datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
firesafety-wildfire
Dataset Summary
This dataset contains images of smoke plumes and fire. The images were taken remotely. The dataset is intended to train computer vision systems to detect forest fires.
balanced_wildfire_datasetFireDetectionDataset-flame-forest-flameye-wildfire
FlamEye — Wildfire Detection Dataset
A merged, deduplicated, and augmented dataset for real-time wildfire detection (fire and smoke) from CCTV/surveillance cameras. Built to train YOLOv8m for early-stage fire detection.
Classes
ID
Name
0
fire
1
smoke
Dataset Statistics
Split
Images
Train
~10,929
Validation
~3,000
Test
~1,500
Sources
Dataset
Source
Notes
D-Fire
Kaggle
Class IDs remapped:… See the full description on the dataset page: https://huggingface.co/datasets/baizhanquan/FireDetectionDataset-flame-forest-flameye-wildfire.Long_Distance_Wildfire_Smoke_Detection_Dataset
Long-Range Wildfire & Smoke Detection Dataset
Long-distance forest monitoring dataset featuring early smoke plumes and wildfire ignitions across global biomes.
🧐 Overview
Long-Range Wildfire & Smoke is a specialized open-source synthetic dataset for Computer Vision (CV) tasks focused on Environmental Monitoring and Early Warning Systems.
Detecting a wildfire before it crowns is the most effective way to prevent ecological disaster. This dataset focuses on the… See the full description on the dataset page: https://huggingface.co/datasets/Simuletic/Long_Distance_Wildfire_Smoke_Detection_Dataset.flameye-wildfire-detection
FlamEye — Wildfire Detection Dataset
A merged, deduplicated, and augmented dataset for real-time wildfire detection (fire and smoke) from CCTV/surveillance cameras. Built to train YOLOv8m for early-stage fire detection.
Classes
ID
Name
0
fire
1
smoke
Dataset Statistics
Split
Images
Train
~10,929
Validation
~3,000
Test
~1,500
Sources
Dataset
Source
Notes
D-Fire
Kaggle
Class IDs remapped: original… See the full description on the dataset page: https://huggingface.co/datasets/Hajorda/flameye-wildfire-detection.wildfire
Dataset Summary
This dataset contains images of smoke plumes and fire. The images were taken remotely. The dataset is intended to train computer vision systems to detect forest fires.
WildfireVLM
WildfireVLM
YOLO-format imagery and bounding-box annotations for wildfire/smoke detection, prepared as the dataset companion to WildfireVLM: AI-powered Analysis for Early Wildfire Detection and Risk Assessment Using Satellite Imagery.
Dataset description
The release contains 3,776 JPEG images and matching YOLO text annotations. It preserves the supplied train, valid, and test directory structure:
Split
Images
train
3,118
valid
57
test
601
Each… See the full description on the dataset page: https://huggingface.co/datasets/Ayanzadeh93/WildfireVLM.wildfire_mmstructure_wildfire_damage_classification
Dataset Card for Structures Damaged by Wildfire
Homepage: Image Dataset of Structures Damaged by Wildfire in California 2020-2022
Dataset Summary
The dataset contains over 18,000 images of homes damaged by wildfire between 2020 and 2022 in California, USA, captured by the California Department of Forestry and Fire Protection (Cal Fire) during the damage assessment process. The dataset spans across more than 18 wildfire events, including the 2020 August Complex Fire, the… See the full description on the dataset page: https://huggingface.co/datasets/kevincluo/structure_wildfire_damage_classification.wildfire-korea-embedded-300m
Wildfire Korea Embedded 300m (16-Channel Environmental Tiles)
A geospatial dataset containing 300 m embedded environmental feature tiles for the Korean Peninsula, designed for wildfire spread prediction and reinforcement-learning research.
This dataset provides model-ready 16-channel tensors representing static and quasi-static environmental conditions needed for wildfire simulation and RL inference.
It is paired with the wildfire episode dataset (wildfire-korea-episodes-300m) and… See the full description on the dataset page: https://huggingface.co/datasets/chaseungjoon/wildfire-korea-embedded-300m.wildfire-predictionafrica-synth-climate-wildfire-smoke-respiratory-all
Wildfire Smoke & Respiratory Outcomes (SSA) | Africa (Electric Sheep Africa metadata inventory)
Size category: 10K<n<100K - Formats: csv - Sector: health - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.
What This Dataset Covers
Health datasets… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-synth-climate-wildfire-smoke-respiratory-all.Wildfire_Images_From_Satilite_For_Azerbaijanbrasa-wildfire-lwir
Brasa — synthetic wildfire LWIR thermal imagery (v1.0)
Labeled synthetic thermal imagery for training and testing wildfire-detection
models — shipped with the evidence that its physics is real, including the
measurement we are worst at.
Real labeled wildfire thermal data is scarce: fires are dangerous to instrument,
aerial campaigns are expensive, and "ground truth" is usually a threshold drawn on
the very pixels a model trains on. Brasa is the synthetic alternative.
Every frame… See the full description on the dataset page: https://huggingface.co/datasets/AY4LA/brasa-wildfire-lwir.wildfire-preventionCSF-Wildfire-Frontline-DatasetWildfire-dataset
MSc Thesis on Wildfire risk assessment using remote sensing data
Introduction
Hi to whoever ends up here on this webpage. Here there is the material I used for my master thesis. You're free to use the code and check what I have done.
What have I done in a few words? I have built a machine learning model (CNN) capable of assessing the risk of a wildfire over the entire globe using remote sensing data.
Abstract
Assessing the risk of wildfires over the entire… See the full description on the dataset page: https://huggingface.co/datasets/beppemar/Wildfire-dataset.fireguard-sentinel2-wildfire-precursorwildfire-datasetwildfire-drone-routing-data
Wildfire Drone Routing — supplementary data
Large supplementary inputs for wildfire_drone_routing (release v1.0.2 of the public-release branch).
This dataset is also archived on Zenodo with a citable DOI: 10.5281/zenodo.21359920 (concept DOI, always resolves to the latest version).
The benchmark-critical ("Tier-1") data — per-year config, mask, Pyrologix static
risk map, and fire scenarios for 2021-2024 — is committed directly in the GitHub
repository and is not duplicated… See the full description on the dataset page: https://huggingface.co/datasets/romainpuech/wildfire-drone-routing-data.wildfire-watchwildfire-preventionLA-Wildfirefireedge-sentinel2-wildfirealaska-wildfire-occurrencewildfire-CA-2023Wildfireseu-wildfires
Dataset Card for EU-Wildfires
This dataset is made up of Sentinel-2 L1C and L2A scenes across Europe. They are meant to be used for testing burned area delineation using AI models.
Dataset Details
Dataset Description
Curated by: [More Information Needed]
Funded by [optional]: [More Information Needed]
Shared by [optional]: [More Information Needed]
Language(s) (NLP): [More Information Needed]
License: [More Information Needed]… See the full description on the dataset page: https://huggingface.co/datasets/risc-conrad/eu-wildfires.