基于小目标复杂环境YOLO算法的瓷质绝缘子低值缺陷识别方法

doi:10.19595/j.cnki.1000-6753.tces.242050

Abstract
Figure/Table
References
Related Citation (4)

Download: PDF (3556 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract When performing real-time detection of insulator low-value defects in the field, one is often faced with a complex infrared image background, which is affected by a variety of factors. This leads to the existence of multi-scale and multi-angle characteristics of the captured small target hotspots. And in order to be applicable to the edge end, it is necessary to maintain a balance between accuracy and lightweighting.
In this paper, to address the above problems, the small targets in complex environments YOLO (SCE-YOLO) algorithm is proposed. The SCE-YOLO algorithm is based on the YOLOv8 algorithm for improvement, and firstly, it uses the self-attention and convolution mix model. Attention and convolution mix (ACmix) to aggregate convolution and self-attention to increase the algorithm’s attention to the insulators, thus reducing the influence of the background on the algorithm. Secondly, the path aggregation network (PANet) is improved by cross-modal contextual feature module (CCFM) to integrate insulator features at different scales and reduce the computational effort of the model. Then the dynamic head (DyHead) is introduced and the second generation of deformable convolutional networks is improved into the third generation of deformable convolutional networks v3 (DCNV3), which significantly improves the expressive ability of the model’s target detecting head and further improves the algorithm’s ability to detect small target defects in complex environments. The algorithm’s ability to detect small target defects in complex environments is further improved by improving DCNv3. The improvement of DCNv3 also makes the model more computationally efficient. Finally, the Inner-SIoU loss function is used to replace the CIoU loss function, and the accuracy of the bounding box regression is improved by introducing an auxiliary bounding box, which improves the recall and accuracy of the detection of insulators with small targets, and enhances the robustness of the model for small targets.
Meanwhile, relevant experiments are carried out to enrich the samples of insulator low-value defect dataset. Different low-value insulators were placed in different positions, and then the humidity and shooting angle were changed to obtain infrared images of low-value insulators heating under different environmental conditions. Based on this dataset, experiments are carried out to evaluate the accuracy and robustness of the algorithm and compare it with other typical algorithms. After experimental verification, the SCE-YOLO algorithm proposed in this paper achieves a detection average accuracy of 0.874 5 while the GFLOPs are reduced to 7.5, which meets the requirements of real-time detection of low-value defects of small-target insulators in the field, and proves the validity and superiority of the algorithm proposed in this paper through the experiments such as ablation and comparison.

Key words： Infrared imaging small target detection target recognition YOLOv8

Received: 15 November 2024

PACS:

TM755

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Pei Shaotong
	Wang Weiqi
	Hu Chenlong
	Wu Mianxiao
	Sun Haichao

Cite this article:

Pei Shaotong,Wang Weiqi,Hu Chenlong等. Identification of Low-Value Defects in Porcelain Insulators Based on the Small Targets in Complex Environments YOLO Algorithm[J]. Transactions of China Electrotechnical Society, 2025, 40(23): 7793-7805.

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.242050 OR https://dgjsxb.ces-transaction.com/EN/Y2025/V40/I23/7793

