Differences between Point Cloud Scanning and Conventional Surveying, and How to Use Them

Table of Contents

• What is point cloud scanning?

• Basics of conventional surveying (TS, level, GNSS)

• Differences in mechanisms between point cloud scanning and conventional surveying

• Advantages and disadvantages of each method

• How to choose by application and accuracy requirements - Use cases in earthworks - Use cases in as-built control - Use cases in disaster recovery - Use cases in ledger/database maintenance

• Barriers to adopting point cloud scanning and simple surveying with LRTK

• Frequently Asked Questions (FAQ)

What is point cloud scanning?

In recent years, 3D scanning technology known as point cloud scanning has attracted attention on surveying sites in the civil engineering and construction industries and in local governments. Point cloud scanning is a method that measures terrain and structures as a collection of countless points (point cloud data) using laser scanners or photogrammetry. Each point includes three-dimensional coordinates (X, Y, Z) and color information, and the density of points can reproduce the actual shape with high accuracy. In other words, it is a technology that obtains a “digital replica of the site” that copies the site’s shape in its entirety.

Representative methods for acquiring point cloud data are measurement with laser scanners and point cloud generation by photogrammetry. Laser scanners rapidly acquire countless range points by emitting laser light from the device, having it reflect off surrounding objects, and detecting the reflections with a sensor. Photogrammetry (SfM and similar techniques), on the other hand, analyzes images taken from multiple angles by drones or cameras and computes three-dimensional coordinates based on common points. Point clouds from laser measurement have the advantage of being highly accurate because distances are measured directly, while photogrammetry-derived point clouds are visually easier to interpret because each point is colored based on images. Both methods share the ability to digitalize site geometry in detail and are used selectively according to purpose.

The biggest advantage of point cloud scanning is that it can measure large areas in a short time. Complex terrain and large-scale structures that were difficult to survey manually can be measured non-contact from a distance with lasers. With drone-mounted laser scanners or vehicle-mounted mobile mapping systems (MMS), even dangerous areas where people cannot enter and vast sites can be quickly digitized in 3D. Recently, small LiDAR sensors have begun to be installed in smartphones and tablets as well, making them useful for scanning narrow interiors and small structures. Acquired point cloud data can be utilized in various tasks by comparing it with the results of conventional surveying or design data described later.

Basics of conventional surveying (TS, level, GNSS)

Compared to point cloud scanning, methods referred to as conventional surveying include total stations (TS), levels, and GNSS surveying. These technologies have been the mainstream of surveying for many years, each with different measurement principles and strengths.

• Total Station (TS): An optical surveying instrument that became widespread after the 1980s. A single unit measures horizontal angle, vertical angle, and slope distance to compute coordinates. It measures one point at a time by targeting prisms or other targets, allowing millimeter-level high-precision measurements (millimeter-level (0.04 in)). However, clear line-of-sight between the instrument and the target is required, and work typically involves two or more people (an operator and a person holding the target). TS is widely used from terrain surveys to setting out on construction sites and remains indispensable today for control point surveys and detailed precision measurements.

• Level (auto level): An optical or digital leveling instrument specialized for measuring height (elevation) differences. By reading graduations on a staff through a telescope leveled horizontally, height relationships among multiple points are determined. Levels offer particularly high vertical measurement accuracy and can keep errors to a few millimeters or less even over survey lines of several kilometers, so they are used for checking road gradients and transferring base elevations. Although they cannot measure horizontal positions, their simple procedures and immediacy make them a standard method for high-precision elevation surveys.

• GNSS surveying: A method of determining position by receiving radio signals from artificial satellites (such as GPS). Practical implementation progressed from the late 1980s, and today RTK relative positioning with reference stations can provide real-time centimeter-level accuracy (centimeter-level (0.4 in)). GNSS surveying enables global positioning, making it powerful for long-distance surveys between sites and linking to public coordinate systems. It does not require line-of-sight and can be operated by a single person, but signal reception can be unstable under tree cover or near tall buildings due to the nature of radio waves. Also, vertical accuracy is somewhat inferior to horizontal, so combining levels for important elevation control is recommended.

As described above, TS, levels, and GNSS each have strengths and weaknesses and are used according to site conditions. In conventional surveying, experienced surveyors typically work in pairs, set up tripods, operate instruments, and measure each point reliably; this basic style remains important even in the current digital age.

