WFF-Net: Trainable weight feature fusion convolutional neural networks for surface defect detection

被引:6
作者
Xiao, Hongyong [1 ]
Zhang, Wenying [1 ]
Zuo, Lei [1 ]
Wen, Long [1 ,2 ,4 ]
Li, Qingzhe [1 ]
Li, Xinyu [3 ]
机构
[1] China Univ Geosci, Sch Mech Engn & Elect Informat, 388 LuMo Rd, Wuhan 430074, Peoples R China
[2] China Univ Geosci, Shenzhen Res Inst, Shenzhen 518057, Peoples R China
[3] Huazhong Univ Sci & Technol, State Key Lab Digital Mfg Equipment & Technol, 1037 Luoyu Rd, Wuhan 430074, Peoples R China
[4] Chongqing Univ, State Key Lab Mech Transmiss, Chongqing 400023, Peoples R China
关键词
Feature fusion; Attention mechanism; Pixel-level segmentation; Surface defect detection;
D O I
10.1016/j.aei.2024.103073
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Deep learning-based surface defect segmentation (SDS) technique is widely used in the field of surface defect detection (SDD) for its high accuracy and robustness. However, the deep learning-based surface defect segmentation method suffers from the interference from the semantic difference between the high-dimensional features and the low-dimensional features during the process of multi-scale feature fusion, which will bring additional noise to the network and thus affect the detection accuracy. For this drawback, this paper investigates a new weight-based feature fusion method, which aims to reduce the semantic differences and information redundancy after multiscale feature fusion in the process of coupling high-dimensional semantic features with low-dimensional semantic features. First, the WFF feature fusion method is proposed in the multi-scale feature fusion stage, which uses a learnable Gate module to assign weight coefficients to neighboring features, and uses an attention mechanism to fuse weighted neighboring features, so that redundant information can be reduced both before neighboring features are coupled. It can also reduce the semantic differences between multi-highdimensional features and low-dimensional features after fusion. Second, a dual decoding module is constructed to reduce the feature loss in the decoding stage, and a structural loss function is designed to optimize the network for the multi-scale output in the dual decoder. The proposed WFF-Net has been conducted on three datasets, and it shows that the proposed WFF-Net outperforms several existing DL methods in mean intersection of union (NEU-SEG: 85.70%, DAGM 2007: 86.12%, MT defects: 82.72%) and F1-measure (NEU-SEG: 94.11 %, DAGM 2007: 96.32 %, MT defects: 93.90 %).
引用
收藏
页数:13
相关论文
共 50 条
[41]   DFSDNet: A dual-branch multi-scale feature fusion network for surface defect detection of copper strips and plates [J].
Wan, Fajia ;
Zhang, Guo ;
Li, Zeteng .
COMPUTERS IN INDUSTRY, 2025, 167
[42]   Feature matching driven background generalization neural networks for surface defect segmentation [J].
Chen, Biao ;
Niu, Tongzhi ;
Zhang, Ruoqi ;
Zhang, Hang ;
Lin, Yuchen ;
Li, Bin .
KNOWLEDGE-BASED SYSTEMS, 2024, 287
[43]   Integrating Weighted Feature Fusion and the Spatial Attention Module with Convolutional Neural Networks for Automatic Aircraft Detection from SAR Images [J].
Wang, Jielan ;
Xiao, Hongguang ;
Chen, Lifu ;
Xing, Jin ;
Pan, Zhouhao ;
Luo, Ru ;
Cai, Xingmin .
REMOTE SENSING, 2021, 13 (05) :1-21
[44]   TAFFNet: Two-Stage Attention-Based Feature Fusion Network for Surface Defect Detection [J].
Cao, Jingang ;
Yang, Guotian ;
Yang, Xiyun .
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 2022, 94 (12) :1531-1544
[45]   FDFNet: An efficient detection network for small-size surface defect based on feature differentiated fusion [J].
Liu, Jiajian ;
Zhang, Zhipeng ;
Alam, M. D. Kawsar ;
Cai, Qing ;
Xia, Chengyi ;
Tang, Youhong .
DIGITAL SIGNAL PROCESSING, 2025, 167
[46]   TAFFNet: Two-Stage Attention-Based Feature Fusion Network for Surface Defect Detection [J].
Jingang Cao ;
Guotian Yang ;
Xiyun Yang .
Journal of Signal Processing Systems, 2022, 94 :1531-1544
[47]   Multi-Scale Feature Fusion Convolutional Neural Networks for Fault Diagnosis of Electromechanical Actuator [J].
Song, Yutong ;
Du, Jinhua ;
Li, Shixiao ;
Long, Yun ;
Liang, Deliang ;
Liu, Yifeng ;
Wang, Yao .
APPLIED SCIENCES-BASEL, 2023, 13 (15)
[48]   Malware Classification Using Convolutional Fuzzy Neural Networks Based on Feature Fusion and the Taguchi Method [J].
Lin, Cheng-Jian ;
Huang, Min-Su ;
Lee, Chin-Ling .
APPLIED SCIENCES-BASEL, 2022, 12 (24)
[49]   A light weight multi-scale feature fusion steel surface defect detection model based on YOLOv8 [J].
Xie, Weining ;
Sun, Xiaoyong ;
Ma, Weifeng .
MEASUREMENT SCIENCE AND TECHNOLOGY, 2024, 35 (05)
[50]   Multilayer Feature Fusion With Weight Adjustment Based on a Convolutional Neural Network for Remote Sensing Scene Classification [J].
Ma, Chenhui ;
Mu, Xiaodong ;
Lin, Renpu ;
Wang, Shuyang .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2021, 18 (02) :241-245