Infrared Image Detection of Substation Insulators Using an Improved Fusion Single Shot Multibox Detector

Zheng, Hanbo; Sun, Yonghui; Liu, Xinghua; Djike, Calvin Laurent Tcheteu; Li, Jin‐Heng; Liu, Yang; Ma, Jun; Xu, Kai; Zhang, Chaohai

doi:10.1109/tpwrd.2020.3038880

Cited by 50 publications

(19 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As opposed to traditional detection methods like region seeds, edge detection, YOLOX boasts self-learning capability and a higher degree of accuracy. Compared with other deep learning methods such as R-CNN, Fast R-CNN, Faster R-CNN, SSD [20][21][22], the advantages of YOLOX are that balance speed and accuracy, and focus on solving the problem of small object detection. Therefore, the YOLOX deep learning network is applied to identify and classify bird pecking damage of composite insulator strings.…”

Section: Methodsmentioning

confidence: 99%

Bird pecking damage risk assessment of UHV transmission line composite insulators based on deep learning

Zhang,

Sun,

et al. 2023

IET Generation Trans & Dist

View full text Add to dashboard Cite

Here, a method for assessing the risk of bird pecking damage of composite insulators in ultra high voltage (UHV) lines is proposed using electric field (E-field) simulation and deep learning. The distribution of E-field on composite insulators is analysed via numerical simulation for different damage locations and damage sizes. Then, using the defective images of composite insulator strings captured in real inspection environments, the corresponding annotation image data set is constructed according to the finite element calculation results of E-field under different damage conditions, and through the application of the YOLOX deep learning neural network, the risk assessment of the damage caused by bird pecking of insulators for UHV lines is conducted. Furthermore, to prevent overfitting of YOLOX with small-scale images, transfer learning, as well as data enhancement, are applied to the YOLOX training process. According to the results, the mean average precision (mAP) of the model is 0.79, indicating that it is capable of high accuracy recognition, provides guidance for operation and maintenance personnel.

show abstract

Section: Methodsmentioning

confidence: 99%

Bird pecking damage risk assessment of UHV transmission line composite insulators based on deep learning

Zhang,

Sun,

et al. 2023

IET Generation Trans & Dist

View full text Add to dashboard Cite

show abstract

“…Horizontal box [20] The detection accuracy and recall rate are as high as 0.91 and 0.96 on the self-built dataset, which meets the detection accuracy requirements of related scenes The detection speed is 0.36s per picture, which can be applied to on-site detection requirements where real-time requirements are not strong Horizontal box [21] Compared with the original model, the detection accuracy on the self-built dataset has been greatly improved, and its average accuracy index has reached 0.82 Two-stage detection method, the detection speed is slow Horizontal box [2] In the self-built dataset, the detection accuracy is slightly higher than the original model, and the average accuracy is increased by 0.72…”

Section: Related Workmentioning

confidence: 99%

“…However, the detection speed as an important evaluation index has not been compared and tested. Reference [2] proposed an improved FSSD detection model for infrared images of substation insulators. However, this model cannot improve the detection accuracy of the model while maintaining a high detection speed.…”

Section: Related Workmentioning

confidence: 99%

Arbitrary-Oriented Detection of Insulators in Thermal Imagery via Rotation Region Network

Zheng

Liu

Sun

et al. 2022

IEEE Trans. Ind. Inf.

Self Cite

View full text Add to dashboard Cite

The precise location of insulators in infrared images is of great significance for insulator condition monitoring and fault diagnosis. Due to the characteristics of insulators themselves and the use of handheld infrared cameras, insulators usually appear in infrared images with different aspect ratios and main axis orientations. Therefore, it is very important and necessary to make full use of the prior knowledge of the insulator itself to accurately locate it. However, most of the existing methods use axial horizontal detection boxes to detect insulators, which cannot take into account the characteristics of the insulator well. When there are large overlapping areas of two horizontal detection boxes, the non-maximum suppression algorithm may lead to missed detection of the object. In order to further improve the accuracy of the detection algorithm, this paper makes full use of the prior features carried by the insulator itself, and optimizes Faster R-CNN from five aspects: rectangular box representation, feature extraction, candidate box generation, anchor design, and feature alignment. An oriented detection model for infrared images of insulators is constructed. Comparative experiments with a variety of mainstream detection methods were carried out on the constructed infrared dataset. The results show that the proposed method is superior to other models in detection accuracy. When the intersection and union ratio is 0.5, the average precision reaches 95.08%. In addition, it can also effectively predict the shape and angle information of insulators in complex scenes, laying a beneficial foundation for subsequent diagnosis automation tasks.

show abstract

“…Moreover, to enhance recognition accuracy, various methods integrating visible light have been explored. On one front, several infrared object detection algorithms have improved performance through transfer learning [9][10][11][12]. Concurrently, multi-spectral detection algorithms combining visible and infrared imagery [13][14][15][16] have demonstrated noteworthy results on datasets like KAIST [17].…”

Section: Object Detection For Infraredmentioning

confidence: 99%

Infrared multi‐target detection and tracking in dense urban traffic scenes

Zha,

Luo,

2024

IET Image Processing

View full text Add to dashboard Cite

Infrared object detection and tracking in dense urban traffic remain a challenge due to factors such as low contrast, small intra‐class differences, and frequent false positives and negatives. To overcome these, the authors introduce YOLO‐IR, an algorithm based on the enhanced YOLOv8s, and YOLO‐DeepOC‐IR, a comprehensive infrared multi‐object tracking method for urban traffic, integrating both detection and tracking. During preprocessing, three infrared image enhancement techniques, local contrast multi‐scale enhancement, non‐local means, and contrast limited adaptive histogram equalization, are applied for better reliability in dense scenes. To further improve the performance, the original YOLOv8s backbone is replaced with MobileVITv3 to enhance detection accuracy and robustness. This infrared feature extraction module, incorporated into the detector, combines canny edge detection, Gabor filtering, and open operation layers, significantly boosting object detection in infrared imagery. The tracker's feature processing capabilities are improved using the learned arrangements of three patch codes descriptor and locality‐sensitive hashing for feature extraction and matching. Experimental results on FLIR ADAS v2 and InfiRay datasets indicate superior performance of this method, achieving 78.6% mAP and 151.1 FPS in detection, and up to 80.8% moving object tracking accuracy, 78.6% identification F1 score, and 62.1% higher order tracking accuracy in multi‐object tracking.

show abstract

Infrared Image Detection of Substation Insulators Using an Improved Fusion Single Shot Multibox Detector

Cited by 50 publications

References 32 publications

Bird pecking damage risk assessment of UHV transmission line composite insulators based on deep learning

Bird pecking damage risk assessment of UHV transmission line composite insulators based on deep learning

Arbitrary-Oriented Detection of Insulators in Thermal Imagery via Rotation Region Network

Infrared multi‐target detection and tracking in dense urban traffic scenes

Contact Info

Product

Resources

About