浏览全部资源
扫码关注微信
Published： 16 November 2023 ，
移动端阅览
梁华刚， 赵慧霞， 刘丽华， 岳鹏， 郑振宇. 2023. 结合跨层特征融合与级联检测器的防震锤缺陷检测. 中国图象图形学报， 28(11):3485-3496
Liang Huagang， Zhao Huixia， Liu Lihua， Yue Peng， Zheng Zhenyu. 2023. Combining cross-layer feature fusion with cascade detectors for anti-vibration hammer defects detection. Journal of Image and Graphics， 28(11):3485-3496
目的
2
防震锤可以减少输电线路的周期性震动以降低线路的疲劳损害，定期对防震锤进行巡检非常必要。针对目前无人机巡检输电线路所得航拍图像背景复杂，而种类较多、形状各异以及特性不一的防震锤在航拍图像中占据像素面积很小，导致防震锤检测过程中出现的检测精度低、无法确定缺陷类型等问题，提出了一种结合跨层特征融合和级联检测器的防震锤缺陷检测方法。
方法
2
本文使用无人机对防震锤部件巡检的航拍图像进行数据扩充建立防震锤缺陷检测数据集，并划分了4种缺陷类型，为研究提供了数据基础。首先，以VGG16（Visual Geometry Group 16-layer network）为基础对1、3、5层特征进行特征融合得到特征图，平衡了语义信息和空间特征；其次，使用3个级联检测器对目标进行分类定位，减小了交并比（intersection over union，IoU）阈值对网络性能的影响；最后，将非极大值抑制法替换为Soft-NMS（soft non-maximum suppression）算法，去除边界框保留了最佳结果。
结果
2
在自建数据集上验证网络模型对4种防震锤缺陷类型的检测效果，与现有基于深度学习的其他6种先进算法相比，本文算法的平均准确率比Fast R-CNN（fast region-based convolutional network）、Faster R-CNN、YOLOv4 （you only look once version 4）分别提高了13.5%、3.4%、5.8%，比SSD300 （single shot MultiBox detector 300）、YOLOv3、RetinaNet分别提高了9.5%、8.5%、8%。与Faster R-CNN相比，本文方法的误检率降低了5.61%，漏检率降低了3.01%。
结论
2
本文提出的防震锤缺陷检测方法对不同背景、不同光照、不同角度、不同尺度、不同种类和不同缺陷种类的防震锤均有较好的检测结果，不但可以更好地提取防震锤的特征，而且还能提高分类和位置预测网络的定位精度， 从而有效提高了防震锤缺陷检测算法的准确率，在满足防震锤巡检工作实际检测要求的同时还具有较好的鲁棒性和有效性。
Objective
2
Anti-vibration hammers can reduce the periodic vibration of transmission lines to minimize line fatigue damage. Therefore， a regular inspection of these hammers is necessary. Previous studies on anti-vibration hammers for transmission lines have mostly focused on the identification or fault classification of anti-vibration hammers with only one case of deficiency. However， the detection methods for such small-scale targets are limited， and their leakage and false detection rates remain very high， which cannot guarantee the safe operation of transmission lines. Therefore， in view of the complex background of the aerial images obtained from the current unmanned aerial vehicle （UAV） inspection of transmission lines， the different types， shapes， and characteristics of anti-vibration hammers occupying a small pixel area in these aerial images can lead to problems， such as inability to determine the types of defects and the low detection accuracy in the anti-vibration hammer detection process. In this paper， we propose an anti-shock hammer defect detection method based on cross-layer feature fusion and cascade detector.
Method
2
Given the lack of a public anti-vibration hammer dataset at home and abroad for targeted research， this paper takes the aerial images taken via the UAV inspection of anti-vibration hammer components as the original data and then expands the dataset by geometric and contrast transformation to ensure the equalization of the number of different types of anti-vibration hammers in the dataset and to establish an anti-vibration hammer defect detection dataset. The anti-vibration hammer defects are then refined and classified into four categories， namely， normal， corroded， broken， and collision， to serve as the data basis for the algorithm research designed in this paper. The anti-shock hammer defect detection algorithm is then designed. The proposed anti-shock hammer defect detection method is mainly divided into two major parts， namely， the feature extraction network and classification location prediction network. The feature extraction network extracts accurate anti-seismic hammer features by fusing the Visual Geometry Group 16-layer network （VGG16）. The main idea is to insert a convolutional kernel of size 1 × 1 into the last convolutional layer of the first， third， and fifth layers， after which the last convolutional layer of the first layer is connected to the maximum pooling layer and fused with the third layer using a deconvolution operation after the fifth layer. These two layers are then fused to form the final feature map， which balances the semantic information and spatial features. To reduce the impact of the intersection over union （IoU） threshold on network performance， in the localization and classification network part， the classification and location prediction network proposed in this paper uses three cascade detectors to gradually increase the IoU threshold and improve the quality of samples and the training effect of the network. The non-maximal suppression method is then replaced by the soft non-maximum suppression （Soft-NMS） algorithm to remove the bounding box.