[1] 郑含博, 李金恒, 刘洋, 等. 基于改进YOLOv3的电力设备红外目标检测模型[J]. 电工技术学报, 2021, 36(7): 1389-1398.
Zheng Hanbo, Li Jinheng, Liu Yang, et al.Infrared object detection model for power equipment based on improved YOLOv3[J]. Transactions of China Electro-technical Society, 2021, 36(7): 1389-1398.
[2] 尹林, 胡京, 王文彬, 等. 环境条件对基于无人机红外检测的零值绝缘子检测判据的影响研究[J]. 电瓷避雷器, 2023(5): 171-177.
Yin Lin, Hu Jing, Wang Wenbin, et al.Influence of environmental conditions on zero-value insulator detection criteria based on infrared detection through UAV[J]. Insulators and Surge Arresters, 2023(5): 171-177.
[3] 陈金法. 绝缘子红外热像检测及诊断技术研究[D]. 杭州: 浙江大学, 2011.
Chen Jinfa.Research on infrared thermal image detecting and diagnostic techniques for insulators[D]. Hangzhou: Zhejiang University, 2011.
[4] 胡晨龙, 裴少通, 刘云鹏, 等. 基于LEE-YOLOv7的输电线路边缘端实时缺陷检测方法[J]. 高电压技术, 2024, 50(11): 5047-5057.
Hu Chenlong, Pei Shaotong, Liu Yunpeng, et al.Real-time edge end defect detection method for transmission line based on LEE-YOLOv7[J]. High Voltage Engineering, 2024, 50(11): 5047-5057.
[5] 孙瑞筱, 胡玉耀, 蒋兴良, 等. 不同污秽影响下劣化瓷绝缘子发热规律与红外特征分析[J]. 电工技术学报, 2025, 40(11): 3591-3603.
Sun Ruixiao, Hu Yuyao, Jiang Xingliang, et al.Analysis of temperature rise and infrared feature of faulty porcelain insulators under different contamina-tion effects[J]. Transactions of China Electrotechnical Society, 2025, 40(11): 3591-3603.
[6] 范晓狄. 基于深度学习的电力绝缘子检测与缺陷识别方法研究[D]. 重庆: 重庆理工大学, 2024.
Fan Xiaodi.Research on defect detection and recognition methods of electrical insulator based on deep learning[D]. Chongqing: Chongqing University of Technology, 2024.
[7] Yang Yang, Geng Sanping, Cheng Chi, et al.An edge algorithm for assessing the severity of insulator discharges using a lightweight improved YOLOv8[J]. Journal of Electrical Engineering & Technology, 2025, 20(1): 807-816.
[8] 张斌. 劣化绝缘子检测技术的国内外研究现状[J]. 电网技术, 2006, 30(增刊1): 275-278.
Zhang Bin.Current study situation of faulty insulator detection method home and abroad[J]. Power System Technology, 2006, 30(S1): 275-278.
[9] 冯旭刚, 阮善会, 王正兵, 等. 基于局部和全局特征的电力设备红外和可见光图像匹配方法[J]. 电工技术学报, 2025, 40(7): 2236-2246, 2305.
Feng Xugang, Ruan Shanhui, Wang Zhengbing, et al.Infrared and visible image matching method for power equipment based on local and global features[J]. Transactions of China Electrotechnical Society, 2025, 40(7): 2236-2246, 2305.
[10] 裴少通, 张行远, 胡晨龙, 等. 基于ER-YOLO算法的跨环境输电线路缺陷识别方法[J]. 电工技术学报, 2024, 39(9): 2825-2840.
Pei Shaotong, Zhang Hangyuan, Hu Chenlong, et al.The defect detection method for cross-environment power transmission line based on ER-YOLO algorithm[J]. Transactions of China Electrotechnical Society, 2024, 39(9): 2825-2840.
[11] 赵利鸿, 高强, 于晓, 等. 基于红外图像的绝缘子提取方法[J]. 红外技术, 2020, 42(9): 840.
Zhao Lihong, Gao Qiang, Yu Xiao, et al.Insulator extraction method based on infrared image[J]. Infrared Technology, 2020, 42(9): 840.
[12] 王媛彬, 李媛媛, 段誉, 等. 基于轻量骨干网络和注意力结构的变电设备红外图像识别[J]. 电网技术, 2023, 47(10): 4358-4369.
Wang Yuanbin, Li Yuanyuan, Duan Yu, et al.Infrared image recognition of substation equipment based on lightweight backbone network and attention mechanism[J]. Power System Technology, 2023, 47(10): 4358-4369.
[13] Zhao Zhenbing, Fan Xiaoqing, Xu Guozhi, et al.Aggregating deep convolutional feature maps for insulator detection in infrared images[J]. IEEE Access, 2017, 5: 21831-21839.
[14] Zhou Ming, Wang Jue, Li Bo.ARG-mask RCNN: an infrared insulator fault-detection network based on improved mask RCNN[J]. Sensors, 2022, 22(13): 4720.
[15] Lin Yu, Tian Lianfang, Du Qiliang.Automatic overheating defect diagnosis based on rotated detector for insulator in infrared image[J]. IEEE Sensors Journal, 2023, 23(21): 26245-26258.
[16] 任毅, 王鹏, 倪彬, 等. 基于改进YOLOv5s的绝缘子定位检测及红外故障识别[J]. 测控技术, 2024, 43(8): 7-14, 22.
Ren Yi, Wang Peng, Ni Bin, et al.Insulator positioning detection and infrared fault recognition based on improved YOLOv5s[J]. Measurement & Control Technology, 2024, 43(8): 7-14, 22.
