uav-image-segmentation

the main objective of this challenge is to segment images acquired by a small UAV (sUAV) in the area of Houston, Texas. These images were acquired to assess the damages on residential and public properties after Hurricane Harvey. In total, there are 25 categories of segments (e.g., roof, trees, pools etc.)

The following classes are present

0: Background
1: Property Roof
2: Secondary Structure
3: Swimming Pool
4: Vehicle
5: Grass
6: Trees / Shrubs
7: Solar Panels
8: Chimney
9: Street Light
10: Window
11: Satellite Antenna
12: Garbage Bins
13: Trampoline
14: Road/Highway
15: Under Construction / In Progress Status
16: Power Lines & Cables
17: Water Tank / Oil Tank
18: Parking Area - Commercial
19: Sports Complex / Arena
20: Industrial Site
21: Dense Vegetation / Forest
22: Water Body
23: Flooded
24: Boat

The dataset is hosted on Kaggle: Hurricane-Harvey-flood-damage-segmentation-dataset