[go: up one dir, main page]

 
 

Advanced Approaches to Wildfire Detection, Monitoring and Surveillance

A special issue of Fire (ISSN 2571-6255).

Deadline for manuscript submissions: 31 October 2024 | Viewed by 2599

Special Issue Editors


E-Mail Website
Guest Editor
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Split, Croatia
Interests: applicatiopn of ICT in environment protection; artificial intelligence; computational inteligence; advanced system modelling and control

E-Mail Website
Guest Editor
Earth Observation and Satellite Image Applications Laboratory (EOSIAL), School of Aerospace Engineering (SIA), Sapienza University of Rome, Via Salaria, Roma, Italy
Interests: land degradation; vegetation mapping; satellite image analysis
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Wildfires are natural phenomena with devastating effects on nature and human properties. Many efforts in fire prevention and protection aim to reduce not only the number of fires, but also the extent of fire damage. It is well-known that early wildfire detection and quick, appropriate interventions are the most important measures for minimizing wildfire damage. Once a wildfire has expanded, it becomes very difficult to control and extinguish. Therefore, over the last couple of decades, there have been many efforts to develop an efficient wildfire monitoring system capable of detecting wildfires at initial stages. The rapid development of artificial neural systems and deep learning methods has shifted research from model-based to data-driven and learning-based approaches, offering a new generation of quite successful methods for the early detection of wildfires.

We are pleased to invite you to submit a paper to the Special Issue of the journal Fire, entitled “Advanced Approaches to Wildfire Detection, Monitoring and Surveillance”. This Special Issue aims to bring together and present recent advanced approaches for wildfire smoke and flame detection, as well as advanced systems for wildfire monitoring and surveillance in various natural environments, from inaccessible forest areas to wildland–urban interfaces (WUIs). After detection, the fire must be continuously monitored. Therefore, potential topics for this Special Issue include remote presence systems at the fire scene, including the application of virtual reality (VR) and augmented reality (AR) techniques, as well as research on post-fire analysis and the assessment of both the burned area and the damage caused by the fire.

In this Special Issue, original research articles and reviews are welcome. Research areas may include (but are not limited to) the following:

  • Wildfire detection, monitoring and surveillance in inaccessible forest areas;
  • Wildfire detection, monitoring and surveillance in wildland–urban interfaces (WUIs);
  • Fire detection and monitoring in open-space areas, such as disposal sites, garbage dumps, warehouses, marinas, dry-land marinas, harbors, parking places, etc.;
  • Wildfire video-based detection, monitoring and surveillance, including model-based, data-driven and learning-based approaches;
  • Wildfire detection and monitoring via satellite and aerial remote sensing systems;
  • Wildfire detection validation and testing;
  • Preparation and presentation of databases for wildfire detection training, validation and testing;
  • Post-fire monitoring of burned areas and their analysis;
  • Integration of wildfire video monitoring with other advanced methods such as the Internet of Things (IoT) and/or Crowdsourcing;
  • Integration of wildfire monitoring and surveillance with advanced ICT systems, such as GIS technologies, meteorological data visualization, wildfire risk estimation, wildfire spread simulation, virtual reality (vR) and augmented reality (AR).

We look forward to receiving your contributions.

Prof. Dr. Darko Stipanicev
Dr. Giovanni Laneve
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Fire is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • wildfire detection
  • fire detection
  • smoke detection
  • flame detection
  • wildfire monitoring
  • wildfire surveillance
  • remote sensing
  • wildfire detection validation and testing
  • burned-area monitoring
  • burned-area analysis
  • wildfire risk estimation
  • wildfire spread simulation
  • virtual reality (VR)
  • augmented reality (AR)

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (3 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

15 pages, 7669 KiB  
Article
Advanced Multi-Label Fire Scene Image Classification via BiFormer, Domain-Adversarial Network and GCN
by Yu Bai, Dan Wang, Qingliang Li, Taihui Liu and Yuheng Ji
Fire 2024, 7(9), 322; https://doi.org/10.3390/fire7090322 (registering DOI) - 15 Sep 2024
Abstract
Detecting wildfires presents significant challenges due to the presence of various potential targets in fire imagery, such as smoke, vehicles, and people. To address these challenges, we propose a novel multi-label classification model based on BiFormer’s feature extraction method, which constructs sparse region-indexing [...] Read more.
Detecting wildfires presents significant challenges due to the presence of various potential targets in fire imagery, such as smoke, vehicles, and people. To address these challenges, we propose a novel multi-label classification model based on BiFormer’s feature extraction method, which constructs sparse region-indexing relations and performs feature extraction only in key regions, thereby facilitating more effective capture of flame characteristics. Additionally, we introduce a feature screening method based on a domain-adversarial neural network (DANN) to minimize misclassification by accurately determining feature domains. Furthermore, a feature discrimination method utilizing a Graph Convolutional Network (GCN) is proposed, enabling the model to capture label correlations more effectively and improve performance by constructing a label correlation matrix. This model enhances cross-domain generalization capability and improves recognition performance in fire scenarios. In the experimental phase, we developed a comprehensive dataset by integrating multiple fire-related public datasets, and conducted detailed comparison and ablation experiments. Results from the tenfold cross-validation demonstrate that the proposed model significantly improves recognition of multi-labeled images in fire scenarios. Compared with the baseline model, the mAP increased by 4.426%, CP by 4.14% and CF1 by 7.04%. Full article
Show Figures

