A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors

Wildfire detection in power transmission corridors is essential for providing timely warnings and ensuring the safe and stable operation of power lines. However, this task faces significant challenges due to the large number of smoke-like samples in the background, the complex and diverse target mor...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaole Wang, Bo Wang, Peng Luo, Leixiong Wang, Yurou Wu
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/13/3882
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849319895693524992
author Xiaole Wang
Bo Wang
Peng Luo
Leixiong Wang
Yurou Wu
author_facet Xiaole Wang
Bo Wang
Peng Luo
Leixiong Wang
Yurou Wu
author_sort Xiaole Wang
collection DOAJ
description Wildfire detection in power transmission corridors is essential for providing timely warnings and ensuring the safe and stable operation of power lines. However, this task faces significant challenges due to the large number of smoke-like samples in the background, the complex and diverse target morphologies, and the difficulty of detecting small-scale smoke and flame objects. To address these issues, this paper proposed an improved Oriented R-CNN model enhanced with metric learning for wildfire detection in power transmission corridors. Specifically, a multi-center metric loss (MCM-Loss) module based on metric learning was introduced to enhance the model’s ability to differentiate features of similar targets, thereby improving the recognition accuracy in the presence of interference. Experimental results showed that the introduction of the MCM-Loss module increased the average precision (AP) for smoke targets by 2.7%. In addition, the group convolution-based network ResNeXt was adopted to replace the original backbone network ResNet, broadening the channel dimensions of the feature extraction network and enhancing the model’s capability to detect flame and smoke targets with diverse morphologies. This substitution led to a 0.6% improvement in mean average precision (mAP). Furthermore, an FPN-CARAFE module was designed by incorporating the content-aware up-sampling operator CARAFE, which improved multi-scale feature representation and significantly boosted performance in detecting small targets. In particular, the proposed FPN-CARAFE module improved the AP for fire targets by 8.1%. Experimental results demonstrated that the proposed model achieved superior performance in wildfire detection within power transmission corridors, achieving a mAP of 90.4% on the test dataset—an improvement of 6.4% over the baseline model. Compared with other commonly used object detection algorithms, the model developed in this study exhibited improved detection performance on the test dataset, offering research support for wildfire monitoring in power transmission corridors.
format Article
id doaj-art-cff9dfd4ed764bada7bd38acb58b0b54
institution Kabale University
issn 1424-8220
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-cff9dfd4ed764bada7bd38acb58b0b542025-08-20T03:50:17ZengMDPI AGSensors1424-82202025-06-012513388210.3390/s25133882A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission CorridorsXiaole Wang0Bo Wang1Peng Luo2Leixiong Wang3Yurou Wu4School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, ChinaWuhan Power Supply Company, State Grid Hubei Electric Power Company, Wuhan 430013, ChinaWuhan Power Supply Company, State Grid Hubei Electric Power Company, Wuhan 430013, ChinaWildfire detection in power transmission corridors is essential for providing timely warnings and ensuring the safe and stable operation of power lines. However, this task faces significant challenges due to the large number of smoke-like samples in the background, the complex and diverse target morphologies, and the difficulty of detecting small-scale smoke and flame objects. To address these issues, this paper proposed an improved Oriented R-CNN model enhanced with metric learning for wildfire detection in power transmission corridors. Specifically, a multi-center metric loss (MCM-Loss) module based on metric learning was introduced to enhance the model’s ability to differentiate features of similar targets, thereby improving the recognition accuracy in the presence of interference. Experimental results showed that the introduction of the MCM-Loss module increased the average precision (AP) for smoke targets by 2.7%. In addition, the group convolution-based network ResNeXt was adopted to replace the original backbone network ResNet, broadening the channel dimensions of the feature extraction network and enhancing the model’s capability to detect flame and smoke targets with diverse morphologies. This substitution led to a 0.6% improvement in mean average precision (mAP). Furthermore, an FPN-CARAFE module was designed by incorporating the content-aware up-sampling operator CARAFE, which improved multi-scale feature representation and significantly boosted performance in detecting small targets. In particular, the proposed FPN-CARAFE module improved the AP for fire targets by 8.1%. Experimental results demonstrated that the proposed model achieved superior performance in wildfire detection within power transmission corridors, achieving a mAP of 90.4% on the test dataset—an improvement of 6.4% over the baseline model. Compared with other commonly used object detection algorithms, the model developed in this study exhibited improved detection performance on the test dataset, offering research support for wildfire monitoring in power transmission corridors.https://www.mdpi.com/1424-8220/25/13/3882wildfire detectiontransmission corridorsmetric learninggroup convolutionscontent-aware up-sampling
spellingShingle Xiaole Wang
Bo Wang
Peng Luo
Leixiong Wang
Yurou Wu
A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors
Sensors
wildfire detection
transmission corridors
metric learning
group convolutions
content-aware up-sampling
title A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors
title_full A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors
title_fullStr A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors
title_full_unstemmed A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors
title_short A Metric Learning-Based Improved Oriented R-CNN for Wildfire Detection in Power Transmission Corridors
title_sort metric learning based improved oriented r cnn for wildfire detection in power transmission corridors
topic wildfire detection
transmission corridors
metric learning
group convolutions
content-aware up-sampling
url https://www.mdpi.com/1424-8220/25/13/3882
work_keys_str_mv AT xiaolewang ametriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT bowang ametriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT pengluo ametriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT leixiongwang ametriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT yurouwu ametriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT xiaolewang metriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT bowang metriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT pengluo metriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT leixiongwang metriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors
AT yurouwu metriclearningbasedimprovedorientedrcnnforwildfiredetectioninpowertransmissioncorridors