Boreal Forest Fire: UAV-collected Wildfire Detection and Smoke Segmentation Dataset

This dataset consists of annotated images and videos of smoke resulting from prescribed burning events in Finnish boreal forests. The dataset was created to train and validate learning-based methods for wildfire detection and smoke segmentation and its effectiveness in doing so was shown in the linked studies.

The data was captured as 4K (3840×2160) videos at four events in Evo, Heinola, Karkkila, and Ruokolahti, using a DJI Phantom 4 drone. Individual frames of the videos were annotated manually with bounding boxes to enable the use of the data to train and test wildfire detection models. A portion of the bounding box annotated image data was resorted and annotated at the pixel level for image segmentation model training, validation, and testing. The training and validation data were annotated automatically using the Segment Anything Model and the manually annotated bounding boxes, while a small test set was annotated with manually drawn pixelwise masks.

The three parts of the dataset are stored in the three separate directories:

* Boreal-Forest-Fire-Subset-A: The bounding box annotated image data
* Boreal-Forest-Fire-Subset-B: 30-second 4K video clips with binary annotations for smoke
* Boreal-Forest-Fire-Subset-C: The segmentation mask annotated image data

In addition, a Jupyter notebook providing an example for visualising and converting the bounding box coordinates is provided in the Visualisation-And-Coordinate-Conversion-Notebook directory. Code for using the segmentation data is found in the linked Gitlab repository.

Some errors were found in the video labels after archiving the data. Refer to the video label directory (Boreal-Forest-Fire/Boreal-Forest-Fire-Subset-B/Video-Ground-Truth) metadata for accurate information.

Detailed data description is found in our Scientific Data article: https://doi.org/10.1038/s41597-025-05634-0

本数据集包含芬兰寒温带森林中计划烧除事件产生的烟雾的标注图像与视频数据。本数据集旨在训练并验证用于野火检测与烟雾分割的基于学习的方法，相关研究已证实了该数据集在该任务中的有效性。

数据采集于埃沃（Evo）、海诺拉（Heinola）、卡尔基拉（Karkkila）与鲁奥科拉赫蒂（Ruokolahti）的4处烧除现场，采用DJI Phantom 4无人机拍摄分辨率达4K（3840×2160）的视频。研究人员手动为视频的每一帧图像标注边界框，以支持野火检测模型的训练与测试。部分标注了边界框的图像数据经重新整理后，进一步完成了像素级标注，用于图像分割模型的训练、验证与测试。其中训练与验证集的像素标注通过分段任意模型（Segment Anything Model）结合手动标注的边界框自动生成，而小型测试集则采用手动绘制的像素级掩码完成标注。

本数据集分为三个部分，分别存储于三个独立目录中：
* 寒温带森林火灾子集A（Boreal-Forest-Fire-Subset-A）：标注有边界框的图像数据
* 寒温带森林火灾子集B（Boreal-Forest-Fire-Subset-B）：带有烟雾二元标注的30秒4K视频片段
* 寒温带森林火灾子集C（Boreal-Forest-Fire-Subset-C）：带有分割掩码标注的图像数据

此外，用于可视化与边界框坐标转换的示例Jupyter笔记本存储于Visualisation-And-Coordinate-Conversion-Notebook目录中。针对分割数据的使用代码可在关联的Gitlab代码仓库中获取。

数据归档后，研究人员发现部分视频标签存在错误。准确的标签信息请参考视频标签目录（Boreal-Forest-Fire/Boreal-Forest-Fire-Subset-B/Video-Ground-Truth）中的元数据。

数据集的详细描述可参阅我们发表于《Scientific Data》的论文：https://doi.org/10.1038/s41597-025-05634-0

搜集汇总

背景与挑战

背景概述

该数据集包含芬兰北方森林规定燃烧事件的烟雾图像和视频，用于训练和验证野火检测及烟雾分割模型。数据分为三个子集，包括边界框注释、视频剪辑和分割掩码注释，采集于四个不同地点，使用4K无人机视频。