Figure 1

Figure 1
<p>Rescaled samples of fire images from CFDB.</p>
Full article ">Figure 2
<p>Rescaled samples of fire images from KT.</p>
Full article ">Figure 3
<p>Rescaled samples of fire images from VOC2012.</p>
Full article ">Figure 4
<p>Model framework diagram.</p>
Full article ">Figure 5
<p>An example of the conditional probability relationship between two labels is provided. Typically, when the image contains “flame”, there is a high likelihood that “smoke” is also present. However, if “smoke” is observed, “flame” may not necessarily be present.</p>
Full article ">Figure 6
<p>BiFormer Block operational flow (<span class="html-fig-inline" id="fire-07-00322-i001"><img alt="Fire 07 00322 i001" src="/fire/fire-07-00322/article_deploy/html/images/fire-07-00322-i001.png"/></span> Represents a residual connection).</p>
Full article ">Figure 7
<p>Domain classification and label classification network architecture.</p>
Full article ">Figure 8
<p>Visualization of results (where red represents a higher level of concern).</p>
Full article ">Figure 9
<p>Visualization of multi-label classification results (where green means the prediction is correct and red means the prediction is incorrect).</p>
Full article ">Figure 10
<p>Accuracy comparisons with different values of τ.</p>
Full article ">Figure 11
<p>Example of predicting sun.</p>
Full article ">Figure 12
<p>Example of predicting clouds.</p>
Full article ">Figure 13
<p>Example of predicting fire and smog.</p>
Full article ">
19 pages, 5011 KiB  
Article
Comparative Analysis between Remote Sensing Burned Area Products in Brazil: A Case Study in an Environmentally Unstable Watershed
by Juarez Antonio da Silva Junior, Admilson da Penha Pacheco, Antonio Miguel Ruiz-Armenteros and Renato Filipe Faria Henriques
Fire 2024, 7(7), 238; https://doi.org/10.3390/fire7070238 - 9 Jul 2024
Viewed by 823
Abstract
Forest fires can profoundly impact the hydrological response of river basins, modifying vegetation characteristics and soil infiltration. This results in a significant increase in surface flow and channel runoff. In response to these effects, many researchers from different areas of earth sciences are [...] Read more.
Forest fires can profoundly impact the hydrological response of river basins, modifying vegetation characteristics and soil infiltration. This results in a significant increase in surface flow and channel runoff. In response to these effects, many researchers from different areas of earth sciences are committed to determining emergency measures to rehabilitate river basins, intending to restore their functions and minimize damage to soil resources. This study aims to analyze the mapping detection capacity of burned areas in a river basin in Brazil based on images acquired by AMAZÔNIA-1/WFI and the AQ1KM product. The effectiveness of the AMAZÔNIA-1 satellite in this regard is evaluated, given the importance of the subject and the relatively recent introduction of the satellite. The AQ1KM data were used to analyze statistical trends and spatial patterns in the area burned from 2003 to 2023. The U-Net architecture was used for training and classification of the burned area in AMAZÔNIA-1 images. An increasing trend in burned area was observed through the Mann–Kendall test map and Sen’s slope, with the months of the second semester showing a greater occurrence of burned areas. The NIR band was found to be the most sensitive spectral resource for detecting burned areas. The AMAZÔNIA-1 satellite demonstrated superior performance in estimating thematic accuracy, with a correlation of above 0.7 achieved in regression analyses using a 10 km grid cell resolution. The findings of this study have significant implications for the application of Brazilian remote sensing products in ecology, water resources, and river basin management and monitoring applications. Full article
Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Study area. Delimitation of the Ottocoded Level-3 river basin located between the Caatinga and Cerrado biomes.</p>
Full article ">Figure 2
<p>Area graph of the monthly burned area (km<sup>2</sup>) in the river basin from 2003 to 2023.</p>
Full article ">Figure 3
<p>Spatial distribution of annual frequency of burned area per month.</p>
Full article ">Figure 4
<p>Map of mean (<b>a</b>), standard deviation (<b>b</b>), maximum frequency (<b>c</b>), trend by Z value (<b>d</b>), and Sen’s slope (<b>e</b>) for the burned area grid based on the years 2003 to 2023.</p>
Full article ">Figure 5
<p>Boxplot of the sample pixel values of the burned and unburned areas of the AMAZÔNIA-1 blue (<b>a</b>), green (<b>b</b>), red (<b>c</b>), NIR (<b>d</b>), BAI (<b>e</b>), and NDVI (<b>f</b>) bands for the years 2021, 2022, and 2023.</p>
Full article ">Figure 6
<p>Training curves for U-Net classification.</p>
Full article ">Figure 7
<p>Spatial distribution of classification errors in the study area. (<b>a</b>–<b>c</b>) for AMAZÔNIA-1, and (<b>d</b>–<b>f</b>) for AQ1KM.</p>
Full article ">Figure 8
<p>Linear regression by 10 km × 10 km grid proportion between the AQ1KM product and AMAZÔNIA-1, labeled as burned by independent reference data for the years (<b>a</b>) 2021, (<b>b</b>) 2022, and (<b>c</b>) 2023.</p>
Full article ">
18 pages, 4000 KiB  
Article
Predicting Wildfire Ember Hot-Spots on Gable Roofs via Deep Learning
by Mohammad Khaled Al-Bashiti, Dac Nguyen, M. Z. Naser and Nigel B. Kaye
Fire 2024, 7(5), 153; https://doi.org/10.3390/fire7050153 - 25 Apr 2024
Viewed by 1033
Abstract
Ember accumulation on and around homes can lead to spot fires and home ignition. Post wildland fire assessments suggest that this mechanism is one of the leading causes of home destruction in wildland urban interface (WUI) fires. However, the process of ember deposition [...] Read more.
Ember accumulation on and around homes can lead to spot fires and home ignition. Post wildland fire assessments suggest that this mechanism is one of the leading causes of home destruction in wildland urban interface (WUI) fires. However, the process of ember deposition and accumulation on and around houses remains poorly understood. Herein, we develop a deep learning (DL) model to analyze data from a series of ember-related wind tunnel experiments for a range of wind conditions and roof slopes. The developed model is designed to identify building roof regions where embers will remain in contact with the rooftop. Our results show that the DL model is capable of accurately predicting the position and fraction of the roof on which embers remain in place as a function of the wind speed, wind direction, roof slope, and location on the windward and leeward faces of the rooftop. The DL model was augmented with explainable AI (XAI) measures to examine the extent of the influence of these parameters on the rooftop ember coverage and potential ignition. Full article
Show Figures