Differences in mechanisms between point cloud scanning and conventional surveying

From the perspective of measurement approach, point cloud scanning and conventional surveying differ significantly. Total stations and GNSS “measure individual points sequentially,” whereas point cloud scanning gives the impression of “measuring surfaces all at once.” With a TS, the operator acquires coordinates of targeted specific points one by one, but with laser scanners or photogrammetry, setting up the device and executing a scan acquires millions of measurement points around the area at once. Therefore, the data volume is orders of magnitude larger, and the information obtained covers the entire site.

Another difference is that point cloud scanning is non-contact and covers wide areas. Conventional surveying often required people to enter the measurement points or set up staffs at each point. With point cloud scanning, lasers can be fired from a distance or images captured from above, enabling measurement of hazardous locations or steep slopes without people approaching. Because large areas can be covered quickly, there is a significant difference in work efficiency.

However, since the positioning principles differ, the nature of the data obtained also differs. GNSS directly provides latitude and longitude in a global coordinate system, while a standalone laser scanner only provides relative shape data (coordinates referenced to the scanner). To give real-world coordinates to acquired point clouds, post-processing georeferencing to known points is necessary. Typically, a few control points measured beforehand by TS or GNSS are used to align the point cloud to that coordinate system. By adding accurate positional information to point cloud results through such hybrid surveying, the data becomes practical 3D survey results.

In summary, point cloud scanning is “a method to acquire the entire space quickly,” while conventional surveying is “a method where people measure each point reliably.” Although the methods differ, using them complementarily enables high-accuracy and efficient surveying.

Advantages and disadvantages of each method

Below is a practical summary of the merits and demerits of point cloud scanning and each conventional method.

• Point cloud scanning: - Advantages: Can acquire terrain and structures over wide areas in a short time, dramatically improving work efficiency. Complex shapes can be digitalized in full detail, and acquired data can be used for 3D models, drawing creation, quantity calculations, and other purposes. Because it allows non-contact measurement without people entering hazardous areas, it is also superior in safety. - Disadvantages: Initial introduction costs are high due to the need for dedicated 3D scanners and high-performance PCs. Point cloud files are very large, requiring skills in data processing and management. Lasers may not reflect well off black or glass surfaces, and areas behind occlusions cannot be acquired. It is not suitable for checking dimensions at very limited single points, where TS or levels may still be required.

• Total Station (TS): - Advantages: Extremely high measurement accuracy per point, measuring distances and angles with millimeter-level accuracy (millimeter-level (0.04 in)). Strong for tasks requiring strict accuracy management, such as setting out from known points and displacement monitoring. The instrument cost is generally lower than that of 3D scanners, making it easier to utilize existing assets. Procedures are well established, and surveyors are familiar with it. - Disadvantages: Because points are measured one by one manually, efficiency is low and large-area surveys require significant time and effort. Capturing the entire terrain requires increasing the number of measured points, but practical limits exist and omitted areas may be discovered later. Line-of-sight is essential, so obstacles can prevent measurement. Typically two or more personnel are required, posing challenges in labor cost and staffing.

• GNSS surveying: - Advantages: Because positioning is obtained from satellites, measurement is not constrained by line-of-sight or distance. Sites far apart can be measured in the same coordinate system, useful for wide-area control point surveys and absolute coordinate checks. Using RTK provides centimeter-level accuracy in a short time, and a single operator can quickly collect many points for terrain mapping or route surveys. - Disadvantages: Accuracy depends on satellite signal reception; in mountainous areas, urban canyons, or under overpasses accuracy may degrade or a fixed solution may not be obtainable. Standalone GNSS needs association with reference elevations, and ensuring elevation accuracy requires additional measures. Although devices have become smaller and cheaper, communication environment setup and satellite positioning knowledge are required, so a certain learning cost is involved.

• Level surveying: - Advantages: Specialized in elevation difference measurement, providing exceptionally high vertical accuracy. It remains indispensable for managing construction base elevations and settlement monitoring. Instruments are robust and simple, so they rarely fail and are relatively easy to operate even by non-experts. Equipment costs are low, and analog models do not require power, eliminating battery concerns. - Disadvantages: Measures only elevation differences, so separate methods are needed for horizontal positioning. Because leveling is typically done between two staffs at a time, it is inefficient for wide-area surveys and requires two or more people. While elevation differences can be obtained, fine terrain irregularities cannot be captured, and detailed surface shape data like point clouds cannot be obtained.

As described, point cloud scanning and each conventional method have their own strengths and weaknesses. For example, point cloud scanning is overwhelmingly superior in efficiency, but conventional methods may win in precision control and pinpoint measurements. On sites, combining these methods complementarily—quickly acquiring wide areas with point clouds and confirming critical points with TS or levels—is ideal.

