A comprehensive survey of research towards AI-enabled unmanned aerial systems in pre-, active-, and post-wildfire management

Information Fusion, Journal Year: 2024, Volume and Issue: 108, P. 102369 - 102369

Published: March 22, 2024

Wildfires have emerged as one of the most destructive natural disasters worldwide, causing catastrophic losses. These losses underscored urgent need to improve public knowledge and advance existing techniques in wildfire management. Recently, use Artificial Intelligence (AI) wildfires, propelled by integration Unmanned Aerial Vehicles (UAVs) deep learning models, has created an unprecedented momentum implement develop more effective Although survey papers explored learning-based approaches wildfire, drone disaster management, risk assessment, a comprehensive review emphasizing application AI-enabled UAV systems investigating role methods throughout overall workflow multi-stage including pre-fire (e.g., vision-based vegetation fuel measurement), active-fire fire growth modeling), post-fire tasks evacuation planning) is notably lacking. This synthesizes integrates state-of-the-science reviews research at nexus observations modeling, AI, UAVs - topics forefront advances elucidating AI performing monitoring actuation from pre-fire, through stage, To this aim, we provide extensive analysis remote sensing with particular focus on advancements, device specifications, sensor technologies relevant We also examine management approaches, monitoring, prevention strategies, well planning, damage operation strategies. Additionally, summarize wide range computer vision emphasis Machine Learning (ML), Reinforcement (RL), Deep (DL) algorithms for classification, segmentation, detection, tasks. Ultimately, underscore substantial advancement modeling cutting-edge UAV-based data, providing novel insights enhanced predictive capabilities understand dynamic behavior.

Language: Английский

---

Channel Attention for Fire and Smoke Detection: Impact of Augmentation, Color Spaces, and Adversarial Attacks

Sensors, Journal Year: 2025, Volume and Issue: 25(4), P. 1140 - 1140

Published: Feb. 13, 2025

The prevalence of wildfires presents significant challenges for fire detection systems, particularly in differentiating from complex backgrounds and maintaining reliability under diverse environmental conditions. It is crucial to address these developing sustainable effective systems. In this paper: (i) we introduce a channel-wise attention-based architecture, achieving 95% accuracy demonstrating an improved focus on flame-specific features critical distinguishing backgrounds. Through ablation studies, demonstrate that our attention mechanism provides 3-5% improvement over the baseline state-of-the-art models; (ii) evaluate impact augmentation detection, performance across varied conditions; (iii) comprehensive evaluation color spaces including RGB, Grayscale, HSV, YCbCr analyze reliability; (iv) assessment model vulnerabilities where Fast Gradient Sign Method (FGSM) perturbations significantly performance, reducing 41%. Using Local Interpretable Model-Agnostic Explanations (LIME) visualization techniques, provide insights into decision-making processes both standard adversarial conditions, highlighting important considerations applications.

Language: Английский

---

Navigating the landscape of global sustainable livelihood research: past insights and future trajectory

Environmental Science and Pollution Research, Journal Year: 2023, Volume and Issue: 30(46), P. 103291 - 103312

Published: Sept. 9, 2023

Language: Английский

---

PyroChroma: advancing wildfire detection with multispectral imaging and explainability insights

Multimedia Tools and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 25, 2025

Language: Английский

---

Estimation of Fire Counts and Fire Radiative Power Using Satellite Optical and Microwave Vegetation Indices With Random Forest Method

Journal of Geophysical Research Atmospheres, Journal Year: 2025, Volume and Issue: 130(3)

Published: Jan. 29, 2025

Abstract The satellite microwave emissivity difference vegetation index (EDVI) has been used in previous studies to estimate FCs and FRP using traditional multivariate linear regression models. However, the nonlinear effects contributions of numerous factors that affect forest fires cannot be disentangled by this model. Using random (RF) model, study utilized multiple EDVIs optical normalized (NDVI) as key fuel properties resolve physical driving mechanisms daily over East Asia. results showed estimated were good agreement with observations, a spatial R 0.59 for 0.63 temporal 0.80 0.81 FRP. integration NDVI into RF model was found improve performance generate overall lower systematic errors than without variables. Model better In addition, greater importance NDVI. This largely due their resolution allowed capture fire dynamics time. combination observations shows great potential FC estimations global danger assessment.

Language: Английский

---

Wildfire Impacts Pinus tabulaeformis Forests on Soil Properties, Actinobacteriota, and Enzyme Activity in Northern China: Direct Effects or Mutual Interactions?

