Experimental Research on Image Recognition of Wire Rope Damage Based on Improved YOLOv5

doi:10.3778/j.issn.1002-8331.2210-0505

Abstract

Abstract: Wire rope plays a very important role in coal mine equipment. In order to find the wire rope damage as early as possible, conduct early warning and fault handling in advance, and protect the safety of personnel under the mine, a method of wire rope damage identification and detection based on depth learning is proposed. The target detection algorithm YOLOv5 is adopted and improved. The fast adaptive weighted median filter is used for image pre-processing to improve the recognition accuracy of wire rope damage images. After the improvement, the running speed is increased to 187?ms/piece, and the enhancement effect is good. It integrates CBAM and Transformer prediction heads（TPH） into YOLOv5, and inputs the expanded dataset into the improved model for training and testing. The experimental results show that the improved model has good detection performance, and the final average accuracy rate reaches 0.893, 0.037 higher than the original algorithm, 0.196, 0.162 and 0.102 higher than the traditional detection algorithm SSD, Faster R-CNN and the original YOLOv3, respectively. It shows that the algorithm in this paper has high accuracy and effectively improves the recognition accuracy of wire rope damage images.

Key words: wire rope damage, target detection, image recognition, YOLOv5

摘要： 钢丝绳在煤矿设备中发挥着非常重要的作用。为了能够及早发现钢丝绳损伤，提前进行预警和故障处理，保护矿下人员安全，提出了一种基于深度学习的钢丝绳损伤识别检测方法，采用目标检测算法YOLOv5并对其进行改进。采用快速自适应加权中值滤波进行图像预处理，提高钢丝绳损伤图像识别精度，改进后运行速度提升到187?ms/张，且增强效果良好；将CBAM模块和Transformer prediction heads（TPH）集成到YOLOv5，数据集扩充后输入到改进的模型中进行训练测试。实验结果表明，改进后的模型检测性能良好，最终平均准确率达到了0.893，比原算法高了0.037，比传统检测算法SSD、Faster R-CNN以及原始YOLOv3分别高0.196、0.162、0.102，表明该算法精度较高，有效提高了钢丝绳损伤图像的识别准确率。

关键词: 钢丝绳损伤, 目标检测, 图像识别, YOLOv5

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.

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.

