Drone LiDAR and photogrammetry can provide complementary information for industrial mapping. LiDAR measures ranges to surfaces; photogrammetry estimates geometry from overlapping images. Their outputs can be combined, but the accuracy of a fused product depends on control, calibration, flight planning, surface conditions, coordinate reference systems and processing. Neither method guarantees centimetre-level accuracy without a defined survey design and independent validation.
For a related explanation of camera sensor and computational capture pipelines, see computational imaging.
Choose the method for the site
LiDAR can provide useful geometry where image texture is weak or where some surfaces are partly obscured. Photogrammetry can provide dense visual texture and colour when image overlap, lighting and surface features are suitable. Reflective, transparent, repetitive or moving surfaces can challenge both methods in different ways.
For the workstation and storage capacity considerations involved in large visual datasets, see studio infrastructure for 12K production.
Survey design and control
Record the aircraft, sensor payload, lens, GNSS/IMU system, flight height, speed, overlap, scan rate, camera settings and coordinate reference system. Establish ground control and independent check points appropriate to the required accuracy class. Record the control survey method and uncertainty. Do not use the same points for adjustment and independent accuracy reporting without clearly distinguishing their roles.
Calibration and fusion
Calibrate the relationship between the camera, LiDAR sensor and navigation system. Account for lever-arm offsets, boresight angles, time synchronisation and coordinate transformations. Validate registration using independent surfaces or control points. A visually aligned point cloud is not necessarily metrically accurate.
For a related architecture covering ingest, active storage, processing and archive, see scalable media architecture.
Accuracy assessment
Report horizontal and vertical residuals at independent checkpoints, point density, completeness, occlusion and any filtering or classification applied. Separate absolute accuracy from relative precision and local surface consistency. State the test distribution, sample size, confidence level and any areas excluded from evaluation.
Quality-control workflow
- Plan mission and control against the project accuracy requirement.
- Capture calibration data and record environmental conditions.
- Process LiDAR and imagery separately before fusion.
- Check registration and coordinate reference systems.
- Validate the fused product against independent checkpoints.
- Document uncertainty, occlusions, gaps and unsuitable surfaces.
For public-sector elevation projects, consult the USGS 3D Elevation Program standards and specifications as a reference for data quality and deliverable requirements. Project-specific survey standards and jurisdictional requirements still apply.
Conclusion
LiDAR and photogrammetry can complement one another, but accuracy is established through survey design and independent validation. Do not publish a centimetre-range result unless the payload, control network, processing and checkpoint statistics support it.