How to choose by application and accuracy requirements

In actual surveying work, it is important to select the optimal method according to the type of project and required accuracy. The following describes how to use point cloud scanning and conventional surveying in typical scenarios.

Use cases in earthworks

In land development and cut-and-fill operations, pre- and post-construction terrain surveys and earthwork volume calculations are important. On large development sites, drone photogrammetry or terrestrial laser scanners can scan the entire area to acquire current terrain point cloud data in a short time. For example, surveys that took several days with conventional TS on sites of several hectares can sometimes be completed in about half a day with a drone. From acquired point clouds, contour maps and longitudinal/cross sections can be automatically generated, and earthwork volume calculation using terrain models is speedy.

However, for the precise setting out of control stakes or accurate installation of structures, high-accuracy positioning by TS or GNSS remains essential. On earthworks sites, it is common to first establish control points with GNSS or TS and then capture the entire surface with drones or lasers. For progress management of cut-and-fill, periodic point cloud scans are used to record construction progress in 3D, and differences from the design model are used to check for excesses or shortages in executed volumes. Because point clouds capture the landform comprehensively, they provide flexibility to accommodate later design changes without additional surveying.

In summary, for wide-area, terrain-focused sites like earthworks, it is effective to improve efficiency with point cloud scanning while combining conventional surveying for control and verification of critical points.

Use cases in as-built control

As-built control is the task of inspecting whether completed structures or landforms conform to design shapes and dimensions. Traditionally, measurements were taken at prescribed locations for height and thickness using TS or tape measures and compared with design requirements. Using point cloud scanning, however, the entire finished structure can be checked in full.

For example, when inspecting pavement thickness or slope on a road, scanning the entire completed site and overlaying the resulting point cloud with the design 3D model allows verification of height and thickness error distributions at all points. Applying a heat map (color-coded error display) to the point cloud makes it intuitive to identify areas that are too high or too low, smoothing inspection meetings. Point clouds enable surface-based error analysis of complex shapes that were difficult to evaluate previously (e.g., curvature of dome tanks or tunnel arch shapes).

Because strict accuracy is required in as-built inspections, point cloud measurements must be performed with adequate control point calibration and high-precision equipment. It is effective to perform additional TS measurements at important locations for double-checking errors against the point cloud. National guidelines now define 3D as-built management requirements, and if accuracy is met, point cloud data can be submitted as as-built inspection deliverables. Introducing point cloud scanning can shorten inspection time and improve quality.

Use cases in disaster recovery

Point cloud scanning is also prominent at disaster recovery sites such as landslides and river flooding. Immediately after a disaster, sites are dangerous and detailed manual surveying into the depths of the area is difficult. Drone aerial photography or terrestrial laser scanning can remotely measure the entire affected area and quickly obtain a point cloud model of the collapsed terrain. This allows rapid estimation of collapsed soil volume and damage extent, aiding consideration of recovery methods and calculation of necessary soil removal quantities. In one large-scale debris flow case, comparing pre- and post-event terrain point clouds allowed calculation of collapse volumes and served as the basis for recovery planning.

Speed and safety are top priorities in disaster recovery, and point cloud scanning effectively delivers both. Helicopter or high-altitude laser measurement can capture current conditions without sending people into hazardous zones. The obtained 3D data can be shared with relevant agencies as a record of damage, aiding consensus in recovery planning. However, in emergencies drones may be grounded by weather, and when local detailed checks are required, additional TS surveying is performed. For example, to measure tilt of damaged houses or displacement of bridge piers, conventional methods remain effective.

Overall, point cloud scanning is effective for initial rapid surveys in disaster recovery, with conventional surveying used in combination for detailed structural evaluation and design. Also noteworthy is that point cloud data makes difference analysis with pre-disaster records (ledgers, etc.) easy.

Use cases in ledger/database maintenance

Point cloud scanning is used in ledger maintenance (updating records of infrastructure conditions) for roads, rivers, and structures. Local governments need to periodically survey sites to update drawings and data for managed road and sewer ledgers. Traditionally, many staff walked sites and measured dimensions with distance meters or TS and manually edited drawings, but introducing point clouds has greatly improved this process.

For example, one municipality used a vehicle-mounted mobile mapping system (MMS) to drive through the city and collect point cloud data and 360-degree imagery of road surfaces and surrounding structures. As a result, they reported approximately a 46% reduction in man-hours for road ledger updates. Because wide-area infrastructure can be digitally measured in a short time, this approach helps address labor shortages. From acquired point clouds, plan views and cross sections of roads can be automatically generated, and pavement wear and distribution of steps can be assessed.

