Use Cases of Point Cloud Data

​

Point cloud data has a very broad range of applications in the civil engineering and construction fields, offering benefits that were not achievable with traditional drawings or surveying data. Below are some of the main use cases.

​

• Design Drawing Creation (Mapping Existing Structures):
  By converting existing structures and terrain on-site into point cloud data, detailed 3D models or 2D drawings can be created. For example, in the case of renovation designs for old bridges or tunnels, the current dimensions may be unclear, or no drawings may be available. In such cases, capturing point clouds with laser scanning allows for the creation of accurate 3D models or CAD drawings of the plan view, cross-sections, etc. This method provides more accurate drawings than manual measurements or individual surveying, and enables the detailed understanding of complex structures. This improves design accuracy and allows for planning based on the actual conditions when working on renovation or expansion projects.

• As-Built Management (Verification of Shape and Dimensions After Construction):
  Point cloud data is used to measure structures or created land after construction to verify if they were built according to the design. Measurements such as dimensions, slopes, and flatness of the components can be checked, and differences from the design model can be identified. For example, by capturing the thickness of tunnel linings after concrete placement or the shaping of roadbeds, point cloud data can identify even small undulations that are difficult to capture by hand. In one case, the use of point cloud data for confirming rebar shape resulted in a 73% reduction in both time and cost compared to traditional methods. Additionally, creating a reference plane from point cloud data and displaying the displacement in color is used for detecting anomalies like bulging, peeling, or leaks on concrete surfaces. This application in as-built management helps with early detection of construction errors and reduces rework, thus ensuring quality and improving efficiency.

• Maintenance and Monitoring (Infrastructure Inspection):
  Point cloud data is also useful for the maintenance and monitoring of infrastructure assets like bridges, tunnels, and roads. By regularly 3D scanning these structures, changes over time can be digitally accumulated and compared. For example, in bridge inspections, combining point cloud data with high-resolution images can help detect concrete cracks or displacement of components, or assess the changes in tunnel cross-sections over time. In an actual highway tunnel project, a system was implemented where daily scans were done using a smartphone-mounted LiDAR, and the data was shared via the cloud. This reduced the need for site visits from headquarters while cutting the time to incorporate changes into the construction plan by 90%. By sharing point cloud data via the cloud, the site’s condition can be remotely understood in 3D, and all stakeholders can verify progress and problem areas. Additionally, by comparing the acquired point cloud data with past design models, signs of deformation can be detected early, which can be useful for maintenance planning. The use of point cloud data in maintenance is gaining attention as it provides long-term digital records of infrastructure while streamlining and enhancing inspection tasks.

• Volume Calculation (Excavation and Embankment Volumes):
  Point cloud data is effective in determining excavation and embankment volumes in earthworks. Traditionally, surveyors would perform polygonal surveys and create cross-section drawings from a few dozen elevation points to calculate volumes. However, by generating a mesh model of the terrain from high-density point clouds obtained through drone photogrammetry or laser scanning, the volumes of embankments or deposited soil can be calculated quickly and accurately. Point cloud-based volume calculation scans the terrain in detail, minimizing oversight, and volumes can be automatically computed for any selected range using software. In one construction case, using point cloud data to calculate excavation volumes showed no issues with accuracy compared to traditional methods. Additionally, for dangerous slopes or large embankments, drones can safely measure the area without personnel needing to enter hazardous zones. Point cloud-based volume calculations not only improve the efficiency of as-built management but are also being used in disaster prevention to estimate runoff volumes during landslides.

• Infrastructure Inspection and Digital Twin:
  Point cloud data is also gaining attention in infrastructure inspection and the context of BIM/CIM. For example, in railways and highways, 3D scans of tunnel interiors can be used for displacement measurement, and point cloud models of bridges can be linked with images of cracks and peeling to document areas of deterioration. Additionally, laser point clouds obtained at a city scale are being used to create 3D models of roads and buildings for use in disaster simulations and maintenance. Point cloud data, being a digital copy of the real world, allows for consistent data use from design to construction and maintenance. For instance, in a large underground station construction project by Shimizu Corporation, BIM models, on-site point clouds, and 360-degree photos were integrated into the cloud, enabling "remote construction management without site visits." In this way, point cloud data is connecting surveying, design, and inspection tasks digitally, making it a foundational technology supporting civil DX (digital transformation).

