How to Turn Drone Photos into a Farm Map — Free (Ortho, NDVI & Elevation)

Field Forge 3D · Field guide

A drone comes home with a few hundred photos. None of them, on its own, is a map — they're overlapping snapshots, each covering a sliver of the field at a slightly different angle. Turning that pile into one seamless, measurable map is a job called photogrammetry, and the drone platforms charge a monthly subscription to do it in their cloud. Here's what the processing step actually produces, the genuinely free way to run it yourself, and what to do with the map once you have it.

What "a drone map" actually is

Processing software matches the overlapping photos against each other, reconstructs where the camera was for every shot, and stitches the result into georeferenced files. Three outputs matter for farm work:

• Orthomosaic (ortho) — one seamless top-down photo of the whole field, corrected so distances are consistent across the frame. This is what most people mean by "drone map."
• Elevation surface (DSM) — a grid of heights for the surface the camera saw. Useful for spotting low ground and, flown twice, for before/after looks at dirt work.
• NDVI / vegetation index — a crop-vigor layer computed from near-infrared imagery. This one has a hardware catch: it needs a camera that captures near-infrared (a multispectral sensor or an NIR-converted camera). A standard RGB drone camera can produce pseudo-vigor layers, but they're not true NDVI.

The format to ask for — or export — is GeoTIFF: an image that carries its own map coordinates, so any GIS or mapping tool can place it on the right ground. A JPEG screenshot of a map has lost that, along with the real data values.

The free route: OpenDroneMap / WebODM

OpenDroneMap (ODM) is free, open-source photogrammetry software, and WebODM is its point-and-click web interface: drag your photos in, press start, and it produces the orthomosaic, elevation surface, plant-health layers, and a 3D model — with GeoTIFF export for the map layers. It's the real answer to "free drone mapping software," and the maps it makes are used in commercial work.

The honest catches:

• Setup is the hill. WebODM runs on your own computer, typically via Docker. The software is free; the projects also sell convenience installers for a one-time fee, which for many people is the sanest path. Either way, expect the first afternoon to go to setup, not mapping.
• Processing takes horsepower and time. Stitching hundreds of photos is heavy work — depending on your machine and photo count, a field can take from tens of minutes to several hours. It's a "start it and go do something else" job.
• You own the pipeline. The upside of the effort: no subscription, no cloud that holds your maps hostage, and nothing stops working when a plan lapses.

What about the paid platforms?

Cloud services like the big drone-mapping platforms make processing genuinely effortless — upload photos, get maps. If mapping is a weekly part of the operation, the subscription can be worth it. The trade-offs are the recurring cost, and that your maps live in their system; some offer free trials or limited free tiers, which are a fine way to see what the outputs look like before committing either way.

Fly so the photos can become a map

Processing can't fix a flight that didn't capture enough. Three things make the difference:

• Overlap. Photogrammetry needs each patch of ground in many photos — a common starting point is around 75% front overlap and 65% side overlap, flown as an automated grid (most drone apps have a mapping mode that handles this). Use more overlap over uniform crops, where the software has fewer features to match.
• Consistent height and one session. Fly the whole field at one altitude, ideally in even light — a cloud front moving through mid-flight shows up as a seam in the ortho.
• Ground control. Marked points with known positions are what pin the map to the real world — they tighten accuracy and, critically, make repeat flights line up with each other. Our field guide to ground control points covers how many you need and where to put them.

Putting the map on your fields — free while in beta

A GeoTIFF sitting in a downloads folder isn't doing much. Field Forge 3D's Ranch Mapper is a free web app that imports drone scans as GeoTIFF files — RGB orthos, NDVI, and elevation layers — and drapes them over your fields on real 3D terrain, next to the boundaries, acreage, and yield records you already keep there. Scans are stored with the field and available offline, so the map you processed in the shop is with you in the truck.

Once a field has scans from more than one date, the Change detection (beta) tool can compare two of them — where vigor improved or declined between NDVI flights, or where dirt was cut and filled between elevation flights. (That workflow gets its own field guide.) One practical tip that pays off later: import the original single-band GeoTIFF exports from your processing software, not screenshots, so the full-precision data is there when you want to compare flights.

The mapper is free while in beta — nothing locked today — no sign-up needed to start. To be clear about the division of labor: the mapper displays and analyzes drone maps; it doesn't process the photos. WebODM (or your platform of choice) makes the map; the mapper puts it to work on your ground.

What the map can't tell you

• Accuracy depends on ground control. Without it, a drone map can be beautifully sharp and still sit several feet from where it belongs — fine for scouting, not for anything that depends on absolute position.
• The elevation surface includes the crop. A DSM is the surface the camera saw, canopy and all — heights over standing crop are crop heights, not ground.
• Vigor layers are relative. NDVI shifts with sensor, light, and crop stage. Treat a vigor map as where to walk first, not a diagnosis or a prescription.
• A map is a snapshot. It shows the field on flight day; the value compounds when you fly consistently and compare over time.

The short version

Fly an automated grid with good overlap and consistent ground control, process the photos with WebODM (free, runs on your own machine, exports GeoTIFF), and load the results somewhere they live next to your field records instead of in a downloads folder. That's a working drone-mapping pipeline with no subscription in it — the only real costs are the drone you already own and one afternoon of setup.