Ledger maintenance requires final organization into drawings and databases, but point clouds enable remote measurement in the office. If something was missed on site, additional dimensions can be extracted from the point clouds, reducing the need for re-surveys. However, some government offices require information to be extracted in existing 2D drawing formats, so converting key points from current point clouds into TS-like deliverables is necessary. Again, combining point clouds with conventional methods is effective: measure main control points with GNSS for absolute coordinates while acquiring detailed shapes with point clouds for drafting.

Thus, point cloud scanning is powerful for ledger tasks requiring wide-area current-condition assessment and document creation, supporting infrastructure management with unprecedented speed and comprehensiveness.

Barriers to adopting point cloud scanning and simple surveying with LRTK

Although point cloud scanning is convenient, many feel barriers such as “equipment looks expensive” or “doesn’t it require advanced expertise?” Indeed, conventional 3D laser scanner systems once cost several million yen, and drone photogrammetry also required high-performance aircraft and software. High-spec PCs and processing skills were required to handle the resulting point clouds, making adoption difficult for small and medium enterprises and local governments.

However, new solutions have recently emerged to lower these barriers. One such solution is simple surveying using LRTK. LRTK is a small positioning device that attaches to a smartphone and turns the phone into a high-precision surveying instrument. Weighing only a few hundred grams and pocket-sized, it contains a high-sensitivity GNSS antenna and an RTK receiver, and in conjunction with a smartphone app can achieve real-time centimeter-level positioning (centimeter-level (0.4 in)). When combined with a smartphone’s built-in LiDAR sensor (on supported models), anyone can easily perform point cloud scanning with positional coordinates.

Using LRTK for smartphone surveying eliminates the need to carry heavy tripods and large instruments, allowing surveys and scans on the spot whenever needed. For example, a single person can collect surrounding point clouds simply by holding up a smartphone, and accurate coordinates are attached simultaneously so post-processing alignment is unnecessary. Acquired data can be shared instantly via the cloud, and 3D point clouds can be viewed in a browser without specialized software. In short, even without an expensive laser scanner, high-precision 3D surveying can be achieved with just a smartphone and LRTK.

With such affordable and easy solutions emerging, point cloud technology adoption is beginning among small-scale contractors and local governments that previously had little contact with it. Tools like LRTK are compatible with ICT construction and i-Construction initiatives promoted by the Ministry of Land, Infrastructure, Transport and Tourism, and acquired point cloud data can meet the accuracy required for as-built management deliverables. Now that “easy point cloud surveying anyone can use” is a reality, opportunities to benefit from point cloud scanning will expand further. With adoption barriers lowered, now is a good time to explore how point cloud scanning can be applied to your organization’s work.

Frequently Asked Questions (FAQ)

Q: Can point cloud scanning completely replace conventional TS or leveling surveys? A: Completely replacing them is difficult. Point cloud scanning can efficiently measure wide areas, but setting high-accuracy control points and measuring specific points precisely are tasks better suited to total stations and levels. In practice, a hybrid operation—using point clouds for overall understanding and conventional surveying to supplement critical parts—is mainstream. It is helpful to think of point cloud scanning as a technology that strengthens, rather than makes obsolete, conventional methods.

Q: What level of accuracy can be obtained with point cloud scanning? A: It depends on the equipment and method used, but laser scanner measurements often fall within an error range from several centimeters to several millimeters (from several centimeters (several 0.4 in) to several millimeters (several 0.04 in)). With high-end instruments and proper post-processing, millimeter-level accuracy (millimeter-level (0.04 in)) suitable for as-built control can be achieved. However, ensuring point cloud accuracy requires calibration with control points and integration of multiple measurements. Photogrammetry-derived point clouds tend to be less accurate than laser-derived ones, so using laser measurement for important inspections may be necessary. In short, point cloud scanning can be highly accurate, but final error management calls for combining conventional surveying and careful quality checks.

Q: Can point cloud surveying be done without an expensive 3D laser scanner? A: Yes. In recent years, point cloud data can be obtained via drone photogrammetry or simplified scans using smartphone LiDAR. Especially by using the small device LRTK attached to a smartphone, you can obtain high-precision point clouds with positional coordinates that were difficult with the smartphone alone. These methods are lower cost and easier than using dedicated scanners, making them suitable for those trying point cloud surveying for the first time. It is advisable to try drone photogrammetry or smartphone-LiDAR scanning on small sites to gain experience before considering full-scale laser scanner adoption. The barriers to point cloud technology are certainly decreasing, so look for opportunities to apply it in your work.