​

Benefits of Using Point Cloud Data

​

By utilizing point cloud data, various benefits are brought to civil engineering sites. Below are the main advantages:
 

• Improved Surveying and Construction Efficiency and Cost Reduction:
  Point cloud measurement allows for the acquisition of a wide range of information in a single pass, directly contributing to improved work efficiency and labor cost reduction. For example, in a case where point cloud data was used for equipment interference checks on a construction site, design work hours and the schedule were reduced by about 80%. In the same case, the accuracy of volume calculations for excavation was also confirmed without issues. This means significant labor-saving while maintaining quality. Traditional surveying and measurements, which could take days with multiple personnel, can now be completed in a short time, enabling quick design and construction decisions using the data. The total cost benefits—through reduced site surveying frequency, minimizing rework, shortening project timelines, and saving on heavy machinery and labor costs—are considerable. In an industry where adhering to timelines and managing budgets is crucial, point cloud data usage becomes a powerful tool for efficiency.

• Enhanced Safety (Unmanned Surveying of Hazardous Areas):
  Point cloud data enables surveying in areas that are difficult or dangerous for humans to access. Using drones or long-range laser scanners, data can be collected from hazardous locations like steep slopes prone to landslides, high-traffic road areas, or the upper parts of aging structures. For example, UAVs can scan from the air, eliminating the need for workers to climb to high or steep areas, thus reducing the risk of falls to zero. Ground-based laser scanners can also scan from a safe distance, allowing for the measurement of high areas such as bridge piers or tunnel ceilings. In this way, point cloud technology contributes to unmanned construction and remote sensing, reducing the risk of labor accidents. The shift from manual measurements in hazardous conditions to safe data acquisition via machines is becoming more prominent.

• Improved Design Accuracy and Reduced Construction Errors:
  By using point cloud data, precise designs based on actual site measurements can be created, reducing inconsistencies and errors during construction. For example, when renovating old structures, a 3D model generated from point cloud data can be used to check the installation positions of new components, allowing interference or dimensional errors to be detected in advance. In one case, by checking for interference using point cloud data and making design corrections, costly rework was avoided. Additionally, verifying the as-built shape immediately after construction with point cloud data allows for on-the-spot corrections, preventing issues from escalating into major defects. Using point cloud data, which faithfully represents the actual site, reduces the reliance on subjective adjustments like "the position doesn't match what we thought" or "adjusting on-site," thereby improving the quality and accuracy of construction. As a result, it leads to cost savings by reducing rework and quality defects.

• Long-Term Data Storage and Reuse (Asset Management):
  Once acquired, point cloud data can be stored as a digital record of the site for long-term use. This data can be reused in the future for planning or maintenance. For example, by scanning an entire structure at the time of completion, there is no need to take new measurements for repairs years later—simply retrieving the original point cloud data and converting it into a 3D model will provide the current site plan. Even if the drawings are lost or scattered, the point cloud data ensures peace of mind. In large-scale infrastructure projects, accumulating time-series point cloud data during construction, completion, and regular inspections allows for quantitative tracking of deformations and deterioration. This is useful for improving future maintenance plans and assessing damages in the event of a disaster (by comparing pre- and post-disaster terrain). Unlike paper drawings, point cloud data does not degrade or deteriorate, making it a digital asset that can be preserved indefinitely if managed well. Recently, cloud services for storing and sharing large amounts of point cloud data have become more common, providing an environment where necessary people can access and utilize past 3D data whenever needed.

​​

As mentioned above, the use of point cloud data demonstrates its value across a wide range of processes, from surveying to design, construction, and maintenance. In particular, 3D modeling using point cloud data is considered a key component of construction DX, which is being promoted by the Ministry of Land, Infrastructure, Transport and Tourism, such as in the i-Construction initiative. However, to fully maximize these benefits, it is important to implement the technology in a way that fits the specific needs of the site, while addressing the challenges outlined below.

