Dataset

This dataset was created by Shubham Karande

Released under Apache 2.0

Contents

D Fire

## Overview

D Fire is a dataset for object detection tasks - it contains Fire annotations for 20,325 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Dataset

This dataset was created by Vishwa Malani

Released under Apache 2.0

Contents

Dataset

This dataset was created by newbee1905

Released under CC0: Public Domain

Contents

D Fire No Night

## Overview

D Fire No Night is a dataset for object detection tasks - it contains Fire Smoke HJ37 annotations for 2,357 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

D-Fire Dataset for Smoke and Fire Detection

This dataset is an enhanced version of the original D-Fire dataset, designed to facilitate smoke and fire detection tasks. It has been restructured to include a validation split, making it more accessible and user-friendly.

Explore Flare Guard

Introducing Flare Guard — an advanced, open-source solution for real-time fire and smoke detection.

This system uses YOLOv11, an advanced object detection model, to monitor live video feeds and detect fire hazards in real-time. Detected threats trigger instant alerts via Telegram and WhatsApp for rapid response.

🔗 Quick Access Links

• 📂 GitHub Repository – Access the full source code and documentation.
• 🎥 LinkedIn Project Demo & Launch – Watch the live demonstration and learn more about the project.

Example of reached Results:

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12748471%2F632cfe5056cc683123c1873547d670ce%2Falert_20250210-034709-167281.jpg?generation=1742122420748481&alt=media" alt="CCVT_EXAMPLE">

Directory Structure

The dataset is organized as follows:

• train/
  • images/: Training images
  • labels/: Training labels in YOLO format
• val/
  • images/: Validation images
  • labels/: Validation labels in YOLO format
• test/
  • images/: Test images
  • labels/: Test labels in YOLO format

Classes

The dataset includes annotations for the following classes:

• 0: Smoke
• 1: Fire

Dataset Statistics

The dataset comprises over 21,000 images, categorized as follows:

Total bounding boxes:

• Fire: 14,692
• Smoke: 11,865

Data Splits

The dataset is divided into training, validation, and test sets to support model development and evaluation.

Citation

If you use this dataset in your research or projects, please cite the original paper:

Pedro Vinícius Almeida Borges de Venâncio, Adriano Chaves Lisboa, Adriano Vilela Barbosa. "An automatic fire detection system based on deep convolutional neural networks for low-power, resource-constrained devices." Neural Computing and Applications, vol. 34, no. 18, 2022, pp. 15349–15368. DOI: 10.1007/s00521-022-07467-z.

Acknowledgments

Credit for the original dataset goes to the researchers from Gaia, solutions on demand (GAIA). The original dataset and more information can be found in the D-Fire GitHub repository.

D Fire Opti 2

## Overview

D Fire Opti 2 is a dataset for object detection tasks - it contains Fs annotations for 11,687 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

9.5AI D Fire

## Overview

9.5AI D Fire is a dataset for object detection tasks - it contains Fire annotations for 11,689 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Dataset

This dataset was created by ryujima sengoku

Contents

Fire and smoke dataset for object detection.

Wildfire Detection With Bounding Boxes + D Fire Smoke Only

## Overview

Wildfire Detection With Bounding Boxes + D Fire Smoke Only is a dataset for object detection tasks - it contains Fire Smoke Olqe annotations for 6,629 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

D-Fire is an image dataset of fire and smoke occurrences designed for machine learning and object detection algorithms with more than 21,000 images. The dataset summary is detailed in the table below.

Number of images Number of bounding boxes | Category |Images | | Only fire | 1,164 | |Only smoke |5,867 | | Fire and smoke | 4,658 | | None |9,838|

| Class | # Bounding boxes | |Fire | 14,692 | | Smoke |11,865 | Annotations follow the YOLO format, with normalized coordinates between 0 and 1. To facilitate usage, we provide a utils.yolo2pixel function to convert these normalized coordinates into pixel coordinates.

