--- license: mit language: - en --- Dataset for RAPID WILDFIRE HOTSPOT DETECTION USING SELF-SUPERVISED LEARNING ON TEMPORAL REMOTE SENSING DATA --------- - **Repository:** https://github.com/links-ads/igarss-multi-temporal-hotspot-detection - **Paper:** https://arxiv.org/abs/2405.20093 --------- ## Dataset Structure The main dataset used in the paper comprises the following inputs: | Suffix | Description | Format | |---------|-------------------------------------------------------------------------------------------|--------------------------| | x | Longitude (EPSG:4326) | float | | y | Latitude (EPSG:4326) | float | | point_id | Pixel id of the AOI | int | | class | Hotspot (1), No hotspot (0) | int | | event_id | Effis event id | str | | time | MSG15 activation timestamp | str | | lc_2018 | ESRI LULC class | int | | IR_016 | Infrared MSG15 channel | float | | IR_039 | Infrared MSG15 channel | float | | IR_087 | Infrared MSG15 channel | float | | IR_097 | Infrared MSG15 channel | float | | IR_108 | Infrared MSG15 channel | float | | IR_120 | Infrared MSG15 channel | float | | IR_134 | Infrared MSG15 channel | float | | VIS_006 | Visible MSG15 channel | float | | VIS_008 | Visible MSG15 channel | float | | WV_062 | Water Vapour MSG15 channel | float | | WV_073 | Water Vapour MSG15 channel | float |