LiDAR Survey

LiDAR, short for Light Detection and Ranging, measures distance by timing laser pulses reflected from surfaces. Millions of such measurements, combined with precise positioning and orientation, form a point cloud: a dense three-dimensional record of terrain, structures and vegetation that can be processed into terrain models, contours and extracted features.

Overview

LiDAR has changed how large areas and complex structures are measured. Instead of selecting and observing individual points, a scanner samples surfaces at very high rates, producing a dense digital record that can be measured, sectioned and re-analysed long after the fieldwork is complete. This makes it well suited to corridors, large sites, complex facilities and situations where the full geometry of a surface matters.

There are three broad ways LiDAR data is captured. Terrestrial laser scanning uses a tripod-mounted scanner set up at multiple positions, with scans registered together; it is used for structures, facades, plant rooms and detailed site areas. Mobile mapping mounts a scanner on a vehicle, trolley or backpack, with positioning from GNSS and an inertial measurement unit; it is efficient for roads, railways and urban corridors. Airborne and drone LiDAR mount the sensor on an aircraft or unmanned aerial vehicle, covering wide areas and, importantly, recording multiple returns that can partly penetrate vegetation canopy to reach the ground.

This page describes LiDAR as a general technology: how it works, the products derived from it, the methodology typically followed and the factors that affect quality. It is educational content to help clients scope LiDAR requirements. Specific capture methods, sensors and platforms for any project are confirmed in the scope of work, and SGTS does not claim ownership of particular scanners or drones on this website.

The quality of a LiDAR deliverable depends on the sensor, the positioning and orientation solution, ground control and check points, point density, the classification process and the experience applied in turning raw points into usable products. A dense point cloud is not automatically an accurate one; control, calibration and independent checks are what make it dependable.

How it works

A LiDAR sensor emits laser pulses and measures the time taken for each pulse to return after striking a surface. Because the speed of light is known, the return time gives the range to that surface. A scanning mechanism sweeps the beam across a field of view, so that a very large number of ranges are measured in different directions every second.

Each range only becomes a three-dimensional coordinate when the position and orientation of the sensor at that instant are known. For terrestrial scanning, the scanner position is fixed and scans from different positions are registered using targets, overlapping geometry or surveyed control. For mobile and airborne systems, a GNSS receiver and an inertial measurement unit continuously record the platform's trajectory and attitude, and the ranges are combined with this trajectory to compute coordinates.

Many sensors record more than one return per pulse. When a pulse passes through vegetation, part of its energy may reflect from leaves and branches while the remainder continues to the ground. Recording first, intermediate and last returns allows ground points to be separated from vegetation, which is why airborne LiDAR is often used to model terrain beneath tree cover. Return intensity, a measure of reflected signal strength, is also recorded and helps distinguish materials such as road markings.

Raw point clouds are processed through several stages: trajectory processing, strip or scan alignment, adjustment to ground control, noise removal and classification. Classification assigns each point to a category such as ground, low vegetation, high vegetation, building or water. The ground class is used to create a Digital Terrain Model (DTM), representing the bare earth. All first returns, including buildings and canopy, are used to create a Digital Surface Model (DSM). Contours, sections, volumes and extracted vector features are then derived from these models.

Accuracy assessment is carried out by comparing the LiDAR-derived surface with independently surveyed check points, typically observed by total station or GNSS on hard, open surfaces. The results of this comparison, rather than the nominal sensor specification, indicate how reliable the final product is for the project's purpose.

Applications

Corridor mapping

Capturing road, rail, canal and transmission line corridors for design, widening and asset assessment.

Terrain modelling

Producing DTMs and contours for planning, drainage design and earthworks, including areas under vegetation.

Volume calculation

Measuring stockpiles, cut and fill and excavation quantities from successive surfaces.

Structure and facility capture

Documenting bridges, buildings, plant and complex structures for retrofit design and as-built records.

Clearance analysis

Measuring clearances between conductors, structures, vegetation and ground in utility and rail corridors.

Flood and drainage studies

Supplying terrain surfaces as input for hydrological and drainage modelling.

