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
- Differentiating weeds from crops and other vegetation.
- Dense field environments made individual weed boundaries difficult to identify.
- Overlapping plants required careful polygon placement.
- Small and partially visible weeds were challenging to annotate accurately.
- Similar-looking plants sometimes made weed identification difficult.
- Soil, crop residue and natural field backgrounds added further complexity.
- Varying lighting conditions and shadows affected weed visibility.
- Maintaining precise segmentation boundaries across the dataset demanded close attention to detail.
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.