FASDD is a largest and most generalized Flame And Smoke Detection Dataset for object detection tasks, characterized by the utmost complexity in fire scenes, the highest heterogeneity in feature distribution, and the most significant variations in image size and shape. FASDD serves as a benchmark for developing advanced fire detection models, which can be deployed on watchtowers, drones, or satellites in a space-air-ground integrated observation network for collaborative fire warning. This endeavor provides valuable insights for government decision-making and fire rescue operations. FASDD contains fire, smoke, and confusing non-fire/non-smoke images acquired at different distances (near and far), different scenes (indoor and outdoor), different light intensities (day and night), and from various visual sensors (surveillance cameras, UAVs, and satellites). FASDD consists of three sub-datasets, a Computer Vision (CV) dataset (i.e. FASDD_CV), a Unmanned Aerial Vehicle (UAV) dataset (i.e. FASDD_UAV), and an Remote Sensing (RS) dataset (i.e. FASDD_RS). FASDD comprises 122,634 samples, with 70,581 annotated as positive samples and 52,073 labeled as negative samples. There are 113,154 instances of flame objects and 73,072 instances of smoke objects in the entire dataset. FASDD_CV contains 95,314 samples for general computer vision, while FASDD_UAV consists of 25,097 samples captured by UAV, and FASDD_RS comprises 2,223 samples from satellite imagery. FASDD_CV contains 73,297 fire instances and 53,080 smoke instances. The CV dataset exhibits considerable variation in image size, ranging from 78 to 10,600 pixels in width and 68 to 8,858 pixels in height. The aspect ratios of the images also vary significantly, ranging from 1:6.6 to 1:0.18. FASDD_UAV contains 36,308 fire instances and 17,222 smoke instances, with image aspect ratios primarily distributed between 4:3 and 16:9. In FASDD_RS, there are 2,770 smoke instances and 3,549 flame instances. The sizes of remote sensing images are predominantly around 1,000×1,000 pixels.FASDD is provided in three compressed files: FASDD_CV.zip, FASDD_UAV.zip, and FASDD_RS.zip, which correspond to the CV dataset, the UAV dataset, and the RS dataset, respectively. Additionally, there is a FASDD_RS_SWIR. zip folder storing pseudo-color images for detecting flame objects in remote sensing imagery. Each zip file contains two folders: "images" for storing the source data and "annotations" for storing the labels. The "annotations" folder consists of label files in four formats: YOLO, VOC, COCO, and TDML. The dataset is divided randomly into training, validation, and test sets, with a ratio of 1/2, 1/3, and 1/6, respectively, within each label format. In FASDD_CV, FASDD_UAV, and FASDD_RS, images and their corresponding annotation files have been individually sorted starting from 0. The flame and smoke objects in FASDD are given the labels "fire" and "smoke" for the object detection task, respectively. The names of all images and annotation files are prefixed with "Fire", "Smoke", "FireAndSmoke", and "NeitherFireNorSmoke", representing different categories for scene classification tasks.When using this dataset, please cite the following paper. Thank you very much for your support and cooperation：################################################################################使用数据集请引用对应论文，非常感谢您的关注和支持：Wang, M., Yue, P., Jiang, L., Yu, D., Tuo, T., & Li, J. (2025). An open flame and smoke detection dataset for deep learning in remote sensing based fire detection. Geo-spatial Information Science, 28(2), 511-526.################################################################################

Combined Forest Fire Datasets

This dataset is a compilation of forest fire-related datasets sourced from various repositories and research platforms. It combines data from multiple sources to provide a comprehensive collection for analysis and research purposes. The datasets included in this compilation are:

1. Dataset 1 (gjmr63rz2r-1): Retrieved from Mendeley
2. Dataset 2 (archive): Retrieved from Kaggle
3. Dataset 3 (ForestFireDataset(Object Detection)): Retrieved from Mendeley
4. Dataset 4 (D-Fire): Retrieved from GitHub
5. Dataset 5 (ForestFireDataset(Classifications)): Retrieved from Mendeley
6. Dataset 6 (mivia-fire): Retrieved from Mivia
7. Dataset 7 (mivia-smoke): Retrieved from Mivia

The datasets cover various aspects of forest fires, including object detection, classifications, and related imagery. By combining these datasets, researchers and analysts can access a diverse range of data for studying forest fires, developing predictive models, and exploring mitigation strategies.

Copyright Notice:

The individual datasets included in this compilation retain their respective copyrights. Users are encouraged to refer to the original sources for specific usage terms and citation requirements. This combined dataset is provided for research and educational purposes only.

Fire And Somke Dfire

## Overview

Fire And Somke Dfire is a dataset for object detection tasks - it contains Fire Smoke 7Hg4 annotations for 3,123 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Earth science remote sensing imagery is rich in structural and spectral information, making such data an ideal platform for benchmarking for a broad range of machine learning (ML) tasks, from pattern retrieval to physics-informed classification to anomaly detection to transfer learning. Nevertheless, the utility of Earth science remote sensing data remains largely unexplored by the broader ML community. Our goal is to bridge this gap and bring a rich variety of multisource multi-resolution Earth image data to a wider range of ML researchers who are non-experts in remote sensing, thereby increasing the utility and societal impact of such data products. In particular, motivated by the emerging wildfire crisis, we present radiometrically and geometrically calibrated radiance data from airborne and orbital instruments from the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), and the Korean Meteorological Administration (KMA).