Industries

Infrastructure
Roads and highways
Rail and metro
Power
Telecom
Industrial
Government
Engineering consultants

Methodology

Requirement definition

Agree the purpose, area, coordinate system, required point density, accuracy expectations and products such as DTM, DSM or contours.

Method selection

Select terrestrial, mobile or airborne capture based on area, terrain, vegetation, access and required detail.

Permissions and planning

Identify permissions needed for site access or flight operations, and plan scan positions, drive routes or flight lines.

Ground control

Establish control and independent check points with total station or GNSS for adjustment and accuracy assessment.

Data capture

Acquire data with sufficient overlap and coverage, recording field conditions and any gaps or occlusions.

Registration and adjustment

Process trajectories, register scans or strips and adjust the point cloud to control.

Classification and product generation

Classify points and derive DTM, DSM, contours, sections or extracted features to the agreed specification.

Quality check and reporting

Compare products with check points, review classification and document accuracy, coverage and limitations.

Equipment and technology

LiDAR systems vary widely in range, pulse rate, accuracy and number of returns. Terrestrial scanners are optimised for detail at short to medium range; mobile systems integrate scanners with GNSS, inertial units and often cameras; airborne and drone sensors are designed for wide-area capture and vegetation penetration. The platform and sensor suitable for a project are confirmed in the scope; SGTS does not state ownership of specific scanners or drones on this website.

Processing relies on trajectory software, point cloud registration and classification tools, and terrain modelling software. Outputs are commonly delivered as LAS or LAZ point clouds, raster DTM and DSM grids, contour lines in CAD or GIS format, and extracted vectors.

Ground control surveyed by total station or GNSS remains essential. It ties the point cloud to the project coordinate system and provides the independent check points used to verify accuracy.

Benefits and considerations

Benefits

Dense capture records complete surface geometry, not just selected points.
Airborne and drone LiDAR can model terrain beneath vegetation through multiple returns.
Data can be revisited to extract additional sections or features without returning to site.
Efficient for long corridors and large areas compared with point-by-point methods.

Considerations

Large data volumes require appropriate processing, storage and software for clients to use.
Water, glass and very dark surfaces can return poor or no data.
Occlusions leave gaps where the scanner could not see; coverage planning reduces but does not remove them.
Airborne operations require permissions and are affected by weather and airspace restrictions.
Accuracy depends on control, calibration and processing, and is verified with check points rather than assumed.

What to prepare

Define the purpose of the data and the products you need, such as DTM, contours, sections or a full point cloud.
Share the area or corridor boundary as a KML, shapefile or marked-up drawing.
Tell us the coordinate system and height datum the point cloud must be delivered in, and share any existing control.
Advise on vegetation, access constraints, security restrictions and sensitive areas.
Indicate the software you will use, so that formats and file sizes are practical for your team.
Allow time for any permissions required for site access or flight operations.

Limitations

LiDAR measures surfaces visible to the sensor. It does not detect underground features.
Dense vegetation can reduce the number of ground returns, lowering terrain model reliability in those areas.
Water bodies generally return little usable data, and bathymetry requires specialised sensors.
Point cloud accuracy depends on control, calibration and processing; nominal sensor specifications alone are not a measure of deliverable accuracy.

Frequently asked questions

What is the difference between a DTM and a DSM?

A Digital Terrain Model represents the bare ground with buildings and vegetation removed. A Digital Surface Model represents the top of everything the sensor detected, including roofs and tree canopy.

Is LiDAR better than photogrammetry?

They suit different needs. LiDAR records multiple returns and can model ground under vegetation, while photogrammetry produces imagery-rich outputs and works well on open, textured surfaces. Many projects combine them.

Which LiDAR method will my project use?

The method depends on area, terrain, vegetation, access, required detail and permissions. It is agreed in the scope after reviewing the requirement.

What file formats are typical?

Point clouds are usually LAS or LAZ, terrain models are delivered as raster grids such as GeoTIFF, and contours and features in CAD or GIS formats. Formats are agreed with the client.

Does LiDAR need ground control?

Yes. Ground control ties the data to the project coordinate system and check points are used to verify the accuracy of the final products.