Forests, Journal Year: 2025, Volume and Issue: 16(2), P. 344 - 344

Published: Feb. 14, 2025

Wildfires are significant disturbances that reshape soil ecosystems, impacting properties, microbial communities, and enzyme activities. In Pinus tabulaeformis forests in northern China, the effects of wildfire on health, particularly Actinobacteriota enzymatic functions, remain poorly understood. This study investigates both direct indirect fire severity these factors examines how fire-induced changes properties mediate responses. Our findings show significantly alters chemical including an increase pH a reduction organic carbon water content, under high severities. These directly impact with showing resilience light moderate intensities but declining severity, especially subsoil layers. Soil enzymes, such as urease protease, played crucial role mitigating negative impacts nutrient cycling. Their activity promoted availability, aiding ecosystem recovery, even intensity reduced overall fertility. Structural Equation Modeling (SEM) further revealed relationships between Actinobacteriota, shaped by thermal complex interactions mediated moisture levels. underscores importance considering mutual activities post-fire recovery. The highlight while high-severity fires disrupt health dynamics, enzymes can help regulate enhancing cycling supporting stability. insights contribute to better understanding wildfire-induced degradation provide actionable strategies for restoration management fire-prone ecosystems.

Language: Английский

---

Interpretable deep one-class model for forest fire detection

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 127657 - 127657

Published: April 1, 2025

---

Forest fire regimes in the Northwestern Himalayas: unravelling microlevel impact of topography, weather, and human activity on fire behaviour

International Journal of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 26

Published: April 20, 2025

Language: Английский

---

Carbon and nitrogen stock in soils of subtropical urban forests: Isotopic δ13C and δ15N indicators for nature-based solutions in a megacity

Ecological Indicators, Journal Year: 2024, Volume and Issue: 160, P. 111743 - 111743

Published: Feb. 21, 2024

Urban forest soils play a pivotal role in enhancing the environmental sustainability of cities, contributing to various natural processes, including plant–microbe interactions, microbial activity, and decomposition organic matter. Consequently, urban emerge as effective NBS, underscoring their potential mitigate challenges foster sustainable ecosystems. In these sense, this manuscript aimed at evaluating how soil attributes forests São Paulo, Brazil, with different adjacent land uses, influence capacity store excess C N from anthropogenic emissions, making ecosystem an important reservoir emissions. Three hundred samples were collected surface depth 50 cm. All analyzed for content (and stable isotopes). addition, granulometric tests also carried out classify soils. It was found that most central fragment has highest contents all depths, probably due association physical aspects texture. For layers, sample, only one clay soil, presented approximately twice many elements when compared other sites. general, stocks isotopes, δ13C δ15N, respectively) varied significantly located center-periphery direction (%N - F = 24.58, p < 0.05; %C 22.48, δ15N 4.27, 19.8, C/N 14.56, 0.05). This more higher vehicle emissions showed greater atmospheric neutralizing efficiency than fragments. together C:N ratio indicated biogeochemical cycling, through decomposition, More peripheral fragments high cycling along profiles, while superficial layer, highly efficient. These shed light results integrating NbS principles into strategic planning city-level climate policies can bolster effectiveness green areas. The integration not promotes carbon sequestration efficient nutrient but fosters practices, resilient landscape.

Language: Английский

---

Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net

Technologies, Journal Year: 2024, Volume and Issue: 12(9), P. 160 - 160

Published: Sept. 12, 2024

Forest ecosystems are critical components of Earth’s biodiversity and play vital roles in climate regulation carbon sequestration. They face increasing threats from deforestation, wildfires, other anthropogenic activities. Timely detection monitoring changes forest landscapes pose significant challenges for government agencies. To address these challenges, we propose a novel pipeline by refining the U-Net design, including employing two different schemata early fusion networks Siam network architecture capable processing RGB images specifically designed to identify high-risk areas through change across time frames same location. It annotates ground truth maps such using an encoder–decoder approach with help enhanced feature learning attention mechanism. Our proposed pipeline, integrated ResNeSt blocks SE techniques, achieved impressive results our newly created cover dataset. The evaluation metrics reveal Dice score 39.03%, kappa 35.13%, F1-score 42.84%, overall accuracy 94.37%. Notably, significantly outperformed multitasking model approaches ONERA dataset, boasting precision 53.32%, 59.97%, 97.82%. Furthermore, it surpassed models HRSCD even without utilizing land maps, achieving 44.62%, 11.97%, 98.44%. Although had lower than methods, performance highlight its effectiveness timely landscape monitoring, advancing deep techniques this field.

Language: Английский

---