Pyramid Attention Based Early Forest Fire Detection Using UAV Imagery

Zhang, Yongtuo; Chen, Sheng; Wang, Weiyi; Zhang, Wei; Zhang, Li

doi:10.1088/1742-6596/2363/1/012021

Cited by 4 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Forest fires, increasingly common today, require rigorous scientific exploration of early detection methods. This is crucial to proactively control their spread and mitigate environmental harm [ 1 , 2 ]. The advent of the Industrial Revolution marked a pivotal moment in Earth’s history, initiating a series of events that significantly altered global fire patterns.…”

Section: Introductionmentioning

confidence: 99%

Development of a deep learning-based surveillance system for forest fire detection and monitoring using UAV

SHAMTA,

Demir

2024

PLoS ONE

View full text Add to dashboard Cite

This study presents a surveillance system developed for early detection of forest fires. Deep learning is utilized for aerial detection of fires using images obtained from a camera mounted on a designed four-rotor Unmanned Aerial Vehicle (UAV). The object detection performance of YOLOv8 and YOLOv5 was examined for identifying forest fires, and a CNN-RCNN network was constructed to classify images as containing fire or not. Additionally, this classification approach was compared with the YOLOv8 classification. Onboard NVIDIA Jetson Nano, an embedded artificial intelligence computer, is used as hardware for real-time forest fire detection. Also, a ground station interface was developed to receive and display fire-related data. Thus, access to fire images and coordinate information was provided for targeted intervention in case of a fire. The UAV autonomously monitored the designated area and captured images continuously. Embedded deep learning algorithms on the Nano board enable the UAV to detect forest fires within its operational area. The detection methods produced the following results: 96% accuracy for YOLOv8 classification, 89% accuracy for YOLOv8n object detection, 96% accuracy for CNN-RCNN classification, and 89% accuracy for YOLOv5n object detection.

show abstract

Section: Introductionmentioning

confidence: 99%

Development of a deep learning-based surveillance system for forest fire detection and monitoring using UAV

SHAMTA,

Demir

2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Fortunately, the development of UAVs has attained a more mature stage, while video, image, and image processing technologies continue to advance [19,20]. UAV-based image analysis for forest fire smoke detection systems can effectively conduct inspections of mountains and forests [21][22][23]. The UAV's camera captures images or videos of the mountains and forests, eliminating the cumbersome process of camera deployment and reducing the allocation of manpower, material resources, and financial resources.…”

Section: Introductionmentioning

confidence: 99%

Efficient Detection of Forest Fire Smoke in UAV Aerial Imagery Based on an Improved Yolov5 Model and Transfer Learning

Yang,

Wang,

Wang

2023

Remote Sensing

View full text Add to dashboard Cite

Forest fires pose severe challenges to forest management because of their unpredictability, extensive harm, broad impact, and rescue complexities. Early smoke detection is pivotal for prompt intervention and damage mitigation. Combining deep learning techniques with UAV imagery holds potential in advancing forest fire smoke recognition. However, issues arise when using UAV-derived images, especially in detecting miniature smoke patches, complicating effective feature discernment. Common deep learning approaches for forest fire detection also grapple with limitations due to sparse datasets. To counter these challenges, we introduce a refined UAV-centric forest fire smoke detection approach utilizing YOLOv5. We first enhance anchor box clustering through K-means++ to boost the classification precision and then augment the YOLOv5 architecture by integrating a novel partial convolution (PConv) to trim down model parameters and elevate processing speed. A unique detection head is also incorporated to the model to better detect diminutive smoke traces. A coordinate attention module is embedded within YOLOv5, enabling precise smoke target location and fine-grained feature extraction amidst complex settings. Given the scarcity of forest fire smoke datasets, we employ transfer learning for model training. The experimental results demonstrate that our proposed method achieves 96% AP50 and 57.3% AP50:95 on a customized dataset, outperforming other state-of-the-art one-stage object detectors while maintaining real-time performance.

show abstract

Exploiting drone images for forest fire detection using metaheuristics with deep learning model

Rajalakshmi,

Sellam,

Kannan

et al. 2023

IFS

View full text Add to dashboard Cite

Forest fires are a global natural calamity causing significant economic damage and loss of lives. Professionals forecast that forest fires would raise in the future because of climate change. Early prediction and identification of fire spread would enhance firefighting and reduce affected zones. Several systems have been advanced to detect fire. Recently, Unmanned Aerial Vehicles (UAVs) can be used for forest fire detection due to their ability, high flexibility, and inexpensive to cover vast areas. But still, they are limited by difficulties like image degradation, small fire size, and background complexity. This study develops an automated Forest Fire Detection using Metaheuristics with Deep Learning (FFDMDL-DI) model. The presented FFDMDL-DI technique exploits the DL concepts on drone images to identify the occurrence of fire. To accomplish this, the FFDMDL-DI technique makes use of the Capsule Network (CapNet) model for feature extraction purposes with a biogeography-based optimization (BBO) algorithm-based hyperparameter optimizer. For accurate forest fire detection, the FFDMDL-DI technique uses a unified deep neural network (DNN) model. Finally, the tree growth optimization (TGO) technique is utilized for the parameter adjustment of the DNN method. To depict the enhanced detection efficiency of the FFDMDL-DI approach, a series of simulations were performed on the FLAME dataset, comprising 6000 samples. The experimental results stated the improvements in the FFDMDL-DI method over other DL models with maximum accuracy of 99.76%.

show abstract

Pyramid Attention Based Early Forest Fire Detection Using UAV Imagery

Cited by 4 publications

References 11 publications

Development of a deep learning-based surveillance system for forest fire detection and monitoring using UAV

Development of a deep learning-based surveillance system for forest fire detection and monitoring using UAV

Efficient Detection of Forest Fire Smoke in UAV Aerial Imagery Based on an Improved Yolov5 Model and Transfer Learning

Exploiting drone images for forest fire detection using metaheuristics with deep learning model

Contact Info

Product

Resources

About