Improved YOLOv5s for outer surface defect detection of elevator traction steel wire rope

doi:10.16265/j.cnki.issn1003-3033.2026.05.0395

Abstract

Abstract:

To improve the detection efficiency and automation level of hidden defects on the outer surface of elevator traction steel wire ropes, and reduce the incidence of elevator accidents, an online detection model for defects on the outer surface of elevator traction steel wire ropes based on improved YOLOv5s is constructed. Firstly, the GhostConv module is introduced into the feature extraction layer C3 module to reduce computational complexity, and the Convolutional Block Attention Module (CBAM) is integrated to enhance the feature extraction capability of small-scale defects. A feature extraction module GC-C3 (GhostConv and CBAM-C3) that integrates GhostConv and CBAM is constructed; Secondly, in the feature fusion layer, Path Aggregation Network (PANet) and Bidirectional Feature Pyramid Network (BiFPN) are used to construct a multi-scale feature fusion network PBNet (PANet BiFPN), which combines multi-scale weight allocation strategy to improve the fusion effect of multi-scale defect feature information; Then, dynamically adjusting the quality weights of prediction boxes using Weighted Intersection over Union(WIoU) loss function, reducing the interference of low-quality samples on training. Finally, the model will be deployed to the developed detection system to perform online testing on the surface defects of the traction steel wire rope on the elevator car roof, verify the improvement effect of the model, and provide grading reminders for the defects. The results showed that the average detection accuracy of the improved model was 96.2%, with a detection speed of 192 f/s, which was 4.1% and 12.3% higher than the original model, respectively. The model volume was reduced by 38.9%. According to the online visualization experiment, under the actual operating environment of the traction steel wire rope (light illumination of 200~400 lx, speed of 1.5~2 m/s), the average accuracy of the system for 8 typical external surface defects is still stable at 94.6% or above, which meets the application requirements of online detection of hidden dangers of external surface defects of the traction steel wire rope in service and reduces elevator accidents caused by external surface defects of the traction steel wire rope.

Key words: YOLOv5s, elevator traction steel wire rope, outer surface defects, defect detection, feature extraction, feature fusion

CLC Number:

X924.3安全监控系统

Li Chunsheng, Sun Weihong, Liang Man, Li Jiefeng. Improved YOLOv5s for outer surface defect detection of elevator traction steel wire rope[J]. China Safety Science Journal, 2026, 36(5): 234-242.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Table 1

