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
- Detecting small trash objects within agricultural fields.
- Distinguishing trash from soil, crops, weeds and natural vegetation.
- Partially hidden objects among plants were easy to overlook.
- Trash varied widely in shape, size and appearance.
- Complex field backgrounds and changing image conditions added difficulty.
- Keeping bounding boxes tight and accurate around very small objects.
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.