With the growing demand for efficient and high-precision quality control in textiles, the limitations of traditional fabric defect detection methods have become increasingly apparent, particularly in handling complex backgrounds, diverse defect types, and small-object detection. To address these challenges, this study proposes a fabric defect detection method based on morphological prior features. By deeply integrating morphological prior information into the feature extraction and fusion modules, the robustness and accuracy of defect detection are improved significantly. Specifically, the divergent path feature enhancement module enhances the detection of small-object defects through fine-grained feature extraction. The oriented space and multiscale feature extraction module combines multiscale and directional feature extraction techniques to improve the recognition of defects with extreme aspect ratios. Furthermore, the efficient multipath feature fusion network achieves comprehensive capture of defect features by integrating shallow and deep features. In addition, the proposed fabric mosaic data augmentation strategy dynamically adjusts the cropping offset points, effectively preserving the pixel and feature integrity of target defects under conditions of data scarcity and imbalance. Experimental results demonstrate that the proposed model achieves a precision of 75.5%, a recall of 75.8%, and an F1 score of 75.6%. The mAP@.5 is improved by 9.8% compared with the baseline model. The proposed approach strikes a favorable balance between detection accuracy, computational complexity, and inference speed, showcasing excellent generalization capability and practical application potential.
WeiBHaoKGaoL, et al.
Detecting textile micro-defects: A novel and efficient method based on visual gain mechanism. Inform Sci2020;
541: 60–74. https://doi.org/10.1016/j.ins.2020.06.035
5.
WeiWDengDZengL, et al.
Real-time implementation of fabric defect detection based on variational automatic encoder with structure similarity. J Real-Time Image Process2021;
18(3): 807–823. https://doi.org/10.1007/s11554-020-01023-5
HuFGongJFuH, et al.
Fabric defect detection method using SA-Pix2pix network and transfer learning. Appl Sci2023;
14(1): 41. https://doi.org/10.3390/app14010041
8.
WeiCLiangJLiuH, et al.
Multi-stage unsupervised fabric defect detection based on DCGAN. Visual Comput2023;
39(12): 6655–6671. https://doi.org/10.1007/s00371-022-02754-1
9.
MohammeSSClarkeHG.Conditional image-to-image translation generative adversarial network (cGAN) for fabric defect data augmentation. Neur Comput Applicat2024; 1–14. https://doi.org/10.1007/s00521-024-10179-1
10.
LinGLiuKXiaX, et al.
An efficient and intelligent detection method for fabric defects based on improved YOLOv5. Sensors2022;
23(1): 97. https://doi.org/10.3390/s23010097
11.
ZhangCChangCJamshidiM.Concrete bridge surface damage detection using a single‐stage detector. Computer‐Aided Civ Infrastruct Eng2020;
35(4): 389–409. https://doi.org/10.1111/mice.12500
12.
ChenMYuLZhiC, et al.
Improved faster R-CNN for fabric defect detection based on Gabor filter with Genetic Algorithm optimization. Comput Ind2022;
134: 103551. https://doi.org/10.1016/j.compind.2021.103551
13.
SongPChenKZhuL, et al.
An improved cascade R-CNN and RGB-D camera-based method for dynamic cotton top bud recognition and localization in the field. Comput Electron Agric2022;
202: 107442. https://doi.org/10.1016/j.compag.2022.107442
14.
JiaoLDongSZhangS, et al.
AF-RCNN: An anchor-free convolutional neural network for multi-categories agricultural pest detection. Comput Electron Agric2020;
174: 105522. https://doi.org/10.1016/j.compag.2020.105522
15.
LiJKangX.Mobile-YOLO: An accurate and efficient three-stage cascaded network for online fiberglass fabric defect detection. Eng Applicat Artif Intell2024;
134, 108690. https://doi.org/10.1016/j.engappai.2024.108690
16.
ZhangYZhangHHuangQ, et al.
DsP-YOLO: An anchor-free network with DsPAN for small object detection of multiscale defects. Expert Syst Applicat2024;
241, 122669. https://doi.org/10.1016/j.eswa.2023.122669
LiuBWangHCaoZ, et al.
PRC-Light YOLO: An efficient lightweight model for fabric defect detection. Appl Sci2024;
14(2): 938. https://doi.org/10.3390/app14020938
20.
ChenCZhouQXiaoL, et al.
