FireRisk Dataset Overview

Basic Information

• Language: English
• Task Category: Image Classification
• Tags: Remote Sensing, Earth Observation, Geospatial, Aerial Imagery, Land‑Cover Classification
• Dataset Name: FireRisk

Dataset Details

• Features:

  • Images: Image type data
  • Labels: Categorical labels covering:
    • high
    • low
    • moderate
    • non‑burnable
    • very_high
    • very_low
    • water

• Data Split:

  • Training Set: 70,331 samples (size: 11,575,141,474.625 bytes)

• Size:

  • Download Size: 11,575,727,336 bytes
  • Dataset Size: 11,575,141,474.625 bytes

Configuration

• Default:
  • Data Files:
    • Training: data/train-*

Description

• Total Images: 91,872
• Bands: 3 (RGB)
• Image Size: 320 × 320
• Resolution: 1 m
• Land‑Cover Classes: 7
• Source: NAIP aerial imagery

Usage

from datasets import load_dataset
load_dataset("blanchon/FireRisk")

Citation

@article{shen2023firerisk,
  title   = {FireRisk: A Remote Sensing Dataset for Fire Risk Assessment with Benchmarks Using Supervised and Self‑supervised Learning},
  author  = {Shuchang Shen and Sachith Seneviratne and Xinye Wanyan and Michael Kirley},
  year    = {2023},
  journal = {arXiv preprint arXiv:2303.07035}
}