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 Datura stramonium from similar-looking weeds and vegetation.
- Identifying plants in dense and complex field backgrounds.
- Handling partially visible or overlapping plants.
- Accurate bounding boxes had to be maintained for plants of different sizes and growth stages.
- Ensuring consistent annotations across varying lighting and image conditions.
Related work
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.
Almond Tree Detection
Bounding boxes around individual almond trees in top-down drone imagery, where shadows and neighbouring vegetation mimicked real trees.
Apple Detection
Bounding boxes around individual apples on tree branches, where dense leaves and occlusion hid much of each fruit.