An insulator self-blast detection method based on YOLOv4 with aerial images

Due to long-term exposure to the natural environment, insulators are prone to self-blast, threatening the safety and reliability of transmission lines. Because of the different sizes of the insulator self-blast area and the complicated background, it is inevitable that the missed and false detection...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Hui He, Xile Huang, Yuxuan Song, Zheng Zhang, Meng Wang, Bo Chen, Guangwei Yan
Formato: article
Lenguaje:EN
Publicado: Elsevier 2022
Materias:
Acceso en línea:https://doaj.org/article/c3583fa8dd9048258df46cb8ca435d86
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:c3583fa8dd9048258df46cb8ca435d86
record_format dspace
spelling oai:doaj.org-article:c3583fa8dd9048258df46cb8ca435d862021-12-04T04:35:03ZAn insulator self-blast detection method based on YOLOv4 with aerial images2352-484710.1016/j.egyr.2021.11.115https://doaj.org/article/c3583fa8dd9048258df46cb8ca435d862022-04-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2352484721012622https://doaj.org/toc/2352-4847Due to long-term exposure to the natural environment, insulators are prone to self-blast, threatening the safety and reliability of transmission lines. Because of the different sizes of the insulator self-blast area and the complicated background, it is inevitable that the missed and false detections occur. To solve this problem, this paper proposes a deep neural network called Mina-Net (Multi-Layer INformation Fusion and Attention Mechanism Network). Mina-Net is based on YOLOv4. First, the shallow feature map with more detailed texture information is fused into the feature pyramid. Then, an improved SENet is applied to recalibrate the features of different levels in the channel direction. The experimental results on the actual dataset show that Mina-Net has increased the average precision by 4.78% compared with the YOLOv4.Hui HeXile HuangYuxuan SongZheng ZhangMeng WangBo ChenGuangwei YanElsevierarticleInsulator self-blastYOLOv4Feature fusionAttention mechanismSENetAerial imagesElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENEnergy Reports, Vol 8, Iss , Pp 448-454 (2022)
institution DOAJ
collection DOAJ
language EN
topic Insulator self-blast
YOLOv4
Feature fusion
Attention mechanism
SENet
Aerial images
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
spellingShingle Insulator self-blast
YOLOv4
Feature fusion
Attention mechanism
SENet
Aerial images
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
Hui He
Xile Huang
Yuxuan Song
Zheng Zhang
Meng Wang
Bo Chen
Guangwei Yan
An insulator self-blast detection method based on YOLOv4 with aerial images
description Due to long-term exposure to the natural environment, insulators are prone to self-blast, threatening the safety and reliability of transmission lines. Because of the different sizes of the insulator self-blast area and the complicated background, it is inevitable that the missed and false detections occur. To solve this problem, this paper proposes a deep neural network called Mina-Net (Multi-Layer INformation Fusion and Attention Mechanism Network). Mina-Net is based on YOLOv4. First, the shallow feature map with more detailed texture information is fused into the feature pyramid. Then, an improved SENet is applied to recalibrate the features of different levels in the channel direction. The experimental results on the actual dataset show that Mina-Net has increased the average precision by 4.78% compared with the YOLOv4.
format article
author Hui He
Xile Huang
Yuxuan Song
Zheng Zhang
Meng Wang
Bo Chen
Guangwei Yan
author_facet Hui He
Xile Huang
Yuxuan Song
Zheng Zhang
Meng Wang
Bo Chen
Guangwei Yan
author_sort Hui He
title An insulator self-blast detection method based on YOLOv4 with aerial images
title_short An insulator self-blast detection method based on YOLOv4 with aerial images
title_full An insulator self-blast detection method based on YOLOv4 with aerial images
title_fullStr An insulator self-blast detection method based on YOLOv4 with aerial images
title_full_unstemmed An insulator self-blast detection method based on YOLOv4 with aerial images
title_sort insulator self-blast detection method based on yolov4 with aerial images
publisher Elsevier
publishDate 2022
url https://doaj.org/article/c3583fa8dd9048258df46cb8ca435d86
work_keys_str_mv AT huihe aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT xilehuang aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT yuxuansong aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT zhengzhang aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT mengwang aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT bochen aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT guangweiyan aninsulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT huihe insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT xilehuang insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT yuxuansong insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT zhengzhang insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT mengwang insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT bochen insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
AT guangweiyan insulatorselfblastdetectionmethodbasedonyolov4withaerialimages
_version_ 1718372993339490304