Result
2
The main contributions of this paper are as follows： 1） the dataset part expands the data using the aerial images taken via the UAV inspection of anti-vibration hammer components to establish the anti-vibration hammer defect detection dataset. To ensure the validity of this dataset， part of those samples that cannot be easily separated even through the naked eye are reasonably removed； 2） the network model is based on VGG16， feature fusion is performed on 1， 3， and 5 layers of features to effectively obtain more features， and three cascade detectors are used to classify the target to reduce the impact of the IoU threshold on network performance and improve the detection capability of the algorithm. The average accuracy of the network model was improved by 13.5%， 3.4%， and 5.8% over the fast region-based convolutional network （Fast R-CNN）， Faster R-CNN， and you only look once version 4 （YOLOv4）， respectively， and by 9.5%， 8.5%， and 8% over single shot MultiBox detector 300 （SSD300）， YOLOv3， and RetinaNet， respectively， when compared with six other advanced algorithms based on deep learning. These results may be explained by the incorporation of multiple layers of features， which enable the model to obtain more information about low- and high-level image features. Moreover， the uses of cascade detectors reduce the impact of the IoU thresholds on network performance and eventually improve the detection accuracy of small-scale targets. Compared with Faster R-CNN， the false detection rate of the proposed method is reduced by 5.61%， whereas the missed detection rate is reduced by 3.01%. Therefore， the proposed method improves its accuracy while effectively reducing its false detection and missed detection rates and works particularly well in practical applications.
Conclusion
2
The proposed anti-shock hammer defect detection method obtains good detection results for different backgrounds， illuminations， angles， scales， and types of anti-shock hammers. Experimental results show that this method not only efficiently extracts the characteristics of anti-shock hammers but also improves the localization accuracy of the network， thus effectively improving the accuracy of the algorithm and satisfying the actual detection requirements of anti-shock hammer inspection work while showing improved robustness and effectiveness.
防震锤缺陷深度学习小尺度目标检测跨层特征融合级联检测器
shockproof hammer defectsdeep learningsmall-scale object detectioncross-layer feature fusioncascade detector
Bodla N， Singh B， Chellappa R and Davis L S. 2017. Improving object detection with one line of code ［EB/OL］. ［2022-08-22］. https://arxiv.org/pdf/1704.04503.pdfhttps://arxiv.org/pdf/1704.04503.pdf
Chen L C， Papandreou G， Kokkinos I， Murphy K and Yuille A L. 2018. DeepLab： semantic image segmentation with deep convolutional nets， atrous convolution， and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence， 40（4）： 834-848 ［DOI： 10.1109/TPAMI.2017.2699184http://dx.doi.org/10.1109/TPAMI.2017.2699184］
Chen X J， Wu Y S and Zhao L. 2010. Identification of OPGW vibration damper based on random Hough transformation. Heilongjiang Electric Power， 32（1）： 1-2， 5
陈晓娟， 吴英石， 赵亮. 2010. 基于随机Hough变换的OPGW防震锤识别. 黑龙江电力， 32（1）： 1-2， 5 ［DOI： 10.13625/j.cnki.hljep.2010.01.021http://dx.doi.org/10.13625/j.cnki.hljep.2010.01.021］
