Sample frames for this project are available under NDA. Write to bilal@wortel.ai and we will share the relevant before and after pairs.
Challenges
- Distinguishing black grass from visually similar grasses and vegetation.
- Creating precise boundaries within dense agricultural fields.
- Overlapping leaves and plants complicated individual masks.
- Small and partially visible plants still had to be segmented accurately.
- Complex backgrounds and varying lighting conditions added difficulty.
- Maintaining consistent polygon boundaries throughout the dataset.
Related work
Aerial Orthomosaic Segmentation
Whole-site orthomosaics segmented into vegetation, water and bare ground across tiles stitched from multiple flights.
Ground Cover Quadrat Segmentation
2,275 cover-crop plot images annotated in Labelbox: plot corners boxed, then every pixel classed as groundcover, weed, residue, crop or soil.
Black Grass Segmentation
Polygon masks isolating black grass from crops and similar-looking grasses across dense agricultural field imagery.