Photogrammetry: verified survey data and clear outputs for confident design, planning and delivery decisions.
Photogrammetry reconstructs three-dimensional information from overlapping photographs taken from different viewpoints. Software identifies corresponding features across the images and uses them to estimate camera positions, generate spatial data and reconstruct visible surfaces.
The resulting model can contain exceptionally rich colour and texture, but visual quality alone does not prove survey accuracy. Scale and geometric confidence depend on the capture geometry, image quality, control, processing and the way the output is checked.
Geo Scan therefore treats photogrammetry as part of a survey methodology. Depending on the project, we can use surveyed targets, known scale, independent check information or TLS to provide a measured reference. Our Calderstones case study shows this hybrid approach in practice.
Photogrammetric performance is project-specific. Image resolution, camera geometry, overlap, focus, lighting, surface texture, control distribution and processing all affect the result, so Geo Scan does not apply one blanket accuracy figure to every photogrammetry survey.
Reflective, transparent, moving, repetitive or low-texture surfaces can be difficult to reconstruct reliably. Occlusion also remains important: a surface that is not adequately photographed cannot be reconstructed simply because neighbouring areas are visible.
Where the project requires stronger dimensional confidence, we can introduce surveyed control or use terrestrial laser scanning as the measured framework. The right method is the one that captures the required information at an appropriate level of accuracy for its intended use.
The capture strategy is designed around the subject, required resolution, accuracy, access and final output.
We establish coverage, lighting, image overlap, control, scale and the required deliverables.
Systematic overlapping imagery is collected from suitable viewpoints, with targets or measured reference where required.
Images are processed into spatial data and related to the agreed control or TLS framework.
Coverage, scale and output suitability are reviewed before meshes, point clouds or orthographic products are issued.
There is no single accuracy figure that applies to every photogrammetry project. Performance depends on image scale and quality, camera geometry, control, subject, processing and checks. We agree the required use first and design the capture methodology around it.
Neither is universally better. Photogrammetry is particularly strong for rich visual surface information, while terrestrial laser scanning provides direct range measurements and is often the stronger framework for dependable building geometry. Many projects benefit from combining them.
Yes. Survey-controlled photogrammetry can be aligned with terrestrial scan data so photorealistic surface information and measured geometry contribute to the same project workflow.
Yes. Where the dataset is appropriately scaled and oriented, it can support orthographic imagery for façades, heritage surfaces and other suitable subjects.
Yes. A scaled orthophoto can be used as a drawing underlay, with vector CAD geometry and annotations added only where the brief needs them. This can retain a detailed visual record while making drawing production more proportionate.
Yes. Planned drone imagery can extend coverage across roofs and high-level areas, while terrestrial laser scanning and survey control can provide a measured reference framework. The capture and checks must be designed around the required output and use.
Both can be image-led, but they are different representations with different strengths. Photogrammetry has established workflows for reconstructing explicit geometry and orthographic products. Gaussian splatting is particularly compelling for novel-view visual rendering. We do not treat visual realism by itself as proof of survey accuracy.