Table of contents

• Challenges and limits of traditional surveying

• Emergence of 3D design data and point-cloud technology

• What is point-cloud difference visualization?

• Innovation in as-built surveying: accurately grasping large areas in a short time

• Use in the design stage: improving planning accuracy and consensus with 3D models

• Use in construction management: catching construction errors with point clouds

• Advanced as-built inspection: visualizing quality with full-coverage measurement

• Simple surveying with smartphones + GNSS: LRTK supports the field

• FAQ

Challenges and limits of traditional surveying

In Japan’s construction industry, a severe labor shortage has become a pressing issue in recent years. Field work is often called the “3 Ks” (kitsu-i, kitanai, kiken — tough, dirty, dangerous), and the decline in young workers and aging workforce mean that traditional methods are becoming increasingly impractical for running sites. The Ministry of Land, Infrastructure, Transport and Tourism is promoting “i-Construction,” which aims to improve productivity on construction sites through ICT technologies and three-dimensional data, with a target of increasing on-site productivity by 20% by fiscal 2025. Against this backdrop, efficiency gains and technological advancement are strongly required across civil engineering processes such as surveying, design, and construction.

Traditional surveying methods in particular had many challenges and limitations. In standard topographic surveying, surveyors use instruments like total stations (electromagnetic distance meters) and levels, walking the site and measuring many individual point heights and distances one by one. Measuring a wide site requires enormous time and manpower, and surveying on steep slopes or poor footing can be hazardous. Also, the data obtained are only spaced point data, so fine undulations and surface irregularities between measured points are often not fully captured. Survey drawings represent the terrain with planar contour lines and numerical values, making it difficult to intuitively grasp complex three-dimensional site shapes.

In as-built (post-construction finish) inspections as well, the conventional approach was to check heights at several points on structures or embankments and confirm errors against design values. A limited number of sample points cannot accurately evaluate the overall finish and risks overlooking locally large deviations. Inspection on drawings required poring over tables of numbers and cross-sections, taking time to explain to the client. In this way, traditional surveying and management methods required considerable effort and time while providing limited information, leaving many issues in site understanding and quality control.

Emergence of 3D design data and point-cloud technology

The trump cards that have emerged to solve these problems are point-cloud surveying with 3D scanners and the digitization of design drawings into three-dimensional data. Advances in laser scanner and drone photogrammetry technologies now allow fast, high-density measurements of entire sites. Using a dedicated 3D laser scanner, you can mount it on a tripod, irradiate 360° with laser, and acquire millions of coordinate points in minutes. Equipping a drone with LiDAR (laser) or cameras and flying it overhead can produce detailed ground-surface point-cloud data for large sites in about 30 minutes to 1 hour. Because enormous numbers of points that would be impossible to measure manually can be acquired in a short time, dramatic efficiency gains are realized—surveys that once took two days can be completed in less than half a day. Dangerous cliffs and steep slopes can be measured remotely from a distance, allowing safe site assessment without personnel entering hazardous areas. Acquired point-cloud data can be immediately visualized as 3D models on office PCs, and you can later measure elevation or cross-sections at arbitrary locations or compute areas and volumes for analysis. It is now possible to digitally capture a safe, detailed “complete copy” of the site with minimal personnel and time.

Meanwhile, the design world is rapidly moving to 3D drawings. While 2D paper plans and cross-sections were traditionally dominant, the use of three-dimensional design data such as BIM/CIM is increasingly recommended even for public works like roads, rivers, and bridges. “3D design,” where design is performed on top of a 3D model of the existing terrain, is becoming widespread. Designers can work in 3D space with detailed terrain models generated from point clouds in the background to consider structure placement and earthworks planning. For example, placing a road alignment three-dimensionally on a surface model generated from point clouds allows accurate calculation of cut-and-fill volumes. In renovations of tunnels or dams, dimensions for new structures can be determined to avoid interference based on point clouds of existing structures. By combining as-built point-cloud data with 3D design data, designers can more easily understand deviations between the actual site and the design from the design stage, greatly reducing rework and mistakes.

What is point-cloud difference visualization?

So how do you compare acquired point-cloud data with 3D design data to see the “differences”? A representative method is point-cloud difference visualization (difference analysis). Simply put, you overlay the as-built point cloud obtained after construction on the ideal design shape in digital space, compute the displacement for each point, and display the results in color so that you can immediately see where and how much deviation exists. If the design data are available as three-dimensional surfaces (for example, a designed ground surface model or structure geometry), software can automatically compute distances between that surface and the point cloud. Coloring the point cloud according to deviation magnitude generates an intuitive heat map of differences.