Given the scarce occurrence of wildfires and complex spatio-temporal dependencies in radiance data, these datasets are especially well suited for benchmarking unsupervised and self-supervised learning tasks both on images and non-Euclidean objects. Our experiments on these datasets indicate that contrastive learning and transfer learning algorithms can capture the structures of views and scenes, map pixel space of multi-sensor imagery to a high-level embedding space for further downstream tasks, and facilitate more cohesive integration of the state-of-the-art ML approaches into wildfire risk analytics.

All NASA-based observations are freely usable under the Creative Commons Zero License.There are also no restrictions on the use of GOES Data. GK2A data are also open data without any restrictions on its use.

For the Planet data, we cannot not share the Radiances, but all masks within this dataset are freely usable with no restrictions.

Use:

An example of programmatic data access and usage can be found in the dataset's associated GitHub repository.

On the data input, input geometrically and radiometrically calibrated radiance data has been pulled from various NASA, NOAA, Planet, and KMA archives. For instruments that have multiple different spatial resolutions within their spectral bands (GOES and GK2A), all bands have been resampled to the lowest collective spatial resolution.

Geometric and radiometric calibration has been done by the science data processing pipelines of the various missions, and would not need to be done by anyone else looking to curate the same data. Further information for each instrument can be found in each of the publicly available Level-1 algorithm theoretical basis documents (ATBDs)

All input and label data have been put in GeoTiff format. Each band is in a separate raster band and each scene is in a separate GeoTiff file. Label files and input files are in separate tar files, labeled respectively, and the file names match for input and labels, with the exception of an additional .fire and .smoke in the respective label filenames and subfolders.

The GeoTiff data format natively contains geolocation metadata internally, and can be interfaced with via C/C++/Python GDAL packages, or other python packages that wrap GDAL, like rasterio and rioxarray . The documentation for SIT-FUSE , the package with which the labels were generated, also has examples on how to read and interface with various data formats, including GeoTiffs. Lastly, this data can be interfaced with using Geographic Information Systems (GIS), like the free and open-source QGIS.

Timing information can be found in the file names, which all use the standard formats from the various instruments' L1B datasets.

V2 includes additional GOES-18 radiance data and associated smoke and fire labels for the recent LA fires (Palisades and Eaton fires in January of 2025).

V3 provides a reorganization of all data, and an inclusion of improved and additional data from airborne and satellite platforms in 2019, associated with this study: https://arxiv.org/pdf/2501.15343 .

Current fire coverage includes:

• 2019: Williams Flats, Sheridan, Horsefly, and Mosquito (US)
• 2022: Uljin Forest Fire (S. Korea; largest fire on record in S. Korea)
• 2025: Palisades and Eaton Fires (US)

Additional data for the 2025 Palisades and Eaton fires from the TEMPO instrument is currently being validated and will be released in a version shortly.

Validation:

These labels have been extensively validated and further information can be referenced in associated publications:
https://doi.org/10.3390/rs13122364
https://doi.org/10.3390/rs17071267

Dfire Small No Night

## Overview

Dfire Small No Night is a dataset for object detection tasks - it contains Fire Smoke 3mjx annotations for 614 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Earth science remote sensing imagery is rich in structural and spectral information, making such data an ideal platform for benchmarking for a broad range of machine learning (ML) tasks, from pattern retrieval to physics-informed classification to anomaly detection to transfer learning. Nevertheless, the utility of Earth science remote sensing data remains largely unexplored by the broader ML community. Our goal is to bridge this gap and bring a rich variety of multisource multi-resolution Earth image data to a wider range of ML researchers who are non-experts in remote sensing, thereby increasing the utility and societal impact of such data products. In particular, motivated by the emerging wildfire crisis, we present radiometrically and geometrically calibrated radiance data from airborne and orbital instruments from the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), and the Korean Meteorological Administration (KMA). Given the scarce occurrence of wildfires and complex spatio-temporal dependencies in radiance data, these datasets are especially well suited for benchmarking unsupervised and self-supervised learning tasks both on images and non-Euclidean objects. Our experiments on these datasets indicate that contrastive learning and transfer learning algorithms can capture the structures of views and scenes, map pixel space of multi-sensor imagery to a high-level embedding space for further downstream tasks, and facilitate more cohesive integration of the state-of-the-art ML approaches into wildfire risk analytics. All NASA-based observations are freely usable under the Creative Commons Zero License.There are also no restrictions on the use of GOES Data. GK2A data are also open data without any restrictions on its use. Use: On the data input, input geometrically and radiometrically calibrated radiance data has been pulled from various NASA, NOAA, and KMA archives. For instruments that have multiple different spatial resolutions within their spectral bands (GOES and GK2A), all bands have been resampled to the lowest collective spatial resolution. Geometric and radiometric calibration has been done by the science data processing pipelines of the various missions, and would not need to be done by anyone else looking to curate the same data. Further information for each instrument can be found in each of the publicly available Level-1 algorithm theoretical basis documents (ATBDs) All input and label data has been put in GeoTiff format. Each band is in a separate raster band and each scene is in a separate GeoTiff file. Label files and input files are in subdirectories labeled respectively, and the file names match for input and labels, with the exception of an additional .fire and .smoke in the respective label filenames and subfolders.

