基于改进ReDet的航拍绝缘子及其缺陷定向检测算法

doi:10.7527/S1000-6893.2025.31825

Abstract

Abstract:

To address the challenges of insulators in UAV aerial images， such as multi-scale， diverse morphologies， arbitrary orientations， and inconspicuous defect features， an improved object detection algorithm based on ReDet is proposed. First， a Rotation Equivariant Attention Module （ReCBAM） is integrated into the backbone network to enhance the focus on target regions. Second， a novel rotation-equivariant recursive feature pyramid is constructed using dilated convolution group， self-gated activation functions， and an adaptive feature fusion module， providing stronger multi-scale and multi-morphological feature modeling capabilities. Finally， the RoI detector is extended to a four-stage hybrid cascade structure， incorporating contextual information fusion and an improved loss function to refine the feature representation of target regions. The improved loss function combines KL divergence， the Focal Loss mechanism， and an IoU-based dynamic weighting mechanism， establishing a geometric-spatial dynamic joint optimization mechanism for rotated bounding boxes， thereby effectively improving the localization accuracy. Experimental results on a self-built dataset show that the proposed algorithm achieves a mAP of 95.04% under oriented bounding box annotations， representing a 3.01% improvement over the baseline ReDet， and demonstrates superior performance in detecting small， rotated targets under complex backgrounds. Additionally， the model’s parameter size is only 34.57 MB， making it suitable for deployment on UAV platforms. Under horizontal bounding box annotations， the algorithm achieves a mAP of 95.56%， with a 1.57% improvement over the baseline ReDet， validating the algorithm’s strong generalization capability.

Key words: insulator defect detection, ReDet, oriented object detection, small object detection, UAV inspection

CLC Number:

V279

Yi ZHENG, Xianghong CHENG, Xingbang TANG, Yi CAO. Oriented detection algorithm for insulator and their defects from aerial images based on improved ReDet[J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(18): 331825.

Figures/Tables 16

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Table 1

Table 2

Impact of different ReCBAM integration methods on model performance

模型	$A P 1 / %$	$A P 2 / %$	$A P 3 / %$	$m A P / %$	Params/MB
ReDet	92.15	92.03	91.91	92.03	31.64
ReDet+Attention Block	92.40	92.20	92.28	92.28	32.98
ReDet+Attention Layer	92.44	92.35	92.23	92.34	31.90
ReDet+Attention Input & Output	92.30	92.20	92.20	92.23	31.92

Table 2

Fig.7

Table 3

Detection accuracy comparison under different cascade configurations

RoI Head	RiRoI Head	$A P 1 / %$	$A P 2 / %$	$A P 3 / %$	$m A P / %$
1	1	92.15	92.03	91.91	92.03
1	2	92.83	92.76	92.45	92.68
1	3	93.64	93.35	93.03	93.34
1	4	93.57	93.27	92.94	93.26
2	1	90.10	89.92	89.53	89.85

Table 3

Table 4

Fig.8

Table 5

Comparison of experimental results for different OBB algorithms

算法	主干网络	$A P 1 / %$	$A P 2 / %$	$A P 3 / %$	$m A P / %$	Params/M	FPS	GFLOPs
Oriented Retinanet^［4］	ResNet-50	81.95	81.76	81.57	81.76	38.78	21.06	203.47
SuperYOLO^［29］	CSPDarknet-53	82.14	82.11	82.05	82.10	7.70	81.90	20.89
$R 3 D e t$ ^［15］	ResNet-50	82.70	82.50	82.38	82.53	41.58	18.33	239.57
Faster R-CNN-O^［8］	ResNet-50	86.22	86.11	86.01	86.12	47.52	17.25	268.25
Oriented R-CNN^［14］	ResNet-50	88.80	88.63	88.45	88.63	41.13	18.69	225.12
RoI Transformer^［13］	ResNet-101	89.23	89.04	88.85	89.04	74.12	15.55	358.34
YOLOv8-O^［31］	CSPDarkNet-53	92.02	91.78	91.25	91.68	11.04	66.20	28.40
ReDet^［17］	ReResNet-50	92.45	92.03	91.61	92.03	31.64	23.63	151.90
Oriented R-CNN	LSKNet^［30］	93.49	93.38	92.75	93.21	21.00	45.53	116.20
Ours	CBAM_ReResNet-50	95.12	95.03	94.97	95.04	34.57	22.38	162.10

Table 5

Fig.9

Table 6

Comparison of experimental results for different HBB algorithms

算法	Backbone	$A P 1 / %$	$A P 2 / %$	$A P 3 / %$	$m A P / %$	Params/MB	FPS	GFLOPS
Retinanet^［4］	ResNet-50	83.94	83.92	83.78	83.88	34.01	25.52	205.16
Faster R-CNN^［8］	ResNet-50	88.20	88.12	87.98	88.10	41.75	21.23	278.53
YOLOv8^［31］	CSPDarkNet-53	93.35	93.31	93.12	93.26	11.04	78.42	28.42
ReDet-H^［17］	ReResNet-50	94.10	94.02	93.79	93.97	31.64	29.34	151.90
Ours-H	CBAM_ReResNet-50	95.82	95.54	95.32	95.56	34.57	27.82	162.10