In the heat map, you might set colors so that points matching the design elevation are blue to green, while areas that are overfilled or over-height are red and those overly excavated or underheight are also red (or otherwise color-coded). The entire as-built point cloud is therefore colorized, making it immediately clear which areas are higher or lower than the standard without consulting drawings or numeric tables. Using point-cloud difference heat maps, site managers and inspectors can grasp the finished condition across the entire space, significantly reducing the effort of comparing against design drawings and checking numbers.

You don’t always need special analysis software to perform this difference visualization. Recently there has been a proliferation of point-cloud processing software and cloud services for construction that can generate heat maps with a button click after loading point-cloud and design data. It is also easy to display point-cloud models and heat maps on tablets and take them to the site to check as-built conditions with stakeholders on the spot. Advanced examples include using AR (augmented reality) to project heat maps onto actual structure surfaces and check finishing accuracy in situ. Visualizing differences with color makes explanations to clients much easier and dramatically speeds up pass/fail judgments and corrective instructions. Because errors that were overlooked by conventional spot checks can be identified, the reliability of as-built management is greatly improved.

Innovation in as-built surveying: accurately grasping large areas in a short time

The introduction of point-cloud surveying using drones and 3D laser scanners has dramatically improved the efficiency of topographic surveying. Wide sites that previously took people days to survey can now yield hundreds of millions of ground points in about 30 minutes to 1 hour with a single drone flight. Since the ground is scanned continuously and measured as surfaces, subtle undulations and depressions that were overlooked before are reliably captured. Acquired data can be visualized on the spot as 3D models, and it’s easy to analyze elevations or cross-sections at arbitrary locations back in the office. The fact that terrain once captured only as discrete points can now be accurately captured as surfaces is revolutionary.

Shorter surveying times directly lead to shorter construction schedules and cost reductions. For example, if a site survey that once required 2–3 days can be completed in about half a day, labor and duration cost savings are significant and subsequent work can proceed earlier. Drone aerial imaging and remote laser measurement also allow safe, non-contact surveying of cliffs and steep slopes where human access is dangerous. Even riverbanks and disaster sites that humans cannot approach can be captured from the air, providing information on areas that were previously unmeasured. Point-cloud surveying, which acquires detailed data over wide areas quickly and safely, can be said to have fundamentally changed how we understand as-built conditions.

Use in the design stage: improving planning accuracy and consensus with 3D models

Detailed 3D terrain models created from as-built point-cloud data are extremely useful during the design stage. Traditionally, designers planned based on 2D drawings and a limited number of survey points, which sometimes led to interference or elevation issues becoming apparent only during construction. With a faithful 3D model of existing conditions generated from point clouds, you can review the actual terrain from the design stage. For example, even in highly uneven terrain, planning roads or earthworks on a surface model derived from point clouds prevents the mistake of “it didn’t match the site after construction.” Cut-and-fill volume calculations also become more accurate by comparing the entire as-built terrain and the design terrain using point-cloud data.

3D design proposals are also effective for building consensus with clients and stakeholders. Completion images that are hard to convey with plan views alone can be intuitively understood by overlaying design data on point clouds in three dimensions. When necessary, VR and AR technologies can be used to superimpose future structure models into the as-built point-cloud space. For example, using AR-capable smart glasses to overlay 3D models of a planned bridge or tunnel on the current site video lets all stakeholders share the completion image on the spot. By combining point-cloud data with 3D design data, design accuracy and stakeholder understanding deepen dramatically, enabling smooth planning without rework.

Use in construction management: catching construction errors with point clouds

During construction, point-cloud data is a powerful quality control tool. By scanning completed portions sequentially and comparing them with design data, construction accuracy can be inspected in detail. For instance, if you scan a concrete structure immediately after pouring and overlay that 3D model with the design BIM data, even slight dimensional differences or omissions can be found at a glance. One major construction company reportedly used this method to detect centimeter-level dimensional errors that previously would have only been discovered at final inspection, enabling early correction and greatly reducing rework. Incorporating continuous point-cloud comparisons into the construction process allows errors to be caught early, ultimately shortening schedules and reducing costs.

In road and earthwork projects, measuring embankment and excavation heights and slopes mid-process using point clouds enables “overfill” or “over-excavation” to be corrected on the same day. Tasks that once relied heavily on the operator’s intuition and experience are shifting toward data-driven management. Even less experienced personnel can evaluate finished results numerically by scanning the site and comparing with the design model, enabling construction management that does not depend on individual intuition. Regular acquisition of point-cloud data also helps overall progress management. If you drone-photograph the site weekly and convert to point clouds, you can accumulate 3D models of progress over time. Comparing this with the project schedule makes it easy to see which parts are ahead or behind. Because the latest point clouds can be shared via the cloud, site status can be confirmed and accurate instructions given remotely. Thus, point-cloud utilization visualizes construction site management in real time and contributes to improved quality and productivity.

