Tecnologías VM
Work
Graduation R&D · TEC · 2017–2019

LiDAR & Photogrammetry

Our 2017–2019 R&D program to map a tropical forest from the air — drones, LiDAR, photogrammetry and deep learning combined to count and classify trees and estimate the carbon the forest stores. Run with the Tecnológico de Costa Rica (TEC) as a staged series of sponsored Computer Engineering graduation projects.

A UAV sweeping LiDAR over the canopy — the point cloud accumulating, dot by dot.
Imagery

From the field

Captures from the 2017–2019 campaigns — flights, point clouds and fieldwork at the TEC sites. Real photos land here soon.

real imagery — coming
slot reserved — real photo (TODO)
real imagery — coming
slot reserved — real photo (TODO)
real imagery — coming
slot reserved — real photo (TODO)
01 / 03
0TEC graduation projects sponsored
0.0%Tree-species CNN accuracy
~0%Point-cloud tree classification
0.0MPoints in one LiDAR pass

Costa Rica had set a national carbon-neutrality goal, and certifying forest carbon meant inventorying biomass by hand — 12 to 24 months of plot-by-plot fieldwork, sometimes dangerous or simply inaccessible. We saw the opening: combine drones, LiDAR, photogrammetry and deep learning to estimate forest carbon faster and at lower cost, so Costa Rican landowners could participate in the carbon market.

The intended pipeline ran end to end: a drone overflies the forest carrying LiDAR and cameras under centimeter-level RTK positioning; photogrammetry and point-cloud processing build terrain and canopy-height models; deep learning detects individual trees, their species and dimensions; and an allometric model turns those variables into biomass and CO₂.

We executed it as sponsored Computer Engineering graduation projects at TEC, in partnership with TEC's UAS Photogrammetry Lab — six students across three cohorts, each owning one slice, with explicit knowledge transfer between them. The company financed and supervised the work and retained the IP.

It is our earliest sustained applied deep-learning and computer-vision program — proof that we were working with CNNs, remote sensing and big data years before “AI” became the label. It also set a pattern we still use: sponsor real R&D, validate honestly, carry the patterns forward.

The pipeline

From flight to carbon estimate.

  1. 01

    Fly

    A UAV overflies the forest carrying LiDAR and cameras, under centimeter-level RTK positioning.

  2. 02

    Capture

    Velodyne VLP-16 point clouds logged in the field — passes of up to 16.4M points.

  3. 03

    Model

    Photogrammetry (SfM) and point-cloud processing build terrain and surface models — subtracted into a canopy-height model.

  4. 04

    Detect

    Deep learning finds and classifies individual trees — species, height, crown — at up to ~95% accuracy.

  5. 05

    Estimate

    An allometric model turns per-tree variables into biomass, and biomass into CO₂.

What we built

Validated, component by component.

LiDAR capture & georeferencing

A Velodyne VLP-16 + RTK + Raspberry Pi capture chain that produced the first 3D point clouds from that sensor at the institution — passes of up to 16.4M points.

Photogrammetry → biomass

Drone photos to 3D canopy models (SfM), a canopy-height model, and >90% detection of separated crowns — feeding an allometric biomass estimate.

Tree-species classifier

A CNN (Inception V3, transfer learning) classifying tree species from photos at 94.4% accuracy — selected over VGG16 after a head-to-head.

Point-cloud tree classification

Algorithms that count and extract individual trees from georeferenced point clouds — about 95% average accuracy on the evaluated samples.

The honest outcome

The program validated its components but was never assembled into a single deployed end-to-end system, and no forest CO₂ figure was corroborated against ground truth. We present it as what it was: an early, ambitious R&D program that built real capability — and taught us how hard the last mile is.

Contact

Let’s build something that works.

Have a real problem where emerging technology might be part of the answer? We’d like to hear about it.