Visualizing the Comparison Results Between As‑Is Point Clouds and Design Data with a Heat Map

In surveying and as‑built management for civil engineering works, the method of comparing the as‑is point cloud (3D point cloud data acquired by scanning the site) with the design data and visualizing the results as a heat map has attracted particular attention in recent years. It allows an overall understanding of whether the finished terrain or structures match the design, greatly contributing to more efficient quality control. This article explains in detail the significance and benefits of visualizing the differences between as‑is point clouds and design data as a heat map, the effects of adoption, and easy ways to realize it.

Contents

• The importance of comparing as‑is point clouds and design data in as‑built management

• Advances in point cloud surveying technology and acquiring as‑is point clouds

• What is heat map analysis? (method of visualizing differences)

• Benefits of heat map visualization

• How point clouds and heat maps drive DX in as‑built management

• 3D as‑built management promoted by the Ministry of Land, Infrastructure, Transport and Tourism

• Simplified surveying with LRTK

• FAQ

The importance of comparing as‑is point clouds and design data in as‑built management

In civil engineering, “as‑built management” refers to the process of verifying and recording that completed structures and prepared land match the shapes and dimensions shown in the design drawings. Especially for public works, it is necessary to demonstrate with measurement data that the work meets the client’s prescribed standards, and as‑built management is a key element of quality assurance. A critical part of as‑built verification is comparing the as‑built (measured shape data) after construction with the design data to check for discrepancies.

However, conventional methods of comparison had several challenges. The main issues are as follows:

• Labor and time burden: As‑built inspections require significant personnel and time, and it was difficult to measure every corner of a large site. With a shortage of personnel, it became practically difficult to inspect every detail using conventional methods, so labor saving was demanded.

• Risk of missed defects and accuracy limitations: Point‑by‑point surveying can only capture part of the shape, risking overlooking intermediate irregularities or subtle dimensional errors. For example, specified survey points might meet standards while unevenness occurs between them and goes unnoticed. Many aspects still relied on the intuition of experienced workers, so the risk of overlooking as‑built defects was always present.

• Safety issues: Measuring is difficult in locations that are hard for people to access, such as high cut slopes, the undersides of bridges, or narrow tunnels. Forcing inspections in such places poses danger to workers, and traditionally those areas’ as‑built verification was often abandoned.

• Time-consuming documentation and sharing: Measurement results had to be recorded and drawn up manually, and site supervisors were occupied with creating photo ledgers and as‑built inspection documents. Human errors such as missed records or forgotten photos were unavoidable, and reporting and sharing consumed time.

As a new approach to solve these problems, as‑built management using as‑is point cloud data has attracted attention. Comparing as‑is point clouds with design data allows planar verification of conformity to the design and helps prevent overlooking defective areas.

Advances in point cloud surveying technology and acquiring as‑is point clouds

Recent point cloud scanning technology has made it easy to capture site geometry as countless points (point cloud data). A point cloud is a set of 3D coordinate data representing many points that compose the surface of the target, and scanning the site is like creating a complete digital copy of the terrain or structure. Whereas conventional surveying inferred shape by connecting distant points, point cloud surveying measures the entire object at high density, allowing accurate capture of complex curved surfaces and fine irregularities. It is also easy to compute distances, thicknesses, and volumes on the acquired point cloud, and use has expanded into as‑built management and various construction management tasks.

There are now diverse ways to acquire point cloud data. Typical methods include:

• 3D laser scanners: Using high‑performance terrestrial laser scanners enables acquisition of high‑density point clouds with millimeter accuracy. However, equipment acquisition costs tend to be high.

• Drone photogrammetry: A camera mounted on a drone photographs the site from above, and photogrammetry generates a point cloud. This covers wide areas in a short time, but ensuring accuracy requires a sufficient number of photos and control points for georeferencing.

• Smartphone LiDAR: Recent smartphones (e.g., LiDAR‑equipped iPhone and iPad) are making close‑range 3D scanning easy. You can obtain a point cloud by simply walking around the site holding up a smartphone, but standalone positioning accuracy is low, on the order of several meters (several ft), so georeferencing to control points after acquisition is essential for practical use. However, by combining with an external RTK‑GNSS receiver, cases have emerged where smartphones can achieve accuracy on the order of several centimeters (cm level accuracy (half‑inch accuracy)).

With these cutting‑edge technologies, the barrier to point cloud surveying has fallen substantially. Site personnel themselves can increasingly acquire detailed as‑is point clouds in a short time without outsourcing to specialists. The next chapter looks at “heat map analysis,” which compares such acquired point cloud data with design data.