Dfire 2 Full

## Overview

Dfire 2 Full is a dataset for object detection tasks - it contains Humo YABP annotations for 7,066 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset series refers to the information on active fire detection provided by the European Forest Fire Information System (EFFIS). ▷_How to cite: see below_◁

Active fires are located on the basis of the so-called thermal anomalies produced by them. The algorithms compare the temperature of a potential fire with the temperature of the land cover around it; if the difference in temperature is above a given threshold, the potential fire is confirmed as an active fire or "hot spot". The mapping of active fires is performed to provide a synoptic view of current fires in Europe and as a means to help the subsequent mapping of burnt fire perimeters. Information on active fires is normally updated 6 times daily and made available in EFFIS within 2-3 hours of the acquisition of the MODIS/VIIRS images. When interpreting the hotspots displayed in the map, the following must be considered. 1. Hotspot location on the map is only accurate within the spatial accuracy of the sensor. 2. Some fires may be small or obscured by smoke or cloud and remain undetected. 3. The satellites also detect other heat sources (not all hotspots are fires). 4. To minimize false alarms and filter out active fires not qualified as wildfires (e.g. agricultural burnings), the system only displays a filtered subset of the hotspots detected by FIRMS. To this end a knowledge based algorithm is applied that takes into account the extent of surrounding land cover categories, the distance to urban areas and artificial surfaces, the confidence level of the hotspot.

How to cite - When using these data, please cite the relevant data sources. A suggested citation is included in the following:

• San-Miguel-Ayanz, J., Houston Durrant, T., Boca, R., Libertà, G., Branco, A., de Rigo, D., Ferrari, D., Maianti, P., Artés Vivancos, T., Schulte, E., Loffler, P., Benchikha, A., Abbas, M., Humer, F., Konstantinov, V., Pešut, I., Petkoviček, S., Papageorgiou, K., Toumasis, I., Kütt, V., Kõiv, K., Ruuska, R., Anastasov, T., Timovska, M., Michaut, P., Joannelle, P., Lachmann, M., Pavlidou, K., Debreceni, P., Nagy, D., Nugent, C., Di Fonzo, M., Leisavnieks, E., Jaunķiķis, Z., Mitri, G., Repšienė, S., Assali, F., Mharzi Alaoui, H., Botnen, D., Piwnicki, J., Szczygieł, R., Janeira, M., Borges, A., Sbirnea, R., Mara, S., Eritsov, A., Longauerová, V., Jakša, J., Enriquez, E., Lopez, A., Sandahl, L., Reinhard, M., Conedera, M., Pezzatti, B., Dursun, K. T., Baltaci, U., Moffat, A., 2017. Forest fires in Europe, Middle East and North Africa 2016. Publications Office of the European Union, Luxembourg. ISBN:978-92-79-71292-0, https://doi.org/10.2760/17690

• San-Miguel-Ayanz, J., Schulte, E., Schmuck, G., Camia, A., 2013. The European Forest Fire Information System in the context of environmental policies of the European Union. Forest Policy and Economics 29, 19-25. https://doi.org/10.1016/j.forpol.2011.08.012

• San-Miguel-Ayanz, J., Schulte, E., Schmuck, G., Camia, A., Strobl, P., Libertà, G., Giovando, C., Boca, R., Sedano, F., Kempeneers, P., McInerney, D., Withmore, C., de Oliveira, S. S., Rodrigues, M., Houston Durrant, T., Corti, P., Oehler, F., Vilar, L., Amatulli, G., 2012. Comprehensive monitoring of wildfires in Europe: the European Forest Fire Information System (EFFIS). In: Tiefenbacher, J. (Ed.), Approaches to Managing Disaster - Assessing Hazards, Emergencies and Disaster Impacts. InTech, Ch. 5. http://doi.org/10.5772/28441

Fire reconstructions for the Williams Creek and Squaretop Mountain study areas.