LWFDD-YOLO: A lightweight defect detection algorithm based on improved YOLOv8. Text Res J2024; 911482844. https://doi.org/10.1177/00405175241285596
21.
MaSLiuYZhangY.Fabric defect detection based on improved lightweight YOLOv8n. Appl Sci2024;
14(17): 8000. https://doi.org/10.3390/app14178000
YaohuiZJiaRYuL.Yolov7-Tinier: Towards high-precision and lightweight detection of fabric defects in textile plant. Fibers Polym2024; 1–14. https://doi.org/10.1007/s12221-024-00662-w
24.
LuBHuangB.A texture-aware one-stage fabric defect detection network with adaptive feature fusion and multi-task training. J Intell Manuf2024;
35(3): 1267–1280. https://doi.org/10.1007/s10845-023-02105-4
25.
ChenRCaiDHuX, et al.
Defect detection method of aluminum profile surface using deep self-attention mechanism under hybrid noise conditions. IEEE Trans Instrum Meas2021;
70: 1–9. https://doi.org/10.1109/TIM.2021.3109723
26.
HuJJiangQCongR, et al.
Two-branch deep neural network for underwater image enhancement in HSV color space. IEEE Signal Process Lett2021;
28: 2152–2156. https://doi.org/10.1109/LSP.2021.3099746
27.
ChenSLaiYKuoC, et al.
A surface defect detection system for golden diamond pineapple based on CycleGAN and YOLOv4. J King Saud Univ Comput Inform Sci2022;
34(10): 8041–8053. https://doi.org/10.1016/j.jksuci.2022.07.018
28.
IkotunAMEzugwuAEAbualigahL, et al.
K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data. Inform Sci2023;
622: 178–210. https://doi.org/10.1016/j.ins.2022.11.139
29.
YanFYangCZhangX, et al.
A 3-D convolution-based burn-through point multistep prediction model for sintering process. IEEE Trans Ind Electron2023;
71(4): 4219–4229. https://doi.org/10.1109/TIE.2023.3279576
30.
JangJQuanCLeeHD, et al.
Falcon: lightweight and accurate convolution based on depthwise separable convolution. Knowledge Inform Syst2023;
65(5), 2225–2249. https://doi.org/10.1007/s10115-022-01818-x
31.
WanWWangLWangB, et al.
Space to depth convolution bundled with coordinate attention for detecting surface defects. Signal Image Video Process2024;
18(5): 4861–4874. https://doi.org/10.1007/s11760-024-03122-3
32.
YangXDel Rey CastilloEZouY, et al.
UAV-deployed deep learning network for real-time multi-class damage detection using model quantization techniques. Automat Construct2024;
159: 105254. https://doi.org/10.1016/j.autcon.2023.105254
33.
MirhajiHSoleymaniMAsakerehA, et al.
Fruit detection and load estimation of an orange orchard using the YOLO models through simple approaches in different imaging and illumination conditions. Comput Electron Agric2021;
191: 106533. https://doi.org/10.1016/j.compag.2021.106533
34.
LiYFanQHuangH, et al.
A modified YOLOv8 detection network for UAV aerial image recognition. Drones2023;
7(5): 304. https://doi.org/10.3390/drones7050304
35.
YuJYaoJZhangJ, et al.
SPRNet: single-pixel reconstruction for one-stage instance segmentation. IEEE Trans Cybernet2020;
51(4): 1731–1742. https://doi.org/10.1109/TCYB.2020.2969046
36.
LinZYunBZhengY.LD-YOLO: A lightweight dynamic forest fire and smoke detection model with Dysample and spatial context awareness module. Forests2024;
15(9): 1630. https://doi.org/10.3390/f15091630
37.
RenZFangFYanN, et al.
State of the art in defect detection based on machine vision. Int J Precision Eng Manuf Green Technol2022;
9(2): 661–691. https://doi.org/10.1007/s40684-021-00343-6
38.
KangMTingCTingFF, et al.
ASF-YOLO: A novel YOLO model with attentional scale sequence fusion for cell instance segmentation. Image Vision Comput2024;
147: 105057. https://doi.org/10.1016/j.imavis.2024.105057
39.
KhanamRHussainM. Yolov11: An overview of the key architectural enhancements. arXiv preprint arXiv:2410.17725, 2024. https://arxiv.org/abs/2410.17725