Table 6

Fig.10

References 31

[1]	刘传洋，吴一全，刘景景. 无人机航拍图像中绝缘子缺陷检测的深度学习方法研究进展［J］. 电工技术学报， 2025， 40（9）： 2897-2916.
	LIU C Y， WU Y Q， LIU J J. Research progress of deep learning methods for insulator defect detection in UAV based aerial images［J］. Transactions of China Electrotechnical Society， 2025， 40（9）： 2897-2916 （in Chinese）.
[2]	罗旭东，吴一全，陈金林. 无人机航拍影像目标检测与语义分割的深度学习方法研究进展［J］. 航空学报， 2024， 45（6）： 028822.
	LUO X D， WU Y Q， CHEN J L. Research progress on deep learning methods for object detection and semantic segmentation in UAV aerial images［J］. Acta Aeronautica et Astronautica Sinica， 2024， 45（6）： 028822 （in Chinese）.
[3]	LIU W， ANGUELOV D， ERHAN D， et al. SSD： Single shot MultiBox detector［M］∥Computer Vision-ECCV 2016. Cham： Springer International Publishing， 2016： 21-37.
[4]	LIN T Y， GOYAL P， GIRSHICK R， et al. Focal loss for dense object detection［C］∥2017 IEEE International Conference on Computer Vision （ICCV）. Piscataway： IEEE Press， 2017： 2999-3007.
[5]	WANG C Y， BOCHKOVSKIY A， LIAO H M. YOLOv7： trainable bag-of-freebies sets new state-of-the-art for real-time object detectors［C］∥2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2023： 7464-7475.
[6]	陈奎，贾立娇，刘晓，等. 基于多尺度特征融合的绝缘子缺陷程度检测［J］. 高电压技术， 2024， 50（5）： 1889-1899.
	CHEN K， JIA L J， LIU X， et al. Insulator defect degree detection based on multi-scale feature fusion［J］. High Voltage Engineering， 2024， 50（5）： 1889-1899 （in Chinese）.
[7]	王韵琳，冯天波，孙宁，等. 融合注意力与多尺度特征的电力绝缘子缺陷检测方法［J］. 高电压技术， 2024， 50（5）： 1933-1942.
	WANG Y L， FENG T B， SUN N， et al. Defect detection method for power insulators based on attention and multi-scale context information［J］. High Voltage Engineering， 2024， 50（5）： 1933-1942 （in Chinese）.
[8]	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.
[9]	HE K M， ZHANG X Y， REN S Q， et al. Spatial pyramid pooling in deep convolutional networks for visual recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2015， 37（9）： 1904-1916.
[10]	李鑫，刘帅男，杨桢，等. 基于改进Cascade R-CNN的输电线路多目标检测［J］. 电子测量与仪器学报， 2021， 35（10）： 24-32.
	LI X， LIU S N， YANG Z， et al. Multi-target detection of transmission lines based on improved cascade R-CNN［J］. Journal of Electronic Measurement and Instrumentation， 2021， 35（10）： 24-32 （in Chinese）.
[11]	MA J Q， SHAO W Y， YE H， et al. Arbitrary-oriented scene text detection via rotation proposals［J］. IEEE Transactions on Multimedia， 2018， 20（11）： 3111-3122.
[12]	赵其昌，吴一全，苑玉彬. 光学遥感图像舰船目标检测与识别方法研究进展［J］. 航空学报， 2024， 45（8）： 029025.
	ZHAO Q C， WU Y Q， YUAN Y B. Progress of ship detection and recognition methods in optical remote sensing images［J］. Acta Aeronautica et Astronautica Sinica， 2024， 45（8）： 029025 （in Chinese）.
[13]	DING J， XUE N， LONG Y， et al. Learning RoI transformer for oriented object detection in aerial images［C］∥2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2019： 2849-2858.
[14]	XIE X， CHENG G， WANG J， et al. Oriented R-CNN for object detection［C］∥2021 IEEE/CVF International Conference on Computer Vision （ICCV）. Piscataway： IEEE Press， 2021： 3520-3529.
[15]	YANG X， YAN J， FENG Z， et al. R 3 D e t ： Refined single-stage detector with feature refinement for rotating object［C］∥Proceedings of the AAAI Conference on Artificial Intelligence， 2021， 35（4）， 3163-3171.
[16]	奉志强，谢志军，包正伟，等. 基于改进YOLOv5的无人机实时密集小目标检测算法［J］. 航空学报， 2023， 44（7）： 327106.
	FENG Z Q， XIE Z J， BAO Z W， et al. Real-time dense small object detection algorithm for UAV based on improved YOLOv5［J］. Acta Aeronautica et Astronautica Sinica， 2023， 44（7）： 327106 （in Chinese）.
[17]	QURESHI M F， MUSHTAQ Z， REHMAN M Z U， et al. E2CNN： An efficient concatenated CNN for classification of surface EMG extracted from upper limb［J］. IEEE Sensors Journal， 2023， 23（8）： 8989-8996.
[18]	HAN J， DING J， XUE N， et al. ReDet： A rotation-equivariant detector for aerial object detection［C］∥2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2021： 2786-2795.
[19]	WOO S， PARK J， LEE J Y， et al. CBAM： Convolutional block attention module［M］∥Computer Vision-ECCV 2018. Cham： Springer International Publishing， 2018： 3-19.
[20]	QIAO S Y， CHEN L C， YUILLE A. DetectoRS： Detecting objects with recursive feature pyramid and switchable atrous convolution［C］∥2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2021： 10208-10219.
[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 （CVPR）. Piscataway： IEEE Press， 2017： 936-944.
[22]	RAMACHANDRAN P， ZOPH B， LE Q V. Searching for activation functions［DB/OL］. arXiv preprint： 1710.05941， 2017.
[23]	GLOROT X， BORDES A， BENGIO Y. Deep sparse rectifier neural networks［C］∥Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics （AISTATS）. Fort Lauderdale： JMLR Workshop and Conference Proceedings， 2011： 315-323.
[24]	YANG X， ZHOU Y， ZHANG G F， et al. The KFIoU loss for rotated object detection［DB/OL］. arXiv preprint： 2201.12558， 2022.
[25]	YANG X， YANG X J， YANG J R， et al. Learning high-precision bounding box for rotated object detection via Kullback-Leibler divergence［DB/OL］. arXiv preprint： 2106.01883， 2021.
[26]	TAO X， ZHANG D P， WANG Z H， et al. Detection of power line insulator defects using aerial images analyzed with convolutional neural networks［J］. IEEE Transactions on Systems， Man， and Cybernetics： Systems， 2020， 50（4）： 1486-1498.
[27]	VIEIRA E SILVA A L B， DE CASTRO FELIX H， SIMÕES F P M， et al. InsPLAD： A dataset and benchmark for power line asset inspection in UAV images［J］. International Journal of Remote Sensing， 2023， 44（23）： 7294-7320.
[28]	SELVARAJU R R， COGSWELL M， DAS A， et al. Grad-CAM： Visual explanations from deep networks via gradient-based localization［C］∥2017 IEEE International Conference on Computer Vision （ICCV）. Piscataway： IEEE Press， 2017： 618-626.
[29]	ZHANG J Q， LEI J， XIE W Y， et al. SuperYOLO： Super resolution assisted object detection in multimodal remote sensing imagery［J］. IEEE Transactions on Geoscience and Remote Sensing， 2023， 61： 5605415.
[30]	LI Y X， HOU Q B， ZHENG Z H， et al. Large selective kernel network for remote sensing object detection［C］∥2023 IEEE/CVF International Conference on Computer Vision （ICCV）. Piscataway： IEEE Press， 2023： 16748-16759.
[31]	WANG G， CHEN Y F， AN P， et al. UAV-YOLOv8： A small-object-detection model based on improved YOLOv8 for UAV aerial photography scenarios［J］. Sensors， 2023， 23（16）： 7190.