What is heat map analysis? (method of visualizing differences)

Heat map analysis is a method of overlaying the acquired as‑is point cloud data with the design data and visualizing the differences by color. By comparing the measured as‑built shape (point cloud) with the design model and representing the error at each point as different colors, the quality of the as‑built work is shown intuitively. In other words, it visualizes as a colored 3D diagram the irregularities of the as‑built result that were hard to see in flat drawings or numerical lists.

Specifically, parts that are higher than the design (excess fill or protrusions) are typically shown in red or warm colors, and parts that are lower (over‑cut or lacking material) are shown in blue or cool colors. Areas that match the design are shown in green or neutral colors, making it immediately obvious which parts are well executed and which require correction. The color gradient also expresses the magnitude of the error, so you can easily grasp whether the overall tendency is positive error (too high), negative error (too low), or whether defects are concentrated in specific areas.

Benefits of heat map visualization

Introducing difference visualization with heat maps dramatically improves the accuracy and efficiency of as‑built inspection. Main benefits include:

• Detection of minute defects: Small irregularities on the order of a few centimeters can be easily detected by color changes on the heat map that were overlooked by numerical comparisons. Localized errors that point sampling could not catch are not missed.

• Understanding error trends: Because the amount of error is visualized by color gradients, you can see at a glance whether the finish tends to be generally high or low, or if only particular areas are insufficient.

• Intuitive pass/fail judgments: Results are presented as a color map, so even non‑specialists can visually understand the situation. It is easy to share as‑built status with site workers and supervisors, and the team can intuitively grasp where corrections are needed.

• Immediate feedback: If you display the heat map on a tablet or smartphone and inspect the site, you can judge on the spot how much to correct each location. Traditionally, workers had to consult drawings or numbers and search the site for defects, but with a heat map you can begin corrective work immediately by matching to the actual object. Using AR technology, it is also possible to overlay the visualization on the live camera view, turning inspection results into a real‑time quality improvement tool.

• Automatic conformity judgment: By setting design tolerances, you can automatically extract defective areas that exceed those ranges on the heat map. For example, you can color only points with errors of ±5 cm (±2.0 in) or more, enabling efficient checks that follow pass/fail criteria.

How point clouds and heat maps drive DX in as‑built management

Introducing point cloud surveying and heat map analysis brings major changes to the entire as‑built management workflow. The main effects of using digital technologies are summarized below:

• Reduced labor and faster surveying: Because wide areas can be scanned at once, as‑built surveying that used to take days can be drastically shortened. Required personnel can be minimized, and dangerous measurements at height or scaffolding installation are often unnecessary, improving safety (there are cases where surveying that used to take two days was completed in half a day).

• Advanced as‑built inspection: With point clouds and heat maps, as‑built quality can be evaluated objectively and quantitatively. Subjective checks based on intuition and experience are reduced, enabling corrective instructions rooted in numerical evidence. Scanning during construction and verifying immediately prevents rework and maintains quality.

• Information sharing and remote supervision: Point cloud data and heat map results can be uploaded to the cloud and shared with stakeholders. Clients and supervisors can view 3D as‑built data from the office via the internet and give instructions remotely. For example, you can measure a dimension on a point cloud acquired by a smartphone and instantly share the result with the client. Reports can also be auto‑generated with one click, smoothing information flow between the site and the office.

• Data accumulation and reuse: Keeping an objective 3D record as point clouds makes it useful for future displacement measurements and root‑cause investigations. Comparing past point clouds makes monitoring subsidence or long‑term deformation of structures easy. As‑built data can be integrated into BIM/CIM models for maintenance planning and future construction planning, making it a valuable digital asset after project completion.

Thus, using as‑is point cloud data and heat map visualization strongly promotes DX (digital transformation) at the site. Digitizing the steps of measuring, verifying, and reporting improves accuracy and efficiency, activates communication through visualization, and ultimately leads to safer, higher‑quality construction.

3D as‑built management promoted by the Ministry of Land, Infrastructure, Transport and Tourism

The government is also backing these advanced as‑built management methods. As part of construction DX initiatives such as i‑Construction, the Ministry of Land, Infrastructure, Transport and Tourism has been developing standards for as‑built management that utilize ICT. In recent years, not only traditional cross‑section and point measurements but also management that measures entire surfaces and as‑built evaluation using 3D data have been officially incorporated into various work procedures.

For example, in earthworks it has become mandatory to measure and evaluate the finished embankment across the entire slope (“surface as‑built management”), and scanning with laser scanners or photogrammetry has been newly accepted for tunnel work and bridge foundations. The March 2025 (Reiwa 7) revision of supervisory manuals and standards explicitly describes as‑built inspection methods using point cloud scans. Going forward, 3D as‑built management that compares as‑is point clouds with design data to evaluate differences will become an indispensable industry standard.

Simplified surveying with LRTK

Hearing “as‑built management using the latest technology” may make you think that expensive 3D scanners and specialist skills are required. However, as technology democratizes, centimeter‑level high‑accuracy surveying is now possible simply by combining a smartphone with a small GNSS receiver. A representative example is the new surveying system “LRTK” offered by Reflexia Co.

LRTK consists of a small RTK‑GNSS device that attaches to a smartphone, a dedicated app, and a cloud service. By scanning with the phone’s built‑in LiDAR while correcting positions with RTK, anyone can obtain high‑accuracy as‑is point clouds with one touch. Acquired data are automatically recorded in a known coordinate system, eliminating complicated post‑processing. Upload point clouds to the LRTK cloud and you can immediately perform distance, area, and volume difference measurements and heat map comparison analysis in a browser. Without installing dedicated software on a PC, you can review and share as‑built data from the field using just a tablet. It also has a feature to display acquired heat maps in AR on site, enabling precise on‑the‑spot repair instructions.

Using LRTK makes 3D surveying and as‑built inspection—tasks that traditionally required specialists or costly equipment—surprisingly simple. It can greatly shorten work time and is expected to dramatically improve productivity. LRTK also supports the Ministry of Land, Infrastructure, Transport and Tourism’s i‑Construction initiatives, strongly supporting site DX. For more details, please check the LRTK official site. Why not take your site to the next stage with LRTK?

FAQ

Q: What is an as‑is point cloud? A: An as‑is point cloud is 3D measurement data of the site (a collection of points) acquired by laser scanners, photogrammetry, etc. It is digital data that represents the surfaces of structures and terrain with countless points—like a 3D copy of the actual site.

Q: What equipment or software is needed for heat map analysis? A: First, you need measurement equipment to acquire as‑is point clouds. High‑accuracy measurement can use 3D laser scanners, RTK‑equipped drones, or smartphone systems with RTK‑GNSS. The comparison of the acquired point clouds with design data can be performed using point cloud processing software or cloud services. For example, using LRTK Cloud you can upload point clouds and automatically generate a design‑vs‑as‑built difference heat map in your browser.

Q: Can 3D point cloud data be handled without specialist knowledge? A: Yes. Recent solutions are designed for intuitive use. Services that scan with a smartphone app and automatically process the data (e.g., LRTK) allow non‑specialists to obtain high‑accuracy point clouds with button operations and then automatically receive visualizations such as heat maps. The results are displayed as color maps, making them easy for anyone to understand.

Q: What level of comparison accuracy can be secured between point cloud data and design data? A: It mainly depends on the measurement accuracy of the point cloud and the accuracy of aligning the two data sets. Using high‑accuracy equipment or RTK, point cloud accuracy can be kept within a few centimeters (a few cm (a few in)). By aligning the acquired point cloud to the design coordinate system, the errors shown on the heat map fall roughly within that range. Some software also offers functions to automatically fine‑tune vertical reference levels for comparison, supporting rigorous accuracy verification.

Q: On what kinds of sites is heat map‑based comparison visualization effective? A: It is widely effective where shape and dimensional quality control are important: earthworks managing cut and fill volumes, erosion control and road works checking slope as‑built, paving works verifying surface flatness and thickness, and structural works checking tunnel or dam shapes. Heat maps can be used from construction progress management to final as‑built inspection and long‑term displacement monitoring. There are also maintenance applications, such as detecting bridge deflection over time or deformation of concrete structures.

Q: Aren’t large point cloud file sizes hard to handle? A: 3D point clouds can contain millions of points and become large files, but recent software and cloud services are designed to handle large data efficiently. Functions such as thinning out unnecessary regions or focusing on the area of interest make it manageable even with limited PC performance. Some services render in the cloud and let you view in a browser, so 3D data can be reviewed on a field tablet without heavy load.

Q: Is there a cost‑benefit to introducing new 3D surveying technology? A: Dedicated 3D laser scanners can be very expensive (on the order of several million yen), but methods using smartphones and small GNSS devices can be introduced at relatively low cost. Early correction of defects that would have been missed with conventional methods, reduced rework, and labor savings can provide benefits that justify the investment. In short, modern point cloud and heat map technologies can deliver productivity improvements and quality assurance benefits that outweigh initial costs.