Attention to Fires: Multi-Channel Deep Learning Models for Wildfire Severity Prediction

Monaco, Simone; Greco, Salvatore; Farasin, Alessandro; Colomba, Luca; Apiletti, Daniele; Garza, Paolo; Cerquitelli, Tania; Baralis, Elena

doi:10.3390/app112211060

Cited by 18 publications

(8 citation statements)

References 67 publications

(68 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These findings agree with previous studies using satellite imagery (Landsat and Sentinel 2), showing that machine learning classifiers are well suited to the broad-scale mapping of fire severity with remotely sensed imagery [48,72,73]. However, even the most recent studies are not sensitive to local scale and spatial patchiness [74,75].…”

Section: Discussionsupporting

confidence: 91%

Total Carbon Content Assessed by UAS Near-Infrared Imagery as a New Fire Severity Metric

et al. 2022

View full text Add to dashboard Cite

The ash produced by forest fires is a complex mixture of organic and inorganic particles with many properties. Amounts of ash and char are used to roughly evaluate the impacts of a fire on nutrient cycling and ecosystem recovery. Numerous studies have suggested that fire severity can be assessed by measuring changes in ash characteristics. Traditional methods to determine fire severity are based on in situ observations, and visual approximation of changes in the forest floor and soil which are both laborious and subjective. These measures primarily reflect the level of consumption of organic layers, the deposition of ash, particularly its depth and color, and fire-induced changes in the soil. Recent studies suggested adding remote sensing techniques to the field observations and using machine learning and spectral indices to assess the effects of fires on ecosystems. While index thresholding can be easily implemented, its effectiveness over large areas is limited to pattern coverage of forest type and fire regimes. Machine learning algorithms, on the other hand, allow multivariate classifications, but learning is complex and time-consuming when analyzing space-time series. Therefore, there is currently no consensus regarding a quantitative index of fire severity. Considering that wildfires play a major role in controlling forest carbon storage and cycling in fire-suppressed forests, this study examines the use of low-cost multispectral imagery across visible and near-infrared regions collected by unmanned aerial systems to determine fire severity according to the color and chemical properties of vegetation ash. The use of multispectral imagery data might reduce the lack of precision that is part of manual color matching and produce a vast and accurate spatio-temporal severity map. The suggested severity map is based on spectral information used to evaluate chemical/mineralogical changes by deep learning algorithms. These methods quantify total carbon content and assess the corresponding fire intensity that is required to form a particular residue. By designing three learning algorithms (PLS-DA, ANN, and 1-D CNN) for two datasets (RGB images and Munsell color versus Unmanned Aerial System (UAS)-based multispectral imagery) the multispectral prediction results were excellent. Therefore, deep network-based near-infrared remote sensing technology has the potential to become an alternative reliable method to assess fire severity.

show abstract

Section: Discussionsupporting

confidence: 91%

Total Carbon Content Assessed by UAS Near-Infrared Imagery as a New Fire Severity Metric

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The experiment results showed that the Double-Step U-Net with BCE loss achieved the best MSE of 0.54. Monaco et al [147] also developed a two-step CNN solution to detect burned areas and predict their damage on satellite data. First, a binary semantic segmentation method-based CNN was used to detect burned areas, and then a regression method-based CNN was applied to predict their damage severity between 0 (no damage) and 4 (completely destroyed).…”

Section: Deep Learning-based Approaches For Fire Susceptibility Using...mentioning

confidence: 99%

“…Using a satellite image collected from Copernicus EMS, DS -UNet, and DS-UNet++ models with BCE loss showed a higher IoU of 75% and 74%, respectively, in delineating the burnt areas compared to DS-AttU and DS-SegU; DS-AttU, DS-UNet, and DS-UNet++ performed better in predicting the damage severity levels of burned areas, obtaining an RMSE of 2.429, 1.857, and 1.857, respectively. Monaco et al [149] also used DS-UNet to detect wildfire and to predict the damage severity level, from 0 (no damage) to 5 (completely destroyed) on Sentinel-2 images. DS-UNet achieved an average RMSE of 1.08, overcoming baseline methods such as DS-UNet++, DS-SegU, UNet++, PSPNet, and SegU-Net.…”

