[1]LI Bing,WEI Letao,ZHANG Yimu,et al.Algorithm for wind turbine blade defect detection by integrating edge enhancement and multi-scale feature aggregation[J].CAAI Transactions on Intelligent Systems,2026,21(3):701-712.[doi:10.11992/tis.202504011]
Copy

Algorithm for wind turbine blade defect detection by integrating edge enhancement and multi-scale feature aggregation

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 21 Number of periods: 2026 3 Page number: 701-712 Column: 学术论文—机器感知与模式识别 Public date: 2026-05-05
Title:
Algorithm for wind turbine blade defect detection by integrating edge enhancement and multi-scale feature aggregation
Author(s):
LI Bing1; 2; 3;  WEI Letao2;  ZHANG Yimu2;  WANG Yue2;  WU Zifeng2;  XIE Zhuofan2;  ZHAI Yongjie1; 2; 3
1. Yanzhao Electric Power Laboratory, North China Electric Power University, Baoding 071003, China;
2. Department of Automation, North China Electric Power University, Baoding 071003, China;
3. Baoding Key Laboratory of Intelligent Robot Perception and Control in Electric Power System, Baoding 071003, China
Keywords:
wind turbine blade; defect detection; irregular defects; complex backgrounds; small targets; adaptive multi-head attention; feature aggregation; deep convolution
CLC:
TP183
DOI:
10.11992/tis.202504011
Abstract:
Unmanned aerial vehicle (UAV) inspections of wind turbine blades yield aerial images characterized by complex backgrounds, small and variably scaled targets, and irregular defect shapes. These factors often cause high rates of missed detections and false alarms in existing object detection methods. To overcome these challenges, this study presents a novel defect detection framework that combines edge enhancement with multi-scale feature aggregation. First, an edge enhancement feature extraction (EEFE) module is proposed to improve the model’s ability to capture defect boundaries, thereby enhancing the representation of irregular defects. Next, an adaptive multi-head attention (AMHA) mechanism is integrated into a hierarchical spatial pyramid pooling efficient layer aggregation network (SPPELAN) architecture, forming the SPPELAN-AMHA module, which strengthens global context modeling and reduces interference from complex backgrounds. In addition, a multi-scale feature aggregation (MFA) module and an aggregation diffusion feature pyramid network (ADFPN) are designed to extract multi-scale contextual information through deep convolutions with diverse receptive fields, improving detection performance on small targets. Experimental results demonstrate that the proposed approach improves average precision by 1.7%, 8.3%, and 4.1% for surface erosion, crack, and gelcoat peeling defects, respectively, and achieves a 4.7% gain in mean average precision at intersection over union threshold 50 (mAP50) compared with the baseline YOLOv8n model, confirming its effectiveness for wind turbine blade defect detection.
References:
[1] 曹锦纲, 杨国田, 杨锡运. 基于RPCA和视觉显著性的风机叶片表面缺陷检测[J]. 图学学报, 2019, 40(4): 704-710 CAO Jingang, YANG Guotian, YANG Xiyun. Surface defect detection of wind turbine blades based on RPCA and visual saliency[J]. Journal of graphics, 2019, 40(4): 704-710
[2] ROGA S, BARDHAN S, KUMAR Y, et al. Recent technology and challenges of wind energy generation: a review[J]. Sustainable energy technologies and assessments, 2022, 52: 102239
[3] WANG Wenjie, XUE Yu, HE Chengkuan, et al. Review of the typical damage and damage-detection methods of large wind turbine blades[J]. Energies, 2022, 15(15): 5672
[4] 白健鹏, 王巍, 陈雨溪, 等. 基于轻量型YOLOv5的风机桨叶检测与空间定位[J]. 智能系统学报, 2022, 17(6): 1173-1181 BAI Jianpeng, WANG Wei, CHEN Yuxi, et al. Detection and spatial location of wind turbine blades based on lightweight YOLOv5[J]. CAAI transactions on intelligent systems, 2022, 17(6): 1173-1181
[5] CHATTERJEE J, DETHLEFS N. Scientometric review of artificial intelligence for operations & maintenance of wind turbines: The past, present and future[J]. Renewable and sustainable energy reviews, 2021, 144: 111051
[6] 贺斌, 贾建华, 赵峰, 等. 无人机在风机叶片检测中的应用[J]. 电工技术, 2019(13): 64-65 HE Bin, JIA Jianhua, ZHAO Feng, et al. Application of UAV in detection of wind turbine blades[J]. Electric engineering, 2019(13): 64-65
[7] TCHERNIAK D, M?LGAARD L L. Active vibration-based structural health monitoring system for wind turbine blade: Demonstration on an operating Vestas V27 wind turbine[J]. Structural health monitoring, 2017, 16(5): 536-550
[8] 郑浩, 周丽婷, 王湘明, 等. 电磁超声检测风机叶片缺陷的有限元仿真研究[J]. 电子测量技术, 2021, 44(23): 24-29 ZHENG Hao, ZHOU Liting, WANG Xiangming, et al. Finite element simulation study on electromagnetic ultrasonic detection of wind turbine blade[J]. Electronic measurement technology, 2021, 44(23): 24-29
[9] 马敏敏, 刘昊, 高毓欣, 等. 基于无人机与人工智能的风机叶片全自主巡检[J]. 电气时代, 2023(S2): 44-48 MA Minmin, LIU Hao, GAO Yuxin, et al. Automatic inspection of fan blades based on UAV and artificial intelligence[J]. Electric age, 2023(S2): 44-48
[10] LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//Computer Vision–ECCV 2016. Cham: Springer, 2016: 21-37.
