Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector
Various surface defects in automated fiber placement (AFP) processes affect the forming quality of the components. In addition, defect detection usually requires manual observation with the naked eye, which leads to low production efficiency. Therefore, automatic solutions for defect recognition hav...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-11-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/11/22/3757 |
_version_ | 1797465513789489152 |
---|---|
author | Yongde Zhang Wei Wang Qi Liu Zhonghua Guo Yangchun Ji |
author_facet | Yongde Zhang Wei Wang Qi Liu Zhonghua Guo Yangchun Ji |
author_sort | Yongde Zhang |
collection | DOAJ |
description | Various surface defects in automated fiber placement (AFP) processes affect the forming quality of the components. In addition, defect detection usually requires manual observation with the naked eye, which leads to low production efficiency. Therefore, automatic solutions for defect recognition have high economic potential. In this paper, we propose a multi-scale AFP defect detection algorithm, named the spatial pyramid feature fusion YOLOv5 with channel attention (SPFFY-CA). The spatial pyramid feature fusion YOLOv5 (SPFFY) adopts spatial pyramid dilated convolutions (SPDCs) to fuse the feature maps extracted in different receptive fields, thus integrating multi-scale defect information. For the feature maps obtained from a concatenate function, channel attention (CA) can improve the representation ability of the network and generate more effective features. In addition, the sparsity training and pruning (STP) method is utilized to achieve network slimming, thus ensuring the efficiency and accuracy of defect detection. The experimental results of the PASCAL VOC and our AFP defect datasets demonstrate the effectiveness of our scheme, which achieves superior performance. |
first_indexed | 2024-03-09T18:22:32Z |
format | Article |
id | doaj.art-7fbfe592e9164eb88ce4b2dfdb90e7d8 |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-09T18:22:32Z |
publishDate | 2022-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-7fbfe592e9164eb88ce4b2dfdb90e7d82023-11-24T08:10:02ZengMDPI AGElectronics2079-92922022-11-011122375710.3390/electronics11223757Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale DetectorYongde Zhang0Wei Wang1Qi Liu2Zhonghua Guo3Yangchun Ji4Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, College of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaAvic Xi’an Aircraft Industry Group Company Ltd., Xi’an 710089, ChinaAvic Xi’an Aircraft Industry Group Company Ltd., Xi’an 710089, ChinaAvic Xi’an Aircraft Industry Group Company Ltd., Xi’an 710089, ChinaAvic Xi’an Aircraft Industry Group Company Ltd., Xi’an 710089, ChinaVarious surface defects in automated fiber placement (AFP) processes affect the forming quality of the components. In addition, defect detection usually requires manual observation with the naked eye, which leads to low production efficiency. Therefore, automatic solutions for defect recognition have high economic potential. In this paper, we propose a multi-scale AFP defect detection algorithm, named the spatial pyramid feature fusion YOLOv5 with channel attention (SPFFY-CA). The spatial pyramid feature fusion YOLOv5 (SPFFY) adopts spatial pyramid dilated convolutions (SPDCs) to fuse the feature maps extracted in different receptive fields, thus integrating multi-scale defect information. For the feature maps obtained from a concatenate function, channel attention (CA) can improve the representation ability of the network and generate more effective features. In addition, the sparsity training and pruning (STP) method is utilized to achieve network slimming, thus ensuring the efficiency and accuracy of defect detection. The experimental results of the PASCAL VOC and our AFP defect datasets demonstrate the effectiveness of our scheme, which achieves superior performance.https://www.mdpi.com/2079-9292/11/22/3757AFP defect detectionmulti-scale object detectionmodel compressionconvolution neural network |
spellingShingle | Yongde Zhang Wei Wang Qi Liu Zhonghua Guo Yangchun Ji Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector Electronics AFP defect detection multi-scale object detection model compression convolution neural network |
title | Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector |
title_full | Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector |
title_fullStr | Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector |
title_full_unstemmed | Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector |
title_short | Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector |
title_sort | research on defect detection in automated fiber placement processes based on a multi scale detector |
topic | AFP defect detection multi-scale object detection model compression convolution neural network |
url | https://www.mdpi.com/2079-9292/11/22/3757 |
work_keys_str_mv | AT yongdezhang researchondefectdetectioninautomatedfiberplacementprocessesbasedonamultiscaledetector AT weiwang researchondefectdetectioninautomatedfiberplacementprocessesbasedonamultiscaledetector AT qiliu researchondefectdetectioninautomatedfiberplacementprocessesbasedonamultiscaledetector AT zhonghuaguo researchondefectdetectioninautomatedfiberplacementprocessesbasedonamultiscaledetector AT yangchunji researchondefectdetectioninautomatedfiberplacementprocessesbasedonamultiscaledetector |