Drone Mapping in Kenya: Photogrammetry, LiDAR, and Open Processing That Delivers

Drone mapping has moved from novelty to standard toolkit on Kenyan sites—construction progress, quarry volumes, corridor reconnaissance, estate topography, and environmental baselines. The question is less “should we use a drone?” and more “which method, which processing stack, and how do we keep the product survey-grade and legally flyable?”

This guide is written for project managers, engineers, surveyors, and county/technical teams in Kenya. It emphasises open processing where it makes sense, honest limits of photogrammetry versus LiDAR, and how aerial products must still tie to ground control.

What drone mapping is good for (and what it is not)

Strong fits: dense topographic surfaces, stockpile and cut/fill volumes, site progress orthophotos, drainage and earthworks checks, rapid corridor context, and visual as-builts when controlled properly.

Weak or inappropriate fits: replacing cadastral beacon definition, “surveying” a title boundary from imagery alone, or claiming centimetre accuracy without surveyed ground control and a documented method.

Treat the UAV as a high-productivity sensor platform. The survey truth still comes from control, calibration, and professional judgement—especially where decisions affect money, safety, or land rights.

Kenya operations: fly legal, fly planned

Commercial UAV work in Kenya sits under civil aviation rules and evolving operational approvals. Exact requirements depend on aircraft class, location, altitude, and whether you fly near aerodromes or restricted areas. Build compliance into the project plan—not as an afterthought the morning of the flight.

A professional flight pack typically includes:

• Site risk assessment (people, roads, power lines, livestock, privacy)
• Airspace and aerodrome checks
• Pilot competency and aircraft airworthiness evidence
• Insurance appropriate to commercial operations
• Emergency procedures and a clear go/no-go weather brief

If a bidder cannot describe how they stay legal on your site, they are not ready for your project.

Photogrammetry vs LiDAR: choose by canopy and accuracy need

Photogrammetry (RGB / multispectral cameras)

Camera-based Structure-from-Motion remains the workhorse for clear sites: estates, roads under construction, open quarries, and many urban parcels. Outputs usually include orthomosaics, dense point clouds, DSMs/DTMs (with caveats), and textured meshes.

Strengths: cost, visual context, fast revisit. Limits: vegetation hides ground; thin objects and uniform surfaces can fail; accuracy collapses without good overlap and GCPs/checkpoints.

UAV LiDAR

LiDAR earns its keep under canopy, for powerline/corridor work, and when you need denser ground returns than photogrammetry can invent. It costs more to acquire and process, and still needs trajectory quality and surveyed control.

Many Kenyan projects start with photogrammetry and escalate to LiDAR only when the canopy or engineering specification demands it.

Hardware vendors: capable airframes, open data out

Common platforms in the region include DJI enterprise and mapping-oriented aircraft, plus other RTK-enabled mapping UAVs from established survey brands. RTK/PPK on the aircraft helps—but it does not remove the need for independent checkpoints.

When evaluating kit or a service provider, ask:

• Can we export raw images, flight logs, and point clouds in open formats?
• Is the camera/lidar calibrated and is that calibration documented?
• How are GCPs measured (and by whom)?
• What GSD and accuracy were achieved on comparable Kenyan sites?

Avoid workflows that only deliver a locked viewer with no GeoTIFF, LAS/LAZ, or CAD/GIS exchange.

Processing stacks: open-source first, commercial when needed

This is where open tooling has matured enough to be a serious default for many projects.

Open and open-friendly processing

• OpenDroneMap / WebODM — strong open photogrammetry pipeline for orthophotos, DEMs, and point clouds; good for teams that want repeatable, self-hosted processing without per-project licence drama.
• PDAL — point cloud pipelines (filter, classify, convert) that sit well beside PostGIS-centric data stores.
• CloudCompare — inspection, QC, and cloud-to-cloud checks.
• QGIS + PostGIS — packaging deliverables, overlays, contours, and long-term project GIS.
• GDAL — COGs, warping, and delivery formats that web maps and cloud analysis can consume.

Commercial photogrammetry (still valuable)

Agisoft Metashape, Pix4D, and vendor-linked cloud processors remain excellent—especially for specialised sensors, polished reporting, or teams already trained on them. Use them when they clearly save time or meet a client specification. Prefer exports that land cleanly in QGIS/PostGIS afterward.

Our bias for 2026 Kenya work: process in whatever engine is fit for purpose, archive open deliverables, manage the project in open GIS.

Ground control: the difference between a pretty map and a survey product

Best practice still looks like this:

1. Establish or densify survey control (see our land surveying guidance)
2. Lay GCPs sized and contrasted for the planned GSD
3. Reserve independent checkpoints (not used in the adjustment) for accuracy reporting
4. Report RMSE in X/Y/Z in the delivery note—not only “RTK drone”

For volumes and engineering surfaces, relative accuracy and a stable site datum often matter as much as absolute national coordinates. Write that into the brief.

Cloud-centric delivery without losing the files

Large orthos and point clouds do not belong only on a field laptop. A practical pattern:

• Store COGs / LAZ in object storage
• Publish web previews (MapLibre/Leaflet or a lightweight 3D viewer)
• Keep PostGIS vectors (breaklines, GCPs, site boundary, progress polygons) as the operational layer
• Optional: push derived indices or change detection into Earth Engine-style cloud analysis for regional context

Clients should receive both a viewing experience and downloadable open data.

Flight planning that protects quality

Automated missions are only as good as their parameters. Define up front:

• Target GSD and overlap (front/side)
• Altitude AGL vs terrain following needs
• Cross-strips on demanding surfaces
• Wind, lighting, and reflectivity risks (new tin roofs, water, bare sand)
• Battery and landing contingencies on congested sites

Re-flying because planning was sloppy costs more than an extra hour of prep.

A buyer’s checklist for Kenyan drone mapping

1. Is the use case photogrammetry, LiDAR, or hybrid?
2. Who provides surveyed GCPs/checkpoints?
3. What accuracy will be reported, and how?
4. Which open formats are delivered (GeoTIFF/COG, LAS/LAZ, DXF/GeoJSON)?
5. Where is processing done (on-prem WebODM, commercial, hybrid)?
6. Are aviation compliance and insurance documented for this site?

How Fayvad Geosolutions approaches drone mapping

At Fayvad Geosolutions, drone mapping is integrated with surveying and GIS—not sold as a standalone spectacle. We plan flights for the decision you need, control products properly, and favour open, reusable deliverables (orthos, clouds, and PostGIS-ready vectors) that plug into the same stacks we recommend for Kenya GIS programmes.

If you need volumes, progress mapping, or a controlled topographic model for design, we can help you scope the right sensor and processing path—without overselling LiDAR or underselling ground control.

Drone mapping services · Discuss a site brief