Figure 1

Figure 1
<p>Flow diagram of research project showing motivation, experimental work, and research questions.</p>
Full article ">Figure 2
<p>Diagram of the model building’s dimensions. The specific values for each dimension are given in <a href="#fire-07-00153-t001" class="html-table">Table 1</a>.</p>
Full article ">Figure 3
<p>Roof image of retained embers. (<b>a</b>) Original image; (<b>b</b>) after processing, where black pixels indicate the presence of embers retained on the rooftop. For this experiment, the wind was approaching at a 45° angle from the upper-left corner of the building.</p>
Full article ">Figure 4
<p>Commonly used activation functions.</p>
Full article ">Figure 5
<p>Illustration of a typical DL model, including one input layer, four hidden layers, and one output layer. Note: input features include wind speed, wind direction, roof slope, and location.</p>
Full article ">Figure 6
<p>Pearson correlation heatmap.</p>
Full article ">Figure 7
<p>Spearman correlation heatmap.</p>
Full article ">Figure 8
<p>Point-biserial correlation heatmap.</p>
Full article ">Figure 9
<p>An illustration of the accuracy of the model prediction. The red line represents exact agreement.</p>
Full article ">Figure 10
<p>Summary plot of Shapley values for the full database.</p>
Full article ">Figure 11
<p>Summary plot of Shapley values for the windward (<b>top</b>) and leeward (<b>bottom</b>) datasets.</p>
Full article ">Figure 11 Cont.
<p>Summary plot of Shapley values for the windward (<b>top</b>) and leeward (<b>bottom</b>) datasets.</p>
Full article ">
Back to TopTop