Forestry AI · Annotation and custom workflows Katam Technologies: Tree species detection from forestry imagery
10,000+forestry images labelled
99%+accuracy
“They’ve shown great adaptability whilst giving valuable input on areas of improvement.”Axel Persson, Katam Technologies
The challenge
Katam needed high-quality annotated data to train computer vision models for forestry optimisation, including detection of different tree species, delivered consistently at scale.
Our solution
- Annotated 10,000+ tree images with bounding boxes for detection model training.
- Assigned top annotators and ran tailored polygon-annotation training for precision.
- Held regular progress reviews and rigorous QA checks.
- Shared actionable insights for continuous model improvement.
Results
- 99%+ annotation accuracy
- Better model performance for forestry optimisation
- A scalable annotation process

More customer stories
Medical AI · SliceVault
Precise organ segmentation across CT, MRI and X-ray
Semantic segmentation of every organ on 10,000+ medical images, at 99%+ precision. Sports AI · SpiideoPlayer and object tracking for sports video
An embedded team annotated 30,000+ sports videos at 99%+ accuracy over a year-long collaboration.Start with a free pilot batch
We agree the batch size on a short call. If the quality isn't there, you've lost nothing.