Section: Deep Learning-based Approaches For Fire Susceptibility Using...mentioning

confidence: 99%

Deep Learning Approaches for Wildland Fires Using Satellite Remote Sensing Data: Detection, Mapping, and Prediction

Ghali

Akhloufi

2023

Fire

View full text Add to dashboard Cite

Wildland fires are one of the most dangerous natural risks, causing significant economic damage and loss of lives worldwide. Every year, millions of hectares are lost, and experts warn that the frequency and severity of wildfires will increase in the coming years due to climate change. To mitigate these hazards, numerous deep learning models were developed to detect and map wildland fires, estimate their severity, and predict their spread. In this paper, we provide a comprehensive review of recent deep learning techniques for detecting, mapping, and predicting wildland fires using satellite remote sensing data. We begin by introducing remote sensing satellite systems and their use in wildfire monitoring. Next, we review the deep learning methods employed for these tasks, including fire detection and mapping, severity estimation, and spread prediction. We further present the popular datasets used in these studies. Finally, we address the challenges faced by these models to accurately predict wildfire behaviors, and suggest future directions for developing reliable and robust wildland fire models.

show abstract

“…Environmental science examples: Attention-based methods have already found application for precipitation mapping (Sønderby et al, 2020;Espeholt et al, 2022), estimating visibility due to coastal fog (Kamangir et al, 2021), generating super-resolution imagery (Liu et al, 2018), wildfire estimation (Monaco et al, 2021), population density estimation (Savner and Kanhangad, 2023), damage assessment (Hao et al, 2021), and land cover estimation (Ghosh et al, 2021;Wang and Sertel, 2021). Many additional examples are provided in the following section.…”

Section: A New Generation Of Neural Networkmentioning

confidence: 99%

Leveraging spatiotemporal information in meteorological image sequences: From feature engineering to neural networks

Bansal

Lee

Hilburn

et al. 2023

Environ. Data Science

View full text Add to dashboard Cite

Atmospheric processes involve both space and time. Thus, humans looking at atmospheric imagery can often spot important signals in an animated loop of an image sequence not apparent in an individual (static) image. Utilizing such signals with automated algorithms requires the ability to identify complex spatiotemporal patterns in image sequences. That is a very challenging task due to the endless possibilities of patterns in both space and time. Here, we review different concepts and techniques that are useful to extract spatiotemporal signals from meteorological image sequences to expand the effectiveness of AI algorithms for classification and prediction tasks. We first present two applications that motivate the need for these approaches in meteorology, namely the detection of convection from satellite imagery and solar forecasting. Then we provide an overview of concepts and techniques that are helpful for the interpretation of meteorological image sequences, such as (a) feature engineering methods using (i) meteorological knowledge, (ii) classic image processing, (iii) harmonic analysis, and (iv) topological data analysis; (b) ways to use convolutional neural networks for this purpose with emphasis on discussing different convolution filters (2D/3D/LSTM-convolution); and (c) a brief survey of several other concepts, including the concept of “attention” in neural networks and its utility for the interpretation of image sequences and strategies from self-supervised and transfer learning to reduce the need for large labeled datasets. We hope that presenting an overview of these tools—many of which are not new but underutilized in this context—will accelerate progress in this area.

show abstract

Attention to Fires: Multi-Channel Deep Learning Models for Wildfire Severity Prediction

Cited by 18 publications

References 67 publications

Total Carbon Content Assessed by UAS Near-Infrared Imagery as a New Fire Severity Metric

Total Carbon Content Assessed by UAS Near-Infrared Imagery as a New Fire Severity Metric

Deep Learning Approaches for Wildland Fires Using Satellite Remote Sensing Data: Detection, Mapping, and Prediction

Leveraging spatiotemporal information in meteorological image sequences: From feature engineering to neural networks

Contact Info

Product

Resources

About