Process

Raw field data become research-ready products through reproducible, automated processing pipelines.

Data Processing

Data we collect, and data brought to us, turned into orthomosaics, 3D models, and classified point clouds — the products research actually starts from.

Photogrammetric reconstruction (SfM/MVS)

Pix4D and OpenDroneMap for sparse and dense 3D reconstruction from drone or ground imagery.

Orthomosaic generation

Corrected, georeferenced top-down mosaics from drone imagery (RGB, multispectral, thermal). GeoTIFF output in any CRS, with optional COG packaging for web tiling.

Multispectral and thermal indices

NDVI, NDRE, GNDVI, thermal anomaly maps, and custom band math from MicaSense Altum or thermal payloads.

Point cloud processing

Classification, ground filtering, canopy height models, registration, and inter-epoch change detection (CloudCompare, lidR, PDAL, Trimble Business Center).

Automated pipelines and cloud archiving

Field data are collected once, then move through shared infrastructure rather than individual workstations: campaigns land in cloud storage, processing runs through automated, version-controlled workflows, and outputs are archived using open standards. The same inputs produce the same products, whether a dataset is processed this year or next.

Deliverables are cloud-native by default — Cloud-Optimized GeoTIFF (COG), Cloud-Optimized Point Cloud (COPC), GeoPackage — with standard formats available on request. Published datasets surface automatically in the lab's open data catalog.

Processing your own imagery

Collaborators can process drone imagery through the lab's hosted WebODM instance; workflows and account setup are covered in the lab documentation.

Have data that needs processing?

Send us a dataset — yours or ours — and we'll scope a pipeline that turns it into research-ready products. Initial scoping conversations are free — see Access & Rates for how projects are priced.

Contact the Lab