Chen X Z. 2021. Application and prospect of transmission line UAV inspection. China Power Enterprise Management， （36）： #86
陈昕哲. 2021. 输电线路无人机巡检应用及展望. 中国电力企业管理， 36： #86 ［DOI： 10.3969/j.issn.1007-3361.2021.36.040http://dx.doi.org/10.3969/j.issn.1007-3361.2021.36.040］
He K M， Zhang X Y， Ren S Q and Sun J. 2016. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas， USA： IEEE： 770-778 ［DOI： 10.1109/CVPR.2016.90http://dx.doi.org/10.1109/CVPR.2016.90］
Huang L L， Xie R and Xu Y. 2015. Invasion detection on transmission lines using saliency computation//Proceedings of 2015 IEEE International Symposium on Signal Processing and Information Technology. Abu Dhabi， United Arab Emirates： IEEE： 320-325 ［DOI： 10.1109/ISSPIT.2015.7394352http://dx.doi.org/10.1109/ISSPIT.2015.7394352］
Jia L Y， Han J and Yu H F. 2021. Shock hammer recognition based on semantic segmentation and spatial context information. Computer Measurement and Control， 29（1）： 200-205
贾立业， 韩军， 余鸿飞. 2021. 基于语义分割和空间上下文信息的防震锤识别. 计算机测量与控制， 29（1）： 200-205 ［DOI： 10.16526/j.cnki.11-4762/tp.2021.01.041http://dx.doi.org/10.16526/j.cnki.11-4762/tp.2021.01.041］
Jia Y F， Chen G D， Yang M， Xing L Y， Zhao L Q and Li S Y. 2022. Damper detection of high voltage transmission line based on deep learning. Machine Tool and Hydraulics， 50（13）： 21-25
贾雁飞， 陈广大， 杨淼， 邢砾云， 赵立权， 李帅洋. 2022. 基于深度学习的高压输电线路防振锤检测. 机床与液压， 50（13）： 21-25 ［DOI： 10.3969/j.issn.1001-3881.2022.13.004http://dx.doi.org/10.3969/j.issn.1001-3881.2022.13.004］
Jiao R T， Ni H X and Wang Z Y. 2021. Identification of Transmission line Damper based on Faster RCNN Algorithm. Journal of Changchun Institute of Technology （Natural Science Edition）， 22（1）： 38-43
焦润童， 倪虹霞， 王智昱. 2021. 基于Faster RCNN算法的输电线路防震锤识别研究. 长春工程学院学报（自然科学版）， 22（1）： 38-43 ［DOI： 10.3969/j.issn.1009-8984.2021.01.010http://dx.doi.org/10.3969/j.issn.1009-8984.2021.01.010］
Jin L J， Yan S J and Liu Y. 2012. Vibration damper recognition based on Haar-Like features and cascade AdaBoost classifier. Journal of System Simulation， 24（9）： 1806-1809
金立军， 闫书佳， 刘源. 2012. 基于类Haar特征与级联AdaBoost算法的防震锤识别. 系统仿真学报， 24（9）： 1806-1809 ［DOI： 10.16182/j.cnki.joss.2012.09.022http://dx.doi.org/10.16182/j.cnki.joss.2012.09.022］
Liang H G， Zuo C and Wei W M. 2020. Detection and evaluation method of transmission line defects based on deep learning. IEEE Access， 8： 38448-38458 ［DOI： 10.1109/ACCESS.2020.297498http://dx.doi.org/10.1109/ACCESS.2020.297498］
Liu X Y， Jiang H， Chen J， Chen J J， Zhuang S B and Miao X R. 2018. Insulator detection in aerial images based on faster regions with convolutional neural network//Proceedings of the 14th IEEE International Conference on Control and Automation （ICCA）. Anchorage， USA： IEEE： 1082-1086 ［DOI： 10.1109/ICCA.2018.8444172http://dx.doi.org/10.1109/ICCA.2018.8444172］
Lyu M， Sheng G H， Zhang W D and Jiang X C. 2012. A tower and line tracking algorithm for power transmission line inspection based on unmanned aerial vehicles. Automation of Electric Power Systems， 36（9）： 92-97
吕明， 盛戈皞， 张卫东， 江秀臣. 2012. 无人飞行器巡检输电线路的杆塔和导线跟踪算法. 电力系统自动化， 36（9）： 92-97
Pan L Z and Xiao X H. 2009. Image recognition for on-line vibration monitoring system of transmission line//Proceedings of the 9th International Conference on Electronic Measurement and Instruments. Beijing， China： IEEE： 1078-1081 ［DOI： 10.1109/ICEMI.2009.5274379http://dx.doi.org/10.1109/ICEMI.2009.5274379］
