Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector

被引:3
|
作者
Zhang, Yongde [1 ,2 ]
Wang, Wei [2 ]
Liu, Qi [2 ]
Guo, Zhonghua [2 ]
Ji, Yangchun [2 ]
机构
[1] Zhejiang Univ, Coll Mech Engn, Key Lab Adv Mfg Technol Zhejiang Prov, Hangzhou 310027, Peoples R China
[2] Avic Xian Aircraft Ind Grp Co Ltd, Xian 710089, Peoples R China
关键词
AFP defect detection; multi-scale object detection; model compression; convolution neural network; CLASSIFICATION; COMPOSITES;
D O I
10.3390/electronics11223757
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
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.
引用
收藏
页数:16
相关论文
共 50 条
  • [1] Multi-scale Research Progress of Manufacturing Defects for Thermoplastic FRP Fabricated by Automated Fiber Placement
    [J]. Fu, Hong-Ya (hongyafu@hit.edu.cn), 1600, Beijing Institute of Aeronautical Materials (BIAM) (45):
  • [2] Multi-scale analysis of interface characteristics for CFRPs made by automated fiber placement
    Sun, Shouzheng
    Han, Zhenyu
    Dhupia, Jaspreet Singh
    Jin, Hongyu
    [J]. MECHANICS OF ADVANCED MATERIALS AND STRUCTURES, 2022, 29 (14) : 2106 - 2117
  • [3] Automated fabric defect detection using multi-scale fusion MemAE
    Wu, Kun
    Zhu, Lei
    Shi, Weihang
    Wang, Wenwu
    [J]. VISUAL COMPUTER, 2024,
  • [4] Research on Automated Fiber Placement Surface Defect Detection Based on Improved YOLOv7
    Wen, Liwei
    Li, Shihao
    Dong, Zhentao
    Shen, Haiqing
    Xu, Entao
    [J]. APPLIED SCIENCES-BASEL, 2024, 14 (13):
  • [5] Insulator Defect Detection Based on Multi-Scale Feature Fusion
    Bin, Li
    Luyao, Qu
    Xinshan, Zhu
    Zhimin, Guo
    Yangyang, Tian
    [J]. Diangong Jishu Xuebao/Transactions of China Electrotechnical Society, 2023, 38 (01): : 60 - 70
  • [6] Steel defect detection based on multi-scale lightweight attention
    Zhou, Yan
    Meng, Jiang-Nan
    Wang, Dong-Li
    Tan, Ya-Qin
    [J]. Kongzhi yu Juece/Control and Decision, 2024, 39 (03): : 901 - 909
  • [7] Defect Detection of Photovoltaic Modules Based on Multi-Scale Feature Fusion
    Tian, Hao
    Zhou, Qiang
    He, Chenlong
    [J]. Computer Engineering and Applications, 2024, 60 (03) : 340 - 347
  • [8] Defect detection of laminated surface in the automated fiber placement process based on improved CenterNet
    Wang, Xuan
    Kang, Shuo
    Zhu, Weidong
    [J]. Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering, 2021, 50 (10):
  • [9] Multi-Scale Low-Entropy Method for Optimizing the Processing Parameters during Automated Fiber Placement
    Han, Zhenyu
    Sun, Shouzheng
    Fu, Hongya
    Fu, Yunzhong
    [J]. MATERIALS, 2017, 10 (09):
  • [10] A Multi-Scale Detector Based on Attention Mechanism
    Zhou, Lukuan
    Wang, Wei
    Wang, Qiang
    Sheng, Biyun
    Yang, Wankou
    [J]. 2020 35TH YOUTH ACADEMIC ANNUAL CONFERENCE OF CHINESE ASSOCIATION OF AUTOMATION (YAC), 2020, : 110 - 115