Advanced as-built inspection: visualizing quality with full-coverage measurement

Point-cloud data also has great power in final as-built inspections. The conventional practice was spot inspections, measuring heights at specified locations and checking differences from design values. However, point-cloud surveying allows full-coverage measurement and inspection of entire structures and developed areas. Overlaying the as-built point-cloud model of the finished terrain or structure with the design 3D data makes it possible to verify in detail whether every location matches the design. Combined with heat-map difference visualization, even small deviations are shown in color and cannot be overlooked, allowing intuitive quality evaluation. Especially for structures like bridges and tunnels, point-cloud analysis can detect minute deformations and misalignments down to the millimeter level, helping to ensure construction quality.

Point-cloud-based as-built inspection also positively affects client inspections and approval processes. The point-cloud data obtained on site itself serves as immutable evidence, enabling objective pass/fail judgments based on data. For clients, colorized inspection results are easier to understand than flat tables of numbers and are highly effective as explanatory materials. Presenting digital 3D as-built data allows more reliable inspections, and there are increasing cases where handovers proceed more smoothly than before. The acquired as-built point-cloud data can also be used directly as electronic deliverables (as-built deliverables) and become digital records useful for future maintenance. The Ministry of Land, Infrastructure, Transport and Tourism, as part of i-Construction, promotes ICT use in as-built management, and point-clouds measured to appropriate accuracy are being accepted as inspection deliverables.

Simple surveying with smartphones + GNSS: LRTK supports the field

The point-cloud measurement and 3D data utilization described so far are extremely useful, but some may worry, “Doesn’t implementation require expensive equipment and specialist skills?” Indeed, high-performance terrestrial 3D laser scanners once cost millions of yen and skilled operators were limited. However, solutions that allow anyone to capture high-precision point clouds via simple surveying using smartphones combined with GNSS have appeared. A representative example is the LRTK series. LRTK consists of a small, high-precision GNSS receiver that attaches to a smartphone and a dedicated app, providing centimeter-level positioning (half-inch positioning) without complicated setup—an easy-to-use surveying system. By utilizing the smartphone’s built-in LiDAR sensor or camera, you can obtain high-precision 3D point clouds with absolute coordinates simply by walking around the site. Without special training or large equipment, you can accurately digitize site conditions.

With such tools, acquired point clouds can be immediately compared and used against design drawings or existing survey coordinates. Modern simple surveying devices like LRTK meet i-Construction requirements of the Ministry of Land, Infrastructure, Transport and Tourism, and point-cloud data with sufficient accuracy can be used as as-built management deliverables. Because initial costs are lower than dedicated equipment, adoption is spreading among small to medium construction firms and local governments, making these tools a trump card that dramatically improves on-site productivity and surveying accuracy. For more details, please also refer to the LRTK official site: https://www.lrtk.lefixea.com/

FAQ

Q: Do I need expensive equipment or specialized knowledge to introduce 3D point-cloud surveying? A: In the past, expensive laser scanners and advanced CAD skills were required, but today there are inexpensive, easy-to-use solutions such as simple surveying devices using smartphones + GNSS (e.g., LRTK). Intuitive dedicated apps and cloud services are also well developed, making it easier to adopt even without specialist surveyors.

Q: Is the accuracy of 3D data obtained by point-cloud surveying comparable to traditional surveying? A: With appropriate methods, point-cloud data accuracy can match traditional surveying levels (on the order of several centimeters to several millimeters). By providing coordinates to control points with high-precision GNSS and generating point clouds with laser scanners or photogrammetry, you can achieve accuracy sufficient for as-built management in civil engineering. Because the entire site can be measured in high detail, local deviations are easier to detect, improving quality control accuracy.

Q: Aren’t point-cloud data files large and difficult to handle? A: Indeed, point-cloud data can contain tens of millions of points and reach several gigabytes in size. However, recent improvements in PC performance and cloud services make it possible to smoothly display and process large point clouds. Techniques such as thinning unnecessary points or converting to meshes to compress data are also available. Using dedicated viewers or web platforms, most users can handle 3D point clouds on typical PCs without problems.

Q: Are point-cloud-based as-built inspections formally accepted by clients and inspectors? A: Yes. The Ministry of Land, Infrastructure, Transport and Tourism is promoting the use of 3D survey data within i-Construction, and acceptance of point-cloud-based as-built management methods is progressing. There are increasing cases where point-cloud data have been submitted as deliverables and passed inspection. Of course, accuracy management and documentation in accordance with surveying standards are required, but if conditions are met, point-cloud inspection results can be treated as formal as-built inspection records. As a digital record that can replace traditional photos and drawings, it is increasingly being accepted by clients.