forest fire classifier dataset

The Fire and Smoke Detection Dataset is a comprehensive collection of aerial and terrestrial images designed to support AI-powered fire, smoke, and no-fire classification. The dataset includes images captured under varying conditions, such as day and night, and is carefully curated from Roboflow, Kaggle projects, and Google Images for reliable performance in real-world applications.

Key Features of the Dataset

• Diverse Image Scenarios: Includes fire, smoke, and no-fire scenarios captured through both aerial and ground-based perspectives.
• Day and Night Coverage: The dataset features images taken in both daylight and nighttime conditions for robust detection across various lighting conditions.
• Total of 4506 Images: This dataset includes 4506 total images, ensuring comprehensive coverage of fire and smoke scenarios.
• Balanced Classes:
  • Fire: 1383 images
  • No Fire: 1390 images
  • Smoke: 1433 images
• Well-Organized Dataset:
  • Training: 3705 images
  • Validation: 396 images
  • Test: 405 images

Applications

• Fire Monitoring: Enhance fire detection systems for wildfire monitoring, disaster management, and emergency response.
• AI and Computer Vision: Useful for image classification, object detection, and surveillance in areas like public safety, agriculture, and forestry.

This dataset is sourced from Kaggle.