Qi Y C， Wu X L， Zhao Z B， Shi B Q and Nie L Q. 2021. Bolt defect detection for aerial transmission lines using faster R-CNN with an embedded dual attention mechanism. Journal of Image and Graphics， 26（11）： 2594-2604
戚银城， 武学良， 赵振兵， 史博强， 聂礼强. 2021. 嵌入双注意力机制的Faster R-CNN航拍输电线路螺栓缺陷检测. 中国图象图形学报， 26（11）： 2594-2604 ［DOI： 10.11834/jig.200793http://dx.doi.org/10.11834/jig.200793］
Redmon J and Farhadi A. 2018. YOLOv3： an incremental improvement ［EB/OL］. ［2022-08-22］. https://arxiv.org/pdf/1804.02767.pdfhttps://arxiv.org/pdf/1804.02767.pdf
Ren S Q， He K M， Girshick R and Sun J. 2017. Faster R-CNN： towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence， 39（6）： 1137-1149 ［DOI： 10.1109/TPAMI.2016.2577031http://dx.doi.org/10.1109/TPAMI.2016.2577031］
Simonvan K and Zisserman A. 2015. Very deep convolutional networks for large-scale image recognition ［EB/OL］. ［2022-08-22］. https://arxiv.org/pdf/1409.1556.pdfhttps://arxiv.org/pdf/1409.1556.pdf
Song W， Zuo D， Deng B F， Zhang H B， Xue K W and Hu H. 2016. Corrosion defect detection of earthquake hammer for high voltage transmission line. Chinese Journal of Scientific Instrument， 37（S1）： 113-117
宋伟， 左丹， 邓邦飞， 张海兵， 薛凯文， 胡泓. 2016. 高压输电线防震锤锈蚀缺陷检测. 仪器仪表学报， 37（S1）： 113-117 ［DOI： 10.19650/j.cnki.cjsi.2016.s1.019http://dx.doi.org/10.19650/j.cnki.cjsi.2016.s1.019］
Wang Q L， Wu B G， Li P F， Li P H， Zuo W M and Hu Q H. 2020. ECA-Net： efficient channel attention for deep convolutional neural network//Proceedings of 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle， USA： IEEE： 11531-11539 ［DOI： 10.1109/CVPR42600.2020.01155http://dx.doi.org/10.1109/CVPR42600.2020.01155］
Xu B Y， Chen Q N， Wang S W， Shen F and Wu Y B. 2021. Implementation and research of automatic line patrol based on 5G UAV in power transmission line. Yunnan Electric Power， 49（6）： 8-12
许保瑜， 陈庆宁， 王胜伟， 沈凤， 吴圆波. 2021. 基于5G无人机在电力输电线路自动巡线的实现与研究. 云南电力技术， 49（6）： 8-12 ［DOI： 10.3969/j.issn.1006-7345.2021.06.002http://dx.doi.org/10.3969/j.issn.1006-7345.2021.06.002］
Yang G， Sun C W， Wang D W， Jin T， Xu C Y， Lu Z B and Zhang X Z. 2020. Comparative study of transmission line component detection models based on UAV front end and SSD algorithm. Journal of Taiyuan University of Technology， 51（2）： 212-219
杨罡， 孙昌雯， 王大伟， 晋涛， 徐澄宇， 卢志博， 张兴忠. 2020. 基于无人机前端和SSD算法的输电线路部件检测模型对比研究. 太原理工大学学报， 51（2）： 212-219 ［DOI： 10.16355/j.cnki.issn1007-9432tyut.2020.02.009http://dx.doi.org/10.16355/j.cnki.issn1007-9432tyut.2020.02.009］
Zhang Y X， Wu G P， Liu Z Y， Yang S and Xu W Z. 2020. Transfer learning of transmission line damper and clamp detection based on YOLOv3 network. Journal of Computer Applications， 40（S2）： 188-194
张永翔， 吴功平， 刘中云， 杨松， 徐玮泽. 2020. 基于YOLOv3网络的输电线路防震锤和线夹检测迁移学习. 计算机应用， 40（S2）： 188-194 ［DOI： 10.11772/j.issn.1001-9081.2020020276http://dx.doi.org/10.11772/j.issn.1001-9081.2020020276］
Zhao Z B， Jiang Z G， Li Y X， Qi Y C， Zhai Y J， Zhao W Q and Zhang K. 2021. Overview of visual defect detection of transmission line components. Journal of Image and Graphics， 26（11）： 2545-2560
赵振兵， 蒋志钢， 李延旭， 戚银城， 翟永杰， 赵文清， 张珂. 2021. 输电线路部件视觉缺陷检测综述. 中国图象图形学报， 26（11）： 2545-2560 ［DOI： 10.11834/jig.200689http://dx.doi.org/10.11834/jig.200689］
相关文章
相关作者
相关机构
京公网安备11010802024621