Case Study

How can a city, region, or road administration obtain an up-to-date overview of the state of vegetation in record time? With roadside.vision, it is very simple. See for yourself and watch a short practical demonstration.

Long story short:

3 days. 200 km. And 10,000 surveyed trees.

While vegetation—especially trees—is typically surveyed and subsequently assessed by field workers, roadside.vision utilizes remote sensing methods (vehicle-mounted camera/LiDAR) or existing databases of panoramic road imagery, similar to Google Street View. The analysis is then handled quickly and reliably by AI thanks to machine learning.

With almost 100% certainty,
we locate and identify individual trees along roadsides,
achieving a 90% success rate in urban areas.

roadside.vision in numbers

Spoiler alert:
Transparent pricing based on road kilometers, not tree count.
A unique advantage for public tenders.

During 9 hours of driving with a roof-mounted camera,

225,000 photos were generated,

and in just 2 additional days,

a complete overview of nearly 10,000 trees was available,

along approximately 200 km of Chrudim roads.

What information does roadside.vision provide?

Detection of existence

Freestanding trees
- Built-up area (municipality) – 91% sensitivity, 16% error rate
- Outside built-up area (outside municipality) – 99% sensitivity, 1% error rate
Groups of trees and continuous stands