Fig.7

Fig.8

Fig.9

Table 2

Fig.10

Fig.11

References 14

[1]	Zhou Ping, Zhou Gongbo, Zhu Zhencai, et al. A review of non-destructive damage detection methods for steel wire ropes[J]. Applied Sciences, 2019, 9(13): 2771-2779. doi: 10.3390/app9132771
[2]	国家市场监督管理总局特种设备安全监察局. 市场监管总局关于2023年全国特种设备安全状况的通告[EB/OL]. (2024-03-23). https://www.samr.gov.cn/zw/zfxxgk/fdzdgknr/tzsbs/art/2024/art_aea38293416e4af382c0136d2e73f8a2.html.
[3]	Ren Mingyue. Design of magnetic flux leakage detector for mine wire rope[J]. Journal of Physics: Conference Series, 2023, 2574(1): 1-11.
[4]	夏慧, 郑军, 邱锦川, 等. 钢丝绳缺陷涡流热成像在线检测方法研究[J]. 中国机械工程, 2022, 33(9): 1044-1050.
	Xia Hui, Zheng Jun, Qiu Jinchuan, et al. Research on ECT for online detection of defects in wire ropes[J]. China Mechanical Engineering, 2022, 33(9): 1044-1050.
[5]	Zhao Kun, Wan Xuanhong, Lin Yenchen, et al. Magnetic field-based non-destructive testing techniques for battery diagnostics[J]. Advanced Energy Materials, 2024, 15(10): 2 404 295-2 404 301.
[6]	苏立鹏, 娄益凡, 杨吴奔, 等. 基于改进Faster-RCNN的起重机钢丝绳表面缺陷识别方法[J]. 机电工程, 2025, 42(7):1341-1349.
	Su Lipeng, Lou Yifan, Yang Wuben, et al. Method of surface defect recognition of crane wire rope based on improved Faster-RCNN[J]. Journal of Mechanical and Electrical Engineering, 2025, 42(7): 1341-1349.
[7]	Zhou Ping, Zhou Gongbo, Wang Shihao, et al. Visual sensing inspection for the surface damage of steel wire ropes with object detection method[J]. IEEE Sensors Journal, 2022, 22(23): 22 985-22 993.
[8]	王红尧, 韩爽, 李勤怡. 改进YOLOv5的钢丝绳损伤图像识别实验方法研究[J]. 计算机工程与应用, 2023, 59(17): 99-106. doi: 10.3778/j.issn.1002-8331.2210-0505
	Wang Hongyao, Han Shuang, Li Qinyi. Experimental research on image recognition of wire rope damage based on improved YOLOv5[J]. Computer Engineering and Applications, 2023, 59(17): 99-106. doi: 10.3778/j.issn.1002-8331.2210-0505
[9]	高嘉, 刘涛, 王显峰, 等. TWRD-Net:一种用于曳引钢丝绳缺陷的实时检测网络算法[J]. 仪器仪表学报, 2023, 44(6): 223-235.
	Gao Jia, Liu Tao, Wang Xianfeng, et al. TWRD-Net: a real-time detection network algorithm for traction wire rope defects[J]. Chinese Journal of Scientific Instrument, 2023, 44 (6): 223-235.
[10]	李刚, 杜亚波, 杨庆贺, 等. 基于多尺度特征融合和注意力机制的矿区道路障碍检测[J]. 中国安全科学学报, 2024, 34(9): 87-98. doi: 10.16265/j.cnki.issn1003-3033.2024.09.1631
	Li Gang, Du Yabo, Yang Qinghe, et al. Obstacle detection on mining roads based on multi-scale feature fusion and attention mechanism[J]. China Safety Science Journal, 2024, 34(9): 87-98. doi: 10.16265/j.cnki.issn1003-3033.2024.09.1631
[11]	陈科圻, 朱志亮, 邓小明, 等. 多尺度目标检测的深度学习研究综述[J]. 软件学报, 2021, 32(4): 1201-1227.
	Chen Keqi, Zhu Zhiliang, Deng Xiaoming, et al. Deep learning for multi-scale object detection: a survey[J]. Journal of Software, 2021, 32(4): 1201-1227.
[12]	Du Yingjie, Liu Xiaofeng, Yi Yuwei, et al. Incorporating bidirectional feature pyramid network and lightweight network: a YOLOv5-GBC distracted driving behavior detection model[J]. Neural Computing and Applications, 2023, 36(17): 9903-9917. doi: 10.1007/s00521-023-09043-5
[13]	Zhen Yan, Zhang An, Zhao Xiaoming, et al. Prediction of deep low permeability sandstone seismic reservoir based on CBAM-CNN[J]. Geoenergy Science and Engineering, 2024, 24(2): 213 241-213 248.
[14]	赵江平, 王欣然, 吴立舟. 改进YOLOv5s的路面坑槽目标检测模型[J]. 中国安全科学学报, 2025, 35(1): 67-74. doi: 10.16265/j.cnki.issn1003-3033.2025.01.0619
	Zhao Jiangping, Wang Xinran, Wu Lizhou. Model of pavement pothole target detection with improved YOLOv5s[J]. China Safety Science Journal, 2025, 35(1): 67-74. doi: 10.16265/j.cnki.issn1003-3033.2025.01.0619

试验	YOLOv5s	GC-C3	PBNet	WIoU	P/%	R/%	mAP@0.5/%	FPS/ (帧/s)	模型体积/MB	浮点运算量/G
1	√	—	—	—	91.0	93.2	92.1	171	14.4	15.8
2	√	√	—	—	91.5	93.8	92.5	214	6.2	7.5
3	√	—	√	—	93.5	93.7	94.0	146	15.9	16.1
4	√	—	—	√	91.4	93.5	92.4	175	14.4	15.7
5	√	√	√	—	93.7	94.5	94.8	189	8.1	8.3
6	√	√	—	√	92.7	94.0	93.2	218	6.2	7.6
7	√	—	√	√	94.5	94.3	95.8	169	16.0	15.9
8	√	√	√	√	94.8	94.2	96.2	192	8.8	9.3