​

Challenges and Solutions in the Use of Point Cloud Data

​

While the use of point cloud data offers many benefits, several challenges have been identified when implementing it on-site. Below, we will explain the main challenges and their solutions.

​

• Processing and Management Burden Due to Large Data Volume:
  Point cloud data is extremely dense, resulting in very large file sizes. Even for smaller sites, file sizes can reach hundreds of MBs, and for larger areas, it is not uncommon to reach tens of GBs. As a result, high-performance PCs for data processing and large-capacity storage are necessary. Additionally, point cloud data often requires specialized software to view and process, which can be a barrier for sharing and viewing between stakeholders. For instance, CAD software cannot directly handle point clouds, requiring conversion to intermediate formats. Furthermore, raw point clouds often contain irrelevant points (noise) such as people and vehicles, which require additional preprocessing to remove and organize. To address these challenges, the use of cloud services has increased in recent years. By uploading point clouds to internet platforms, automated processing services have emerged that allow for easy display and measurement directly via browsers. Additionally, efforts to use international standard point cloud formats (such as LAS or E57) make it easier to share data between different software systems. On the data management side, establishing naming conventions and version control for point cloud data by project, and sharing only the clipped data within the necessary range, can make the data easier to handle. Furthermore, downsampling (reducing the number of points) and utilizing compression formats can help reduce the size of the data. Recently, AI-based technologies for automatically detecting and removing unwanted objects (such as cars and people) from point clouds are also emerging, which will further reduce the processing and management burden in the future.

• Improvements for Higher Accuracy (Measurement Errors and Alignment):
  Although point cloud measurements offer high accuracy, if not carefully managed, errors can accumulate. For example, in ground-based laser scanning, multiple scans from different locations are used to cover blind spots and align point clouds (registration), and small misalignments may occur during this process. In drone photogrammetry, the absolute accuracy can vary by several centimeters to tens of centimeters, depending on GPS accuracy and flight altitude. To address positioning accuracy issues, known reference points or target markers are typically set up on-site for measurement and used to correct the point clouds. Alternatively, combining the system with centimeter-level positioning tools such as RTK-GNSS, which assigns absolute coordinates during data acquisition, is also effective (this is where the LRTK technology, which will be introduced in the next chapter, offers a major solution). Additionally, removing noise points is essential for improving accuracy. Unwanted points caused by people, machinery movements, or laser reflections from glass surfaces can be filtered out in software, thereby enhancing the model's accuracy. Careful planning for measurements is also important; for example, when measuring inside a bridge, scanning from both the top and bottom of the arch, or scanning from both the entrance and inside of a tunnel, helps minimize blind spots. Furthermore, SLAM-based handheld scanners use techniques to suppress error propagation by ensuring that the scanning path closes. Regular calibration of equipment and attention to errors caused by temperature and humidity are also necessary. By following these practices, point cloud data can be used to achieve surveying-grade accuracy and reliable results, making it suitable for design and construction tasks.

• Software Integration (Data Utilization Barriers):
  Point cloud data is not in a format that is universally accessible like CAD or Excel files, so without in-house expertise, it can easily become an underused resource. For example, if you want to create drawings from point cloud data, most regular CAD software cannot handle it directly, and specialized point cloud processing software is required to convert the data into meshes or surfaces. This conversion and editing require specialized knowledge, creating a barrier to learning and implementation. Software compatibility is also an issue; if you try to open point cloud data edited in one software with another system, differences in supported formats can cause delays. A solution is to first select the appropriate software for the intended use. In recent years, major CAD vendors have made progress in supporting point clouds, with tools like Autodesk's Civil3D and Revit, and Bentley's ContextCapture enabling direct referencing of point clouds for civil BIM. Alternatively, domestic point cloud processing software (e.g., TREND-POINT, ScanX) can be introduced to manage everything from noise processing to drawing creation. If in-house skills are lacking, outsourcing data processing alongside surveying tasks is also an option. In fact, services that handle 2D/3D drawings from point clouds have emