[1]黄荷,黄靖,廖佳祥.基于机器视觉的电力塔杆锈蚀程度检测[J].福建理工大学学报,2024,22(06):582-589.[doi:10.3969/j.issn.2097-3853.2024.06.011]
　HUANG He,HUANG Jing,LIAO Jiaxiang.Corrosion degree detection of power tower poles based on machine vision[J].Journal of Fujian University of Technology;,2024,22(06):582-589.[doi:10.3969/j.issn.2097-3853.2024.06.011]
点击复制

基于机器视觉的电力塔杆锈蚀程度检测
分享到：

《福建理工大学学报》[ISSN:2097-3853/CN:35-1351/Z]

卷:
第22卷
期数:
2024年06期
页码:
582-589
栏目:
出版日期:
2024-12-25

文章信息/Info

Title:
Corrosion degree detection of power tower poles based on machine vision
作者:
黄荷; 黄靖; 廖佳祥
福建理工大学电子电气与物理学院
Author(s):
HUANG He;  HUANG Jing;  LIAO Jiaxiang
School of Electronic, Electrical Engineering and Physics, Fujian University of Technology
关键词:
机器视觉; 电力塔杆; YOLACT; 锈蚀检测; 线性投影
Keywords:
machine vision; power tower poles; YOLACT; corrosion detection; linear projection
分类号:
TP391.4
DOI:
10.3969/j.issn.2097-3853.2024.06.011
文献标志码:
A
摘要:
针对电力塔杆在不同环境下运行容易产生表面锈蚀的问题，利用机器视觉技术引入一种基于YOLACT的实例分割方法。将ResNet作为主干网络来提取图像特征，使用多损失函数和Fast NMS来提高模型的分割性能；以YOLACT建立基线，在不同主干网络和分辨率下进行训练，选择平均精度较好的ResNet101-700×700模型；从采集的图像中推断锈蚀的RGB 颜色特征并划分锈蚀等级。使用边界框策略对塔杆进行锈蚀程度检测，提出线性投影的概念以更好地定位出锈蚀区域。在测试集上对该方法进行验证，其识别平均精确率达到97.3%、平均召回率为94.1%。
Abstract:
In response to the problem of surface corrosion of power tower poles in different environments, an instance segmentation method based on YOLACT is introduced by using machine vision technology. ResNet is used as the backbone network to extract image features, and multi-loss functions and Fast NMS are used to improve the segmentation performance of the model. YOLACT is used to establish the baseline, training is conducted at different backbone networks and resolutions, the ResNet101-700×700 model with better average accuracy is selected. The RGB color characteristics of corrosion are deduced from the acquired images and the corrosion degree is divided. A boundary box strategy is employed for corrosion detection on the towers. The concept of linear projection is introduced to better localize the corroded areas. Validation results on the test dataset achieve an average precision of 97.3% and an average recall of 94.1%.

参考文献/References:

［1］ TIAN G S，MENG S，BAI X J，et al. Electric tower target identification based on high-resolution SAR image and deep learning［J］. Journal of Physics：Conference Series，2020，1453（1）：012117. ［2］ 卢君妥. 飞行机器人巡检图像中关键部件缺陷特征量化［D］. 北京：华北电力大学，2016. ［3］ 商俊平，李储欣，陈亮. 基于视觉的绝缘子定位与自爆缺陷检测［J］. 电子测量与仪器学报，2017，31（6）：844-849. ［4］ NASH W，DRUMMOND T，BIRBILIS N. Quantity beats quality for semantic segmentation of corrosion in images［EB/OL］. 2018：1807.03138.https：arxiv.org/abs/1807.03138v1.［5］ HE K M，GKIOXARI G，DOLLAR P，et al. Mask R-CNN［C］∥2017 IEEE International Conference on Computer Vision （ICCV）. Venice: IEEE，2017：2961-2969. ［6］ 张洪才. 直升机巡检输电线路锈蚀缺陷检测识别方法研究［D］. 大连：大连海事大学，2009. ［7］ BOLYA D, ZHOU C, XIAO F S, et al. YOLACT: Real-time instance segmentation［C］∥2019 IEEE/CVF International Conference on Computer Vision （ICCV）. Seoul, Korea（South）: IEEE,2019:9157-9166. ［8］ 李瑾，范佳能，刘屹然. 复杂环境基于YOLACT电缆识别与定位［J］. 电子测量技术，2023，46（4）：114-120. ［9］ REN S Q，HE K M，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. ［10］ 张冬明，靳国庆，鲁鼎煜，等. 注意力引导的标志检测与识别［J］. 软件学报, 2024, 35(11): 5116-5132.［11］ 王凤随，王启胜，陈金刚，等. 基于注意力机制和SoftNMS的改进Faster RCNN目标检测算法［J］. 激光与光电子学进展，2021，58（24）：405-416. ［12］ 陈肇隆. 基于无人机图像的输电线路附件识别与缺陷检测［D］. 广州：华南理工大学，2020. ［13］ 国家质量监督检验检疫总局，中国国家标准化管理委员会. 涂覆涂料前钢材表面处理 表面清洁度的目视评定 第1部分：未涂覆过的钢材表面和全面清除原有涂层后的钢材表面的锈蚀等级和处理等级：GB/T 8923.1—2011［S］. 北京：中国标准出版社，2013.［14］ 逯鹏，赵天淞，王剑. 基于计算机视觉的钢结构表面锈蚀程度检测方法［J］.工业建筑, 2024, 54(8): 133-139.

相似文献/References:

[1]檀甫贵、邹复民、刘丽桑、李建兴.基于机器视觉的软包锂电池表面缺陷检测[J].福建理工大学学报,2020,18(03):267.[doi:10.3969/j.issn.1672-4348.2020.03.012]
　TAN Fugui,ZOU Fumin,LIU Lisang,et al.Surface defect detection of soft-pack lithium battery based on machine vision[J].Journal of Fujian University of Technology;,2020,18(06):267.[doi:10.3969/j.issn.1672-4348.2020.03.012]
[2]戴福全、刘路杰.基于视觉引导的机器人抓取分类系统设计[J].福建理工大学学报,2020,18(06):530.[doi:10.3969/j.issn.1672-4348.2020.06.004]
　DAI Fuquan,LIU Lujie.Design of vision-guided robotic grasping classification system[J].Journal of Fujian University of Technology;,2020,18(06):530.[doi:10.3969/j.issn.1672-4348.2020.06.004]
[3]李建兴、林华良、俞斌、陈炜、林晨煌、黄诗婷.基于机器视觉的汽车角窗玻璃混线检测算法[J].福建理工大学学报,2021,19(03):223.[doi:10.3969/j.issn.1672-4348.2021.03.004]
　LI Jianxing,LIN Hualiang,YU Bin,et al.Machine vision-based non-congeneric product detection algorithm for vehicle quarter glass[J].Journal of Fujian University of Technology;,2021,19(06):223.[doi:10.3969/j.issn.1672-4348.2021.03.004]
[4]李济泽,位威,张凯凯.基于机器视觉的养殖鱼摄食行为识别方法[J].福建理工大学学报,2022,20(04):378.[doi:10.3969/j.issn.1672-4348.2022.04.012]
　LI Jize,WEI Wei,ZHANG Kaikai.Recognition method of fish feeding behavior based on machine vision[J].Journal of Fujian University of Technology;,2022,20(06):378.[doi:10.3969/j.issn.1672-4348.2022.04.012]
[5]仓大健,吴选忠,李占福.基于R-Y通道冗余去除钢筋层阴影方法[J].福建理工大学学报,2022,20(04):391.[doi:10.3969/j.issn.1672-4348.2022.04.014]
　CANG Dajian,WU Xuanzhong,LI Zhanfu.Spatial ghosting elimination method of reinforcement layer based on R-Y channel redundancy removal[J].Journal of Fujian University of Technology;,2022,20(06):391.[doi:10.3969/j.issn.1672-4348.2022.04.014]
[6]刘佳鑫,孔令华,郑积仕,等.基于机器视觉的木材径级测量系统设计[J].福建理工大学学报,2022,20(06):607.[doi:10.3969/j.issn.1672-4348.2022.06.016]
　LIU Jiaxin,KONG Linghua,ZHENG Jishi,et al.Design of wood diameter classification system based on machine vision[J].Journal of Fujian University of Technology;,2022,20(06):607.[doi:10.3969/j.issn.1672-4348.2022.06.016]
[7]董世超,陈丙三,连长伟,等.基于机器视觉的烟梗长度测量方法[J].福建理工大学学报,2024,22(01):74.[doi:10.3969/j.issn.2097-3853.2024.01.011]
　DONG Shichao,CHEN Bingsan,LIAN Changwei,et al.Measurement method of tobacco stem length based on machine vision[J].Journal of Fujian University of Technology;,2024,22(06):74.[doi:10.3969/j.issn.2097-3853.2024.01.011]
[8]戴福全,池生烨.金刚石线表面缺陷视觉检测方法[J].福建理工大学学报,2024,22(03):267.[doi:10.3969/j.issn.2097-3853.2024.03.009]
　DAI Fuquan,CHI Shengye.Visual inspection method for surface defects of diamond wire[J].Journal of Fujian University of Technology;,2024,22(06):267.[doi:10.3969/j.issn.2097-3853.2024.03.009]
[9]易思敏,曾绍锋,于明源,等.基于机器视觉的斜齿轮廓参数测量方法[J].福建理工大学学报,2024,22(06):549.[doi:10.3969/j.issn.2097-3853.2024.06.006]
　YI Simin,ZENG Shaofeng,YU Mingyuan,et al.Measurement method of helical gear profile parameters based on machine vision[J].Journal of Fujian University of Technology;,2024,22(06):549.[doi:10.3969/j.issn.2097-3853.2024.06.006]

更新日期/Last Update: 2024-12-25