[17] 郑含博, 胡思佳, 梁炎燊, 等. 基于YOLO-2MCS的输电线路走廊隐患目标检测方法[J]. 电工技术学报, 2024, 39(13): 4164-4175.
Zheng Hanbo, Hu Sijia, Liang Yanshen, et al.A hidden danger object detection method for transmis-sion line corridor based on YOLO-2MCS[J]. Trans-actions of China Electrotechnical Society, 2024, 39(13): 4164-4175.
[18] 汤璐, 王淑青, 王年涛, 等. 基于改进YOLOX网络的雾天绝缘子缺陷检测[J]. 高压电器, 2024, 60(3): 223-228.
Tang Lu, Wang Shuqing, Wang Niantao, et al.Insulator defect detection in foggy condition based on improved YOLOX network[J]. High Voltage Apparatus, 2024, 60(3): 223-228.
[19] Ge Zheng, Liu Songtao, Wang Feng, et al.YOLOX: Exceeding YOLO series in 2021[J/OL]. ArXiv, 2021: 2107.08430v2[2024-10-30]. https://doi.org/10.48550/arXiv.2107.08430.
[20] Pan Xuran, Ge Chunjiang, Lu Rui, et al.On the integration of self-attention and convolution[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, 2022: 805-815.
[21] Jung H, Salam G P.Hadronic final state predictions from CCFM: the hadron-level Monte Carlo generator CASCADE[J]. The European Physical Journal C-Particles and Fields, 2001, 19: 351-360.
[22] Dai Xiyang, Chen Yinpeng, Xiao Bin, et al.Dynamic head: unifying object detection heads with attentions[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 2021: 7369-7378.
[23] Wang Wenhai, Dai Jifeng, Chen Zhe, et al.InternImage: exploring large-scale vision foundation models with deformable convolutions[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, 2023: 14408-14419.
[24] Zhang Hao, Xu Cong, Zhang Shuaijie. Inner-IoU: more effective intersection over union loss with auxiliary bounding box[J/OL]. ArXiv, 2023: 2311. 02877[2024-10-30]. https://arxiv.org/abs/2311.02877v4.
[25] 刘云鹏, 张凯元, 付炜平, 等. 不同湿度下低值瓷质绝缘子的发热特性[J]. 高电压技术, 2018, 44(6): 1741-1749.
Liu Yunpeng, Zhang Kaiyuan, Fu Weiping, et al.Heating characteristics of deteriorated porcelain insulator under different humidities[J]. High Voltage Engineering, 2018, 44(6): 1741-1749.
[26] 蒋兴良, 夏强峰. 污秽对红外成像法检测劣化绝缘子的影响[J]. 高电压技术, 2010, 36(7): 1619-1624.
Jiang Xingliang, Xia Qiangfeng.Influence of conta-mination to infrared imaging method detecting deterioration insulators[J]. High Voltage Engineering, 2010, 36(7): 1619-1624.
[27] 韩永霞, 唐力, 郝艳捧, 等. 特高压直流线路复合绝缘子悬挂方式对其冲击闪络特性的影响[J]. 高电压技术, 2015, 41(4): 1272-1277.
Han Yongxia, Tang Li, Hao Yanpeng, et al.Influence of suspension patterns of UHVDC line composite insulator on its impulse flashover characteristics[J]. High Voltage Engineering, 2015, 41(4): 1272-1277.
[28] Tian Zhi, Shen Chunhua, Chen Hao, et al.FCOS: fully convolutional one-stage object detection[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea, 2019: 9626-9635.
[29] 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 (CVPR), Las Vegas, NV, USA, 2016: 779-788.
[30] Zhao Yian, Lv Wenyu, Xu Shangliang, et al.DETRs beat YOLOs on real-time object detection[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2024: 16965-16974.
[31] Wang Yuanyuan, Wang Chao, Zhang Hong, et al.Automatic ship detection based on RetinaNet using multi-resolution Gaofen-3 imagery[J]. Remote Sensing, 2019, 11(5): 531.
[32] Wang Ao, Chen Hui, Lin Zijia, et al.Rep ViT: revisiting mobile CNN from ViT perspective[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2024: 15909-15920.
[33] Liu Haoyu, Gao Shuguo, Tian Xu, et al.Transformer abnormal heat accurate identification method based on AHIPDNet[J]. Scientific Reports, 2024, 14(1): 29456.
[34] Ou Jianhua, Wang Jianguo, Xue Jian, et al.Infrared image target detection of substation electrical equipment using an improved faster R-CNN[J]. IEEE Transactions on Power Delivery, 2023, 38(1): 387-396.
[35] Li Changwu, Shi Ying, Lu Ming, et al.A composite insulator overheating defect detection system based on infrared image object detection[J]. IEEE Trans-actions on Power Delivery, 2025, 40(1): 203-214.