类别		标签	来源	样本数	总数
正常绝缘子	正常复合绝缘子类型1	Polymer_insulator	CPLID	1 104	4 215
	正常玻璃绝缘子	Glass_insulator	InsPLAD	1 133
	正常陶瓷绝缘子	Ceramic_insulator	电网数据集	968
	正常复合绝缘子类型2	Composite_insultaor	电网数据集	1 010
缺陷绝缘子	缺陷复合绝缘子类型1	Defective_polymer_insulator	CPLID	1 472	2 707
缺陷绝缘子	缺陷玻璃绝缘子	Defective_glass_insulator	InsPLAD	1 235	2 707
缺陷	复合绝缘子类型1缺陷	Polymer_defect	CPLID	1 472	2 976
缺陷	玻璃绝缘子缺陷	Glass_defect	InsPLAD	1 504	2 976

序号	ReDet	M1	M2	M3	mAP/%	Params/MB	FPS
1	√				92.03	31.64	23.63
2	√	√			92.34	31.92	23.44
3	√	√	√		93.27	32.49	23.01
4	√	√	√	√	95.04	34.57	22.38

Oriented detection algorithm for insulator and their defects from aerial images based on improved ReDet

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 16

References 31

Related Articles 4

Recommended Articles

Metrics

Comments

[1]	Yiquan WU, Kang TONG. Research advances on deep learning-based small object detection in UAV aerial images [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(3): 30848-030848.
[2]	Junyu LI, Qiankun LIU, Ying FU. Infrared small object detection based on attention mechanism [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(14): 628959-628959.
[3]	Zhiqiang FENG, Zhijun XIE, Zhengwei BAO, Kewei CHEN. Real⁃time dense small object detection algorithm for UAV based on improved YOLOv5 [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(7): 327106-327106.
[4]	Zihao LI, Zhengping WANG, Yuntao HE. Aerial-photography dense small target detection algorithm based on adaptive cooperative attention mechanism [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(13): 327944-327944.