模型	P/%	mAP@0.5/%	FPS/(帧/s)
Faster RCNN	80.4	89.6	41
SSD	79.7	87.3	76
YOLOv4	91.8	89.4	102
YOLOv5s	91.0	92.1	171
YOLOv7	91.4	91.3	98
YOLOv8	91.3	91.6	131
改进模型	94.8	96.2	192

[1]	KE Binbin, SUN Chenchen. Facepiece detection model based on feature fusion for personnel in tunnel operation scenarios [J]. China Safety Science Journal, 2026, 36(1): 267-274.
[2]	MENG Guangxiong. Research on fire source visual recognition method based on multi-feature fusion [J]. China Safety Science Journal, 2025, 35(S1): 40-46.
[3]	ZHOU Yinhui, DING Yong, WU Yulong, LI Denghua, GE Dalong. Research on multi-classification detection method of wall hollow drum based on Bayesian algorithm optimization and feature fusion [J]. China Safety Science Journal, 2025, 35(11): 131-138.
[4]	HAO Qinxia, ZHEN Haolong. Unsafe behavior recognition of miners in coal mine belt area based on multimodal feature fusion [J]. China Safety Science Journal, 2025, 35(11): 32-41.
[5]	HAO Qinxia, ZHANG Jiaqian. Identification model of miners' unsafe behaviors in coal mine conveyor belt [J]. China Safety Science Journal, 2025, 35(10): 98-105.
[6]	ZHAO Jiangping, WANG Xinran, WU Lizhou. Model of pavement pothole target detection with improved YOLOv5s [J]. China Safety Science Journal, 2025, 35(1): 67-74.
[7]	LIU Huilin, FANG Qiong, JIANG Yu, WEI Huazhang, WANG Tao, ZHANG Shuchuan. A lightweight forest fire detection algorithm based on YOLOv5s [J]. China Safety Science Journal, 2025, 35(1): 75-83.
[8]	LI Gang, DU Yabo, YANG Qinghe, MAO Mengying, JIA Dongping. Obstacle detection on mining roads based on multi-scale feature fusion and attention mechanism [J]. China Safety Science Journal, 2024, 34(9): 87-98.
[9]	JIA Guimin, CHENG Yu, QI Mengfei. Multi-scale attention feature-enhanced fusion of a new network for infrared small object detection [J]. China Safety Science Journal, 2024, 34(6): 90-98.
[10]	FAN Xiaoxuan, DUAN Lixiang, ZHANG Na, LI Xingtao, JIANG Lumeng. Conjoint localization dense networks for fault feature extraction of variable load gearbox [J]. China Safety Science Journal, 2024, 34(10): 166-173.
[11]	SHANG Shuhong, CAO Liang, LU Wei. Fault diagnosis algorithm for medium draining screen based on wavelet transform feature extraction [J]. China Safety Science Journal, 2023, 33(S2): 233-237.
[12]	ZHANG Qi, JIRGEL. Detection algorithm of slime water overflow based on convolutional neural network [J]. China Safety Science Journal, 2023, 33(S1): 192-195.
[13]	YIN Yibing, ZHANG Yongliang, LIU Zunmin, ZHANG Qiang. Detection method and experiment of small hole leakage based on electrostatic sensing [J]. China Safety Science Journal, 2023, 33(9): 142-149.
[14]	LI Shanmei, ZHOU Xiangzhi. Prediction of airport departure delay based on S2S-CNN-GRU [J]. China Safety Science Journal, 2023, 33(8): 93-100.
[15]	ZHANG Lei, LI Xiwei, YAN Qianru, WANG Haosheng, LEI Weiqiang. Personnel detection algorithm in fully mechanized coal face based on improved YOLOv5s [J]. China Safety Science Journal, 2023, 33(7): 82-89.