Dataclusterlabspvtltd/Fire-and-Smoke-Dataset

该数据集包含真实世界的火灾和烟雾场景图像，通过手机在不同条件下拍摄。适用于智能摄像头、火灾报警系统、监控和安全应用的早期火灾和烟雾检测模型训练。场景包括典型的家庭情况，如垃圾燃烧、纸张和塑料燃烧、田间作物燃烧和家庭烹饪，涵盖室内外环境以及不同的光照和天气条件。数据集包含fire和smoke两个类别。样本数据集包含约100-200张图像，注释格式为Pascal VOC (XML)。

This dataset contains real-world images of fire and smoke scenarios, captured on mobile phones under diverse conditions. It is well-suited for training early fire and smoke detection models used in smart cameras, fire alarm systems, surveillance, and safety applications. Scenes include typical domestic situations such as garbage burning, paper and plastic burning, field crop burning, and domestic cooking — covering both indoor and outdoor environments under varied lighting and weather. The dataset includes two classes: fire and smoke. The sample dataset contains approximately 100-200 images with annotations in Pascal VOC (XML) format.