9 datasets found

The FireNet dataset is a publicly available dataset for wildfire detection, containing 108 video clips and 160 images that are prone to error detection.

The FireFlame dataset is a publicly available dataset for wildfire detection, containing three categories: Fire (flame), smoke, and neutral, with 1000 images each.

The Forest-Fire Dataset is a publicly available dataset for wildfire detection, containing 1900 images in the training set and the test set, of which 950 are images with fire.

The Bowfire dataset is a publicly available dataset for wildfire detection, containing 226 images, of which 119 contain flames.

The Fire Detection Dataset is a publicly available dataset for wildfire detection, containing 149 video clips, of which 74 videos contain smoke.

The SKLFS-WildFire Test dataset is a large-scale test set for wildfire detection, containing 50,735 images from 3,309 video clips, of which 3,588 images are positive images...

SYN70K dataset for smoke segmentation, containing 70k synthetic images.

SMOKE5K dataset for smoke segmentation, containing 400 real images.

SmokeSeg dataset for early smoke segmentation, containing 6144 real images with pixel-wise annotations.