基于Gromov-Wasserstein最优传输的输电线路小目标检测方法

doi:10.6040/j.issn.1672-3961.0.2023.098

摘要/Abstract

摘要： 针对输电线路无人机线路巡检场景中目标检测算法在处理线路缺陷、零部件缺失等小目标时性能严重下降的问题,从标签分配角度提出新的损失函数,提高小目标检测的准确性和效果。区别于传统目标检测方法,将每个目标预测框视为高斯感受野,将真实值视为高斯热图,通过计算2个高斯分布之间的距离进行标签分配;提出利用Gromov-Wassertein最优传输引导模型学习,该方法可以建立在现有的检测模型之上。对多个输电线路目标检测数据集进行试验,结果表明,采用高斯感受野和最优传输的标签分配方案在输电线路巡检中的小目标检测方面具有良好的效果。

关键词: 输电线路, 小目标检测, 深度学习, 最优传输, 标签分配

中图分类号:

TP391

索大翔,李波. 基于Gromov-Wasserstein最优传输的输电线路小目标检测方法[J]. 山东大学学报 (工学版), 2024, 54(3): 22-29.

SUO Daxiang, LI Bo. Transmission line small object detection based on Gromov-Wassertein optimal transport[J]. Journal of Shandong University(Engineering Science), 2024, 54(3): 22-29.

参考文献

[1] 裴伟, 许晏铭, 朱永英, 等. 改进的SSD航拍目标检测方法[J]. 软件学报, 2019, 30(3): 738-758. PEI Wei, XU Yanming, ZHU Yongying, et al. The target detection method of aerial photography images with improved SSD[J]. Journal of Software, 2019, 30(3): 738-758.
[2] 周霞, 陈旦, 李霞. 我国供电可靠性的现状分析与展望[J].供用电, 2021, 38(1): 49-54. ZHOU Xia, CHEN Dan, LI Xia. Analysis and prospect of power supply reliability in China[J]. Distribution & Utilization, 2021, 38(1): 49-54.
[3] WU H B, XI Y P, FANG W M, et al. Damper detection in helicopter inspection of power transmission line[C] //2014 Fourth International Conference on Instrumentation and Measurement, Computer, Communication and Control. Harbin, China: IEEE, 2014: 628-632.
[4] GUO Z, TIAN Y, MAO W. A robust faster R-CNN model with feature enhancement for rust detection of transmission line fitting[J]. Sensors, 2022, 22: 7961-7977.
[5] 李海峰, 武剑灵, 李立学. YOLOv3-ad在输电线路监控中提高小目标物体检测准确度的研究[J]. 广东电力, 2021, 34(2): 92-100. LI Haifeng, WU Jianling, LI Lixue. Research on YOLOv3-ad in improving detection accuracy of small objects in transmission line monitoring[J]. Guangdong Electric Power, 2021, 34(2): 92-100.
[6] 戚银城, 江爱雪, 赵振兵, 等. 基于改进SSD模型的输电线路巡检图像金具检测方法[J]. 电测与仪表, 2019, 56(22): 7-12. QI Yincheng, JIANG Aixue, ZHAO Zhenbing, et al. Fittings detection method in patrol images of transmission line based on improved SSD[J]. Electrical Measurement & Instrumentation, 2019, 56(22): 7-12.
[7] 邓天华, 赵曙光, 刘西钉. 基于改进 SSD 算法的输电线异物附着故障检测识别技术研究[J]. 智能电网, 2022, 12(2): 36-42. DENG Tianhua, ZHAO Shuguang, LIU Xiding. Research on detection and recognition technology of foreign object adhesion fault in transmission line based on improved SSD algorithm[J]. Smart Grid, 2022, 12(2): 36-42.
[8] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C] //2017 IEEE International Conference on Computer Vision(ICCV). Venice, Italy: IEEE, 2017: 2999-3007.
[9] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C] //Proceedings of the 14th European Conference on Computer Vision. Amsterdam, Netherlands: Springer, 2016: 21-37.
[10] 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, USA: IEEE, 2016: 779-788.
[11] TIAN Z, SHEN C, CHEN H, et al. FCOS: fully convolutional one-stage object detection[C] //2019 IEEE/CVF International Conference on Computer Vision(ICCV). Seoul, Korea: IEEE, 2019: 9626-9635.
[12] KONG T, SUN F, LIU H, et al. Foveabox: beyound anchor-based object detection[J]. IEEE Transactions on Image Processing, 2020, 29: 7389-7398.
[13] REN S, HE K, 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.
[14] DAI J, LI Y, HE K, et al. R-FCN: object detection via region-based fully convolutional networks[C] //Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona, Spain: NIPS, 2016: 29-38.
[15] LI Z, PENG C, YU G, et al. Light-head R-CNN: in defense of two-stage object detector[EB/OL].(2017-11-23)[2023-4-15]. https://arxiv.org/abs/1711.07264.
[16] 李辉, 董燕, 刘祥, 等. 基于两阶段深度网络的输电线路异常目标检测方法[J]. 控制与决策, 2022, 37(7): 1873-1882. LI Hui, DONG Yan, LIU Xiang, et al. Transmission line abnormal object detection method based on deep network of two-stage[J]. Control and Decision, 2022, 37(7): 1873-1882.
[17] 吴军, 白梁军, 董晓虎, 等. 基于Cascade R-CNN算法的输电线路小目标缺陷检测方法[J]. 电网与清洁能源, 2022, 38(4): 19-27. WU Jun, BAI Liangjun, DONG Xiaohu, et al. Transmission line small target defect detection method based on Cascade R-CNN algorithm[J]. Power System and Clean Energy, 2022, 38(4): 19-27.
[18] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C] //Proceedings of the 16th European Conference on Computer Vision. Glasgow, UK: Springer, 2020: 213-229.
[19] ZHU X, SU W, LU L, et al. Deformable DETR: deformable transformers for end-to-end object detection[EB/OL].(2021-03-18)[2023-04-17]. https://arxiv.org/abs/2010.01459v1.
[20] GE Z, LIU S, LI Z, et al. OTA: optimal transport assignment for object detection[C] //2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Nashville, TN, USA: IEEE, 2021: 303-312.
[21] CUTURI M. Sinkhorn distances: lightspeed computation of optimal transport[C] //Proceedings of the 26th International Conference on Neural Information Processing Systems. Stateline, USA: NIPS, 2013: 2292-2300.
[22] LUO W, LI Y, URTASUN R, et al. Understanding the effective receptive field in deep convolutional neural networks[C] //Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona, Spain: NIPS, 2016: 4905-4913.
[23] XU C, WANG J, YANG W, et al. RFLA: Gaussian receptive field based label assignment for tiny object detection[C] //Proceedings of the 17th European Conference on Computer Vision. Tel Aviv, Israel: Springer, 2022: 526-543.
[24] ABDELFATTAH R, WANG X, WANG S. TTPLA: an aerial-image dataset for detection and segmentation of transmission towers and power lines[C] //Proceedings of the 15th Asian Conference on Computer Vision. Kyoto, Japan: ACCV, 2020: 17-33.
[25] 飞桨. 输电线路金具数据集[DS/OL].(2020-12-29)[2023-02-15]. https://aistudio.baidu.com/aistudio/datasetdetail/66310.
[26] ZHANG Z D, ZHANG B, LAN Z C, et al. FINet: an insulator dataset and detection benchmark based on synthetic fog and improved YOLOv5[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 6006508.
[27] ZHAI Y, YANG X, WANG Q, et al. Hybrid knowledge R-CNN for transmission line multifitting detection[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 5013312.
[28] ZHAI Y, WANG Q, GUO C, et al. CGRM: a context-based graph reasoning model for fitting detection[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 2518213.

