Individual Tree Detection in Airborne LiDAR (ALS)

Many digital forest inventories run into the same problem: you want to be able to look at things at the level of individual trees, but your data is just a bit too coarse for that. Drone and terrestrial LiDAR deliver beautifully detailed trees, but are too expensive or too slow for an entire municipality or province. Airborne LiDAR does scale well, but point density is often low and crowns tend to merge into one another.

That’s exactly where our challenge lies: how do you still extract individual trees from standard airborne datasets that weren’t specifically collected for research?

With our latest AI models, we’re able to segment individual trees in low-density point clouds, even in areas where tree crowns strongly overlap. We don’t capture the forest down to the millimetre, but we do achieve a reliable tree-level inventory.

What this delivers:

✔️ Scalable maps for entire regions

✔️ Use of existing or affordable flight data

✔️ Height, volume and crown attributes per tree, without manual digitizing

If you’re curious how this works or how it performs in your forest or area type, we’d be happy to tell you more.