. The method was validated on two self-made fabric datasets. Experimental results indicate that our method significantly improved the accuracy and efficiency of detecting fabric defects and is more suitable to the automatic fabric defect detection. |
first_indexed | 2024-03-10T14:33:06Z |
format | Article |
id | doaj.art-7d9b774a73894994b5afd6f1b16e4d36 |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-10T14:33:06Z |
publishDate | 2020-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-7d9b774a73894994b5afd6f1b16e4d362023-11-20T22:27:07ZengMDPI AGApplied Sciences2076-34172020-11-011023843410.3390/app10238434Automatic Fabric Defect Detection Method Using PRAN-NetPeiran Peng0Ying Wang1Can Hao2Zhizhong Zhu3Tong Liu4Weihu Zhou5Institute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaInstitute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaInstitute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaInstitute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaSchool of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing 100083, ChinaInstitute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaFabric defect detection is very important in the textile quality process. Current deep learning algorithms are not effective in detecting tiny and extreme aspect ratio fabric defects. In this paper, we proposed a strong detection method, Priori Anchor Convolutional Neural Network (PRAN-Net), for fabric defect detection to improve the detection and location accuracy of fabric defects and decrease the inspection time. First, we used Feature Pyramid Network (FPN) by selected multi-scale feature maps to reserve more detailed information of tiny defects. Secondly, we proposed a trick to generate sparse priori anchors based on fabric defects ground truth boxes instead of fixed anchors to locate extreme defects more accurately and efficiently. Finally, a classification network is used to classify and refine the position of the fabric defects. The method was validated on two self-made fabric datasets. Experimental results indicate that our method significantly improved the accuracy and efficiency of detecting fabric defects and is more suitable to the automatic fabric defect detection.https://www.mdpi.com/2076-3417/10/23/8434fabric defect detectiondeep learningFaster R-CNNextreme and tiny defectspriori anchor |
spellingShingle | Peiran Peng Ying Wang Can Hao Zhizhong Zhu Tong Liu Weihu Zhou Automatic Fabric Defect Detection Method Using PRAN-Net Applied Sciences fabric defect detection deep learning Faster R-CNN extreme and tiny defects priori anchor |
title | Automatic Fabric Defect Detection Method Using PRAN-Net |
title_full | Automatic Fabric Defect Detection Method Using PRAN-Net |
title_fullStr | Automatic Fabric Defect Detection Method Using PRAN-Net |
title_full_unstemmed | Automatic Fabric Defect Detection Method Using PRAN-Net |
title_short | Automatic Fabric Defect Detection Method Using PRAN-Net |
title_sort | automatic fabric defect detection method using pran net |
topic | fabric defect detection deep learning Faster R-CNN extreme and tiny defects priori anchor |
url | https://www.mdpi.com/2076-3417/10/23/8434 |
work_keys_str_mv | AT peiranpeng automaticfabricdefectdetectionmethodusingprannet AT yingwang automaticfabricdefectdetectionmethodusingprannet AT canhao automaticfabricdefectdetectionmethodusingprannet AT zhizhongzhu automaticfabricdefectdetectionmethodusingprannet AT tongliu automaticfabricdefectdetectionmethodusingprannet AT weihuzhou automaticfabricdefectdetectionmethodusingprannet |