References

[1] 赵文，薛涛，凡成华，等.矿井提升机钢丝绳损伤在线检测方法研究[J].矿山机械，2022，50（6）：22-26.
ZHAO W，XUE T，FAN C H，et al.Research on online detection method for damage of wire rope of mine hoist[J].Mining & Processing Equipment，2022，50（6）：22-26.
[2] 靳志强.矿用提升机钢丝绳损伤检测装置的设计[J].机械管理开发，2021（8）：253-255.
JIN Z Q.Design of the steel wire rope damage detection device for the mine hoist[J].Mechanical Management and Development，2021（8）：253-255.
[3] HUA L J.Non-destructive testing of coal mine equipment[J].Machinery Management Development，2017，32（4）：64-65.
[4] LI X H，HE L，CUI T S，et al.Research on optimisation and extension of comprehensive detection methord of steel wire rope for mine[J].Science and Technology Innovation Duide，2018，15（2）：36-38.
[5] 高升，王巧华，付丹丹，等.红提糖度和硬度的高光谱成像无损检测[J].光学学报，2019，39（10）：347-356.
GAO S，WANG Q H，FU D D，et al.Nondestructive detection of sugar content and firmness of red globe grape by hyperspectral imaging[J].Acta Optica Sinica，2019，39（10）：347-356.
[6] YAN X L，ZHANG D L，PAN S M，et al.Online nondestructive testing for fine steel wire rope in electromagnetic interference enviroment[J].NDT&E International，2017，92：75-81.
[7] 马宏伟，毛清华，张旭辉.矿用强力带式输送机智能监控技术研究进展[J].振动、测试与诊断，2016，36（2）：213-219.
MA H W，MAO Q H，ZHANG X H.Research progress of intelligent monitoring technology for mining power belt conveyor[J].Journal of Vibration，Measurement and Diagnosis，2016，36（2）：213-219.
[8] 王倩.钢丝绳故障检测与实验分析[D].北京：北京工业大学，2014.
WANG Q.Fault detection and experimental analysis of wire rope[D].Beijing：Beijing University of Technology，2014.
[9] 冯军政，王军，谭继文.基于磁桥路的钢丝绳LMA型损伤检测试验研究[J].青岛理工大学学报，2012，33（5）：73-77.
FENG J Z，WANG J，TAN J W.Experimental study on LMA damage detection of wire rope based on magnetic bridge[J].Journal of Qingdao Technological University，2012，33（5）：73-77.
[10] 戴勇磊，吴遐，张雷，等.基于机器视觉的电梯钢丝绳缺陷检测方法[J].中国电梯，2018，29（7）：10-12.
DAI Y L，WU X，ZHANG L，et al.Inspection method of elevator wire rope defects based on machine vision[J].China Elevator，2018，29（7）：10-12.
[11] ZOU P，ZHOU G B，HE Z Z，et al.A novel texture-based damage detection method for wire ropes[J].Measurement，2019，148：1-14.
[12] 梁波，卢军，曹阳.基于改进U-Net卷积神经网络的钢轨表面损伤检测方法[J].激光与光电子学进展，2021，58（10）：334-340.
LIANG B，LU J，CAO Y.Rail surface damage detection method based on improved U-Net convolution neural network[J].Laser and Optoelectronics Progress，2021，58（10）：334-340.
[13] HUANG X Y，LIU Z L，ZHANG X Y，et al.Surface damage detection for steel wire ropes using deep learning and computer vision techniques[J].Measurement，2020，161：1-12.
[14] 曲诚，陈景龙，常远洪，等.面向钢丝绳微弱损伤智能识别的多尺度注意力网络[J].西安交通大学学报，2021，55（7）：141-150.
QU C，CHEN J L，CHANG Y H，et al.Multi-scale attention network for intelligent identification of weak damage of wire rope[J].Journal of Xi’an Jiaotong University，2021，55（7）：141-150.
[15] 毛晓琦.钢绳表面缺陷的视觉检测方法研究[D].西安：西安理工大学，2021.
MAO X Q.Research on visual inspection method of steel rope surface defects[D].Xi’an：Xi’an University of Technology，2021.
[16] 李希，徐翔，李军.面向航空飞行安全的遥感图像小目标检测[J].航空兵器，2020，27（3）：54-61.
LI X，XU X，LI J.Small target detection in remote sensing image for aviation flight safety[J].Aero Weaponry，2020，27（3）：54-61.
[17] REDMON J，FARHADI A.Yolov3：an incremental improvement[J].arXiv：1804.02767，2018.
[18] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot multibox detector[C]//European Conference on Computer Vision.Cham：Springer，2016：31-37.
[19] REN S，HE K，GIRSHICK R，et al.Faster R-CNN：towards realtime object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2016，39（6）：1137-1149.
[20] 蒋镕圻，彭月平，谢文宣，等.嵌入scSE模块的改进YOLOv4小目标检测算法[J].图学学报，2021，42（4）：546-555.
JIANG R Q，PENG Y P，XIE W X，et al.Improved YOLOv4 small target detection algorithm with embedded scSE module[J].Journal of Graphics，2021，42（4）：546-555.
[21] LIN T Y，DOLLáR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition.New York：IEEE Press，2017：936-944.
[22] REZATOFIGHI H，TSOI N，GWAK J，et al.Generalized intersection over union：a metric and a loss for bounding box regression[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington DC，USA：IEEE Press，2019：658-666.
[23] 胡均平，王鸿树，戴小标，等.改进YOLOv5的小目标交通标志实时检测算法[J].计算机工程与应用，2023，59（2）：185-193.
HU J P，WANG H S，DAI X B，et al.Real-time detection algorithm for small-target traffic signs based on improved YOLOv5[J].Computer Engineering and Applications，2023，59（2）：185-193.