相关文章 15

[1]	聂秀山,巩蕊,董飞,郭杰,马玉玲. 短视频场景分类方法综述[J]. 山东大学学报 (工学版), 2024, 54(3): 1-11.
[2]	高泽文,王建,魏本征. 基于混合偏移轴向自注意力机制的脑胶质瘤分割算法[J]. 山东大学学报 (工学版), 2024, 54(2): 80-89.
[3]	李璐,张志军,范钰敏,王星,袁卫华. 面向冷启动用户的元学习与图转移学习序列推荐[J]. 山东大学学报 (工学版), 2024, 54(2): 69-79.
[4]	陈成,董永权,贾瑞,刘源. 基于交互序列特征相关性的可解释知识追踪[J]. 山东大学学报 (工学版), 2024, 54(1): 100-108.
[5]	李家春,李博文,常建波. 一种高效且轻量的RGB单帧人脸反欺诈模型[J]. 山东大学学报 (工学版), 2023, 53(6): 1-7.
[6]	王旭晴,魏伟波,杨光宇,宋金涛,吕婷,潘振宽. 基于算法展开的图像盲去模糊深度学习网络[J]. 山东大学学报 (工学版), 2023, 53(6): 35-46.
[7]	王碧瑶,韩毅,崔航滨,刘毅超,任铭然,高维勇,陈姝廷,刘嘉巍,崔洋. 基于图像的道路语义分割检测方法[J]. 山东大学学报 (工学版), 2023, 53(5): 37-47.
[8]	周晓昕,廖祝华,刘毅志,赵肄江,方艺洁. 融合历史与当前交通流量的信号控制方法[J]. 山东大学学报 (工学版), 2023, 53(4): 48-55.
[9]	于畅,伍星,邓秋菊. 基于深度学习的多视角螺钉缺失智能检测算法[J]. 山东大学学报 (工学版), 2023, 53(4): 104-112.
[10]	宋佳芮,陈艳平,王凯,黄瑞章,秦永彬. 基于Affix-Attention的命名实体识别语义补充方法[J]. 山东大学学报 (工学版), 2023, 53(2): 70-76.
[11]	李旭涛,杨寒玉,卢业飞,张玮. 基于深度学习的遥感图像道路分割[J]. 山东大学学报 (工学版), 2022, 52(6): 139-145.
[12]	袁钺,王艳丽,刘勘. 基于空洞卷积块架构的命名实体识别模型[J]. 山东大学学报 (工学版), 2022, 52(6): 105-114.
[13]	孟令灿,聂秀山,张雪. 基于遮挡目标去除的公交车拥挤度分类算法[J]. 山东大学学报 (工学版), 2022, 52(4): 83-88.
[14]	杨霄,袭肖明,李维翠,杨璐. 基于层次化双重注意力网络的乳腺多模态图像分类[J]. 山东大学学报 (工学版), 2022, 52(3): 34-41.
[15]	王心哲,邓棋文,王际潮,范剑超. 深度语义分割MRF模型的海洋筏式养殖信息提取[J]. 山东大学学报 (工学版), 2022, 52(2): 89-98.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed