How to Accelerate On-site DX by Introducing Point Cloud Scanning

Table of Contents

• Introduction: The need for DX at construction sites and the turning point for surveying work

• What is point cloud scanning? Overview and technical background

• Labor savings and quality improvements in surveying enabled by point cloud scanning

• Steps to introduce point clouds: planning, equipment selection, measurement, processing, and utilization

• Concrete effects at sites and implementation examples

• Common challenges and points to overcome them

• Storage, utilization, and future prospects of point cloud data

• Introduction to LRTK: an integrated smartphone simple point cloud scanning solution (price not stated)

• FAQ: Questions from the field about introducing point cloud scanning

Introduction: The need for DX at construction sites and the turning point for surveying work

In recent years, the construction industry has been emphasizing the importance of on-site DX (digitalization of construction sites). With severe labor shortages, the need to correct long working hours (the so-called “2024 problem”), and the urgency to reverse productivity declines, on-site process reform through DX has become imperative. One of the technologies attracting attention as a trump card is point cloud scanning.

Traditional surveying work involved surveyors measuring individual points one by one with a transit or GPS surveying instrument and drafting drawings. This required experience and effort, and large sites demanded significant time and personnel. Today, however, it is possible to create comprehensive 3D data of a site using laser scanners, drones, or even smartphones. Surveying methods have dramatically evolved, and we have entered an era where the actual conditions of a site can be digitally recorded as-is.

Since 2016, the Ministry of Land, Infrastructure, Transport and Tourism has promoted the i-Construction initiative, aiming to improve construction site productivity by 20% by 2025. Use of 3D data and automation of on-site tasks are viewed as key, and the use of point cloud data for as-built management and other purposes has been stipulated in various guidelines. In other words, realizing a digital twin through point cloud data is expected to be a pillar of construction DX. Indeed, if completed roads and structures are recorded as point clouds, the data can be useful for post-completion drawing creation, quality inspections, and future renovation planning.

Against this backdrop, surveying work is now at a major turning point. The new method of scanning an entire site and digitizing it is having an impact that overturns prior conventions. But many may still wonder, “What exactly is a point cloud? Can it be used on our site?” This article clearly explains—from the basics of point cloud scanning to its benefits, concrete procedures and case studies, and even the latest solution, the smartphone-integrated point cloud scanner “LRTK”—for those involved in surveying. Use this as a hint for taking the first step toward on-site DX.

What is point cloud scanning? Overview and technical background

Point cloud data is data that represents the shape of objects or terrain in three-dimensional space using countless points. Each point includes X, Y, and Z coordinate values indicating position, and when acquired by a camera may also include color information (RGB values). For example, when a building or terrain is converted into a point cloud, millions of measurement points on its surface are reproduced on a computer and displayed as a three-dimensional “collection of points.” In short, a point cloud is the raw 3D model data that copies real space wholesale.

The method of obtaining this point cloud data is called “point cloud scanning.” Representative methods include measurement by laser scanner (LiDAR) and photogrammetry. Laser scanners emit laser light at high speed and generate a point cloud by calculating distances from the return time of the light. Photogrammetry, on the other hand, reconstructs a three-dimensional point cloud by analyzing multiple photographs taken from different angles with software. Recently, methods have diversified beyond large ground-based 3D laser scanners to include drone-mounted LiDAR, mobile mapping systems mounted on vehicles that measure while driving, and handheld scanners. In addition, point clouds can now be easily obtained using built-in LiDAR or cameras in smartphones and tablets. In other words, point cloud measurement that previously required specialized equipment is becoming feasible with familiar devices thanks to technological advances.

Labor savings and quality improvements in surveying enabled by point cloud scanning

• Significant efficiency gains and labor reduction in surveying work: By utilizing point cloud scanning, surveying that previously required several people several days can be completed quickly by a single person. Because a large number of points can be acquired at once even on vast sites, measurement omissions are reduced and the need for re-measurement later is minimized. For example, at a site that introduced tablet-based point cloud measurement, as-built surveying that previously took more than half a day was completed in about 30 minutes, dramatically shortening work time. In the face of severe labor shortages, this leads to labor savings that allow small teams to manage sites.

• Improved measurement accuracy and data quality: Point cloud data provides extremely dense and high-precision measurements. The shape of an object can be recorded down to the millimeter level, and it is easy to remeasure dimensions at any location after acquisition. As a result, as-built conditions, which were previously inferred from a subset of survey points, can be accurately captured, enabling quality control that minimizes discrepancies with design drawings. Once acquired, point cloud data is stored as a digital record, allowing you to review the point cloud later to understand “what was that part of the site like?” during inspections or maintenance. It becomes an asset data set that is useful for inspections and maintenance.

• Improved safety for hazardous work: Laser measurement and photogrammetry using drones or smartphones are non-contact, enabling safe surveying in areas where personnel cannot enter or are dangerous. Measurements of high structures or unstable slopes can be performed remotely, reducing the risk to workers. The need to set up scaffolding or restrict access is reduced, greatly lowering the risk of accidents during surveying.

• Smoother consensus building through 3D data sharing: Acquired point cloud data can be shared among stakeholders as a three-dimensional model. Uploading to the cloud allows supervisors or clients in remote offices to view the site in 3D on their computers. Information that was difficult to convey with drawings or photos becomes immediately clear with 3D point cloud visuals. Showing a point cloud on a tablet during an as-built inspection increases persuasiveness, and discussing design changes while viewing lifelike data makes explanations and approval processes smoother.

• Advanced construction planning and simulation: Point cloud data is powerful for construction planning and pre-simulation. For example, you can plan heavy equipment ingress routes on the point cloud model to check for obstacles in advance, or overlay design data (BIM/CIM models) on the point cloud to check for clashes. In some cases, temporary structures or expected-completion models can be placed in the point cloud and viewed on site with AR technology. These uses improve planning accuracy and help prevent rework and construction errors.

Steps to introduce point clouds: planning, equipment selection, measurement, processing, and utilization

• Planning: First, clarify why you are introducing point cloud scanning. Identify which on-site tasks have issues and how point cloud technology can improve them—for example, “as-built measurement takes too long” or “we want to speed up earthwork volume calculations.” It’s useful to research similar cases to learn from successes and failures. For initial introduction, start small with a trial on a limited site (a small-start approach) so your organization can experience the benefits. Also prepare rough cost estimates, schedules, and an in-house training plan to gain the understanding and cooperation of management and site staff.

• Equipment selection: Next, choose measurement equipment and methods that match your objectives. The optimal method varies with the scale and shape of the target, required accuracy, and site environment. For wide-area terrain surveys, drone + photogrammetry is efficient; for precise dimensions of structures, a static laser scanner is appropriate. If ease of use is a priority, smartphone or tablet scanning is a candidate. Consider sensor performance on metallic or low-light surfaces. Compare each device’s measurement range, accuracy, portability, and software requirements, and select within your budget. If necessary, demo or rent equipment to confirm fit with your operations.

• Measurement: Once equipment is chosen, acquire point cloud data on site. Prepare in advance by placing control points and checking weather conditions. For laser scanning, stabilize the instrument and place alignment targets if scanning from multiple stations. For drone photogrammetry, plan flight paths and altitudes to ensure sufficient ground resolution. For smartphone or tablet measurement, tidy the surroundings to make movement easier and check battery levels to ensure uninterrupted continuous shooting. During measurement, check in real time for blind spots or omissions and add scans from different angles or positions as needed. Consider safety by restricting access during high-elevation work or flights, and conduct activities within reasonable limits.

• Processing: Process and analyze acquired point cloud data with dedicated software or cloud services. When multiple point clouds are taken, perform registration to align them into a single coordinate system. Filter out noise or perform thinning if measurement errors exist. High-resolution point clouds can become massive, so high-performance PCs or GPU environments help processing run smoothly. Some software offers cloud-based automated processing. After processing, perform tasks such as comparing with design data, creating CAD drawings, generating contours or cross-sections, and verifying that coordinate systems and scales are correct and consistent with control points. Back up and securely store the completed point cloud data and deliverables.

• Utilization: Use processed point cloud data in actual operations. For example, in as-built management you can measure dimensions on the point cloud and create quality reports; for earthworks, compare point clouds before and after excavation to accurately calculate volumes. Overlay design drawings or BIM/CIM models to detect discrepancies and perform 3D as-built inspections. Sharing data with site agents or clients enables remote site verification and aids consensus building. The range of point cloud applications is broad—by starting with one use case and verifying effectiveness, you can gradually expand utilization and drive various on-site DX initiatives.

Concrete effects at sites and implementation examples

Here are two real-site cases where point cloud technology was introduced and achieved significant results. Both succeeded without large-scale investment by using practical, site-oriented approaches.

Case 1: 30-minute survey and same-day volume calculation with smartphone LiDAR – On a small earthwork project in Gifu Prefecture (excavation area approximately 150 m²), as-built measurement had previously been performed by UAV (drone) photogrammetry, but issues included “too much time required for preparation and processing” and “quantities could not be calculated until the point cloud was ready, forcing temporary pauses in construction.” The team trialed a 3D scanning app compatible with the iPad Pro’s built-in LiDAR and conducted tablet-based point cloud measurement. As mentioned earlier, they completed point cloud acquisition in 30 minutes—15 minutes to set control points and 15 minutes to scan—and calculated earthwork volume on-site to report to the client immediately. Eliminating processing wait time allowed arrangements for subsequent steps (ordering backfill materials and disposing of surplus soil) within the same day, significantly shortening the schedule and improving logistics. Post-process accuracy verification showed a volume difference from drone survey of only within 0.1%, demonstrating high accuracy and earning client approval. The keys to success were choosing the tablet LiDAR to meet the specific goal of rapid as-built measurement and immediate quantity calculation, and conducting site-led trials to achieve short measurement times. This example shows that site DX can be realized by creatively using familiar devices without large capital investment.

Case 2: Immediate 3D recording and sharing of disaster sites with LRTK – Following the 2023 earthquake in the Noto region of Ishikawa Prefecture, rapid assessment and recording of damaged sites became urgent. Large surveying instruments could not be brought in immediately, and communication infrastructure was partially disrupted. The smartphone fitted with LRTK proved invaluable. A smartphone-based measuring device weighing only a few hundred grams is easy for workers to carry anywhere. LRTK uses augmentation signals from Japan’s quasi-zenith satellite system (Michibiki), allowing positioning accuracy on the order of a few centimeters (a few inches) even in areas without cellular reception; thus, high-precision point clouds with positional coordinates were obtained in radio-dark disaster zones. Investigators walked the site holding the smartphone and performed continuous positioning and scanning, uploading acquired data to the LRTK cloud for immediate sharing with offices and support headquarters. This greatly accelerated damage assessment and restoration planning. Stakeholders commented that “in disasters, even a single small LRTK unit can dramatically improve situational sharing,” highlighting its value as a new emergency tool. The strengths of this case are portability for everyday carry and instant data-sharing capability. LRTK demonstrated value not only for routine construction management but also for quick initial surveys after disasters, enabling anyone to perform high-precision 3D measurement easily.

Common challenges and points to overcome them

• Concerns about introduction cost: Many worry that the latest 3D scanners are prohibitively expensive. While high-performance laser scanners can indeed cost millions of yen, low-cost solutions have become more available. Options include outsourcing drone surveys, renting equipment, or starting with simple smartphone scanners and validating effectiveness before full-scale introduction. Some companies use subsidies or leasing. Prove effects with a small investment first, then present a cost-recovery outlook to gain internal buy-in.

• Difficulty of operation and management: Concerns like “Is specialized knowledge required?” or “Will ICT fit on-site?” are common. Although new technologies can initially be confusing, modern point cloud systems have become user-friendly. Some smartphone apps allow scanning as simply as taking photos, and cloud services can automatically stitch point clouds. Begin with trial-and-error, train tech-savvy younger staff to build proficiency, and leverage vendor support and workshops. Embracing new technology can create opportunities for younger employees and strengthen the company’s digital momentum.

• Data size and PC environment issues: Worries about large point cloud files are frequent. Indeed, point clouds of hundreds of millions of points can be several gigabytes in size, and high-performance PCs are desirable for processing. However, cloud storage and compression technologies have reduced barriers to storing and sharing large datasets. Using point-cloud-specific compressed formats such as LAZ can greatly reduce file size, and online platforms for viewing and sharing point clouds have emerged. Upgrading internal PCs with better GPUs and more memory can also help. By gradually enhancing equipment and selectively scanning high-density areas, data-size issues can be effectively managed.

• Accuracy and reliability of results: Some question whether point cloud measurements are precise enough or acceptable as official survey results. Many validation cases show that high accuracy can be achieved if conditions are properly controlled. Even simple measurements with smartphones or drones can reach centimeter-level accuracy when combined with ground control points, and there are reports of millimeter-level results in small-scale surveys. The Ministry of Land, Infrastructure, Transport and Tourism also recommends using point cloud data, and when surveys follow guidelines they can be used for as-built management deliverables. Start with trial sections comparing conventional methods to verify suitability, and share findings internally and externally. With proper procedures, point clouds are a reliable technology that is becoming an industry standard.

Storage, utilization, and future prospects of point cloud data

Point cloud data is a valuable digital record that preserves a site in its entirety and serves as the data foundation for construction DX. To realize its true value, appropriate storage and sharing are essential. Even precisely acquired point clouds become worthless if they are lost, corrupted, or cannot be retrieved when needed. Common site problems include “overwriting and losing past point clouds,” “not knowing which reference coordinate system data uses,” “file sizes overwhelming PC or tablet storage,” and “files too large to send by email.” To prevent such issues, adopt smart point cloud management practices.

First, establish file-naming and version-control rules to avoid accidentally overwriting historical data. Record surveying control points and coordinate system information as metadata so the spatial context is clear later. Choose storage formats thoughtfully: standard LAS, compressed LAZ, or vendor-specific formats exist, but selecting a versatile format improves future compatibility. For capacity, prepare high-performance external HDDs or internal servers and perform regular backups. Cloud storage facilitates smooth external data sharing. When exchanging huge files is difficult, split files by region or prepare lightweight, lower-density versions alongside the full dataset. Properly stored and shared point cloud data becomes a legacy that transfers site knowledge and continuously supports DX benefits.

Finally, future prospects: Point cloud technology is expected to penetrate the construction industry further and broaden its range of applications. The Ministry of Land, Infrastructure, Transport and Tourism is promoting the principal application of BIM/CIM to direct-managed projects, and building digital twins that combine 3D design models and construction-time point cloud data is becoming mainstream. Examples of automatic as-built inspection by overlaying point clouds and 3D design data and using point clouds in construction simulations have already been demonstrated on sites. In the future, sequential point cloud accumulation of on-site events and AI-based automatic detection of changes or deterioration by comparing past data will enable smart maintenance. Research on machine learning to recognize and classify structures and features from point clouds is underway. Measurement devices are also evolving, making it likely that everyone will soon be able to 3D scan with mobile devices. Beyond construction sites, point clouds will become commonplace in infrastructure inspection, disaster prevention, urban planning, and environmental fields. Point cloud data will increasingly stand out as a key technology supporting the future of the construction industry.

Introduction to LRTK: an integrated smartphone simple point cloud scanning solution (price not stated)

Finally, for those who want to start point cloud scanning easily, we introduce the low-cost, smartphone-integrated 3D measurement solution “LRTK.” LRTK consists of a compact high-precision GNSS receiver that attaches to a smartphone called the LRTK Phone, a dedicated surveying app, and a cloud service. Combined with a smartphone’s built-in LiDAR sensor, it enables the simple capture of centimeter-level accuracy (half-inch accuracy) point cloud data. Traditionally, smartphone LiDAR scans produced point clouds in arbitrary local coordinates that required post-processing alignment to control points. With LRTK, accurate coordinates (latitude, longitude, and elevation) in the world geodetic system (WGS84) can be attached during scanning, eliminating cumbersome alignment steps. For example, if a known control point exists at the site, scanning the surrounding terrain and structures from nearby with a smartphone provides a 3D point cloud model with global coordinates in real time.

LRTK dramatically simplifies point cloud measurement that previously required specialized equipment. Point clouds are displayed on the screen in real time as you move the smartphone, allowing immediate verification of omissions. After acquisition, you can measure distances and areas or calculate volumes on the smartphone, so scanning an embankment can yield on-the-spot volume calculations. Tasks that used to require manual calculations from drawings or volume tables can be completed on site in minutes. Captured images and point cloud data are automatically synchronized to the cloud, making detailed office analysis and stakeholder sharing easy. Positioning accuracy is on the order of a few centimeters (a few inches), and because it supports augmentation signals from Japan’s quasi-zenith satellite system, accuracy is maintained even at sites without communication coverage. With these features, LRTK is a smartphone-based solution that anyone can use to dramatically improve on-site survey accuracy and work efficiency. It is compatible with the Ministry of Land, Infrastructure, Transport and Tourism’s i-Construction standards and represents an excellent choice for promoting DX in construction.

FAQ: Questions from the field about introducing point cloud scanning

Q1. Aren’t point cloud scanners expensive and difficult to introduce? A1. It used to be assumed that you had to buy an expensive 3D laser scanner to perform point cloud measurement, but today there are more options such as outsourced drone survey services, equipment rental, and low-cost smartphone-based products. Start with inexpensive methods on a small scale to test effects, and then invest incrementally. Subsidies can sometimes be used. You don’t have to buy a full set of the latest equipment at once—creativity and phased introduction make implementation feasible.

Q2. Are point clouds obtained by smartphones or drones really accurate enough? A2. When measured appropriately, point clouds from smartphones or drones can provide accuracy sufficient for practical use. For example, smartphone measurements aligned with known control points have confirmed centimeter-level accuracy, and millimeter-level results have been reported in small-scale surveys. For cases requiring higher precision, combine ground-based laser scanners or precision GNSS. Also, validate acquired point clouds against conventional survey results to confirm they meet required accuracy.

Q3. Processing and operation of point cloud data seems difficult. Can ordinary sites handle it? A3. Recent point cloud tools have improved usability and are easier to handle even without specialized technicians. Smartphone-app-based systems allow intuitive scanning like taking photos, and cloud services can automatically process and share point clouds. You may be confused at first, but hands-on training and vendor support will help. Letting younger staff lead initial operations and adopting a “try-first” attitude will steadily raise in-house skills.

Q4. Point cloud files are large—how should we store and share them? A4. Higher-resolution point clouds result in larger files, but there are countermeasures. Use compressed formats such as LAZ to significantly reduce size, and trim unnecessary regions. Internally, maintain large external storage or NAS and implement backup routines. For sharing, use cloud storage or point cloud viewers to facilitate smooth transfer of large files. Choosing appropriate formats and media and applying simple measures makes data-size issues manageable.

Q5. Can data obtained by point cloud scanning be used for official drawings or inspection documents? A5. Yes. When point cloud as-built measurements are acquired according to the Ministry of Land, Infrastructure, Transport and Tourism’s guidelines, they can be used as inspection materials and deliverables. The use of point clouds for as-built management and quantity calculation is recommended in several standards, and many projects have adopted them. It’s important to cross-check with traditional measurements as needed, but point cloud data is becoming an established form of electronic deliverable. Clients are increasingly receptive to point cloud use, so you can proceed with confidence.