[11] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 779-788.
[12] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 6517-6525.
[13] REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. (2018-04-08)[2025-01-01]. https://arxiv.org/abs/1804.02767.
[14] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 2014: 580-587.
[15] GIRSHICK R. Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision. Santiago: IEEE, 2016: 1440-1448.
[16] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(6): 1137-1149
[17] ZHANG Chao, WEN Chuanbo, LIU Jihui. Mask-MRNet: a deep neural network for wind turbine blade fault detection[J]. Journal of renewable and sustainable energy, 2020, 12(5): 053302
[18] LYU Liang, YAO Zhongyuan, WANG Enming, et al. Efficient and accurate damage detector for wind turbine blade images[J]. IEEE access, 2022, 10: 123378-123386
[19] RAN Xiukang, ZHANG Shang, WANG Hengtao, et al. An improved algorithm for wind turbine blade defect detection[J]. IEEE access, 2022, 10: 122171-122181
[20] 汤占军, 张朝杰, 王健, 等. 基于改进YOLOv7的风机叶片缺陷检测研究[J/OL]. 北京航空航天大学学报, 1-15. TANG Zhanjun, ZHANG Chaojie, WANG Jian, et al. Research on wind turbine blade defect detection based on improved Yolov7[J/OL]. Journal of Beijing University of Aeronautics and Astronautics, 1-15.
[21] 付军军, 陈林, 童华敏, 等. 基于改进EfficientNet的海上风机叶片早期缺陷检测及分类[J]. 电子测量技术, 2024, 47(16): 110-119 FU Junjun, CHEN Lin, TONG Huamin, et al. Early defect detection and classification of offshore wind turbine blades based on improved EfficientNet[J]. Electronic measurement technology, 2024, 47(16): 110-119
[22] BOCHKOVSKIY A, WANG C Y, LIAO H M. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. (2020-04-23)[2025-01-01]. https://arxiv.org/abs/2004.10934.
[23] LIN T Y, DOLL?R P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 936-944.
[24] 闫建红, 冉同霄. 基于YOLOv8的轻量化无人机图像目标检测算法[J]. 图学学报, 2024, 45(6): 1328-1337 YAN Jianhong, RAN Tongxiao. Lightweight UAV image target detection algorithm based on YOLOv8[J]. Journal of graphics, 2024, 45(6): 1328-1337
[25] 赵振兵, 郭广学, 王艺衡, 等. 融合边缘感知与统计纹理知识的输电线路金具锈蚀检测[J]. 智能系统学报, 2024, 19(5): 1228-1237 ZHAO Zhenbing, GUO Guangxue, WANG Yiheng, et al. Rust detection in transmission line fittings via fusion of edge perception and statistical texture knowledge[J]. CAAI transactions on intelligent systems, 2024, 19(5): 1228-1237
[26] CHOLLET F. Xception: deep learning with depthwise separable convolutions[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1800-1807.
[27] 赵振兵, 韩钰, 唐辰康. 基于改进YOLOv8的配电线路绝缘子缺陷级联检测方法[J]. 图学学报, 2025, 46(1): 1-12 ZHAO Zhenbing, HAN Yu, TANG Chenkang. Cascade detection method for insulator defects in distribution lines based on improved YOLOv8[J]. Journal of graphics, 2025, 46(1): 1-12
[28] WANG C Y, YEH I H, MARK LIAO H Y. YOLOv9: learning what you want toLearn using programmable gradient information[C]//Computer Vision–ECCV 2024. Cham: Springer, 2025: 1-21.
[29] YANG Guoyu, LEI Jie, ZHU Zhikuan, et al. AFPN: asymptotic feature pyramid network for object detection[C]//2023 IEEE International Conference on Systems, Man, and Cybernetics. Honolulu: IEEE, 2024: 2184-2189.
[30] TAN Mingxing, PANG Ruoming, LE Q V. EfficientDet: scalable and efficient object detection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2020: 10778-10787.
[31] 凌均健, 徐东华, 朱胜山, 等. 改进YOLOv8的无人机建筑外墙缺陷检测方法[J/OL]. 光通信研究, 1-9. LING Junjian, XU Donghua, ZHU Shengshan, et al. Improved UAV-Based Building Facade Defect Detection Method Using YOLOv8[J/OL]. Study on optical communications, 1-9.
[32] ZHAO Yian, LYU Wenyu, XU Shangliang, et al. DETRs beat YOLOs on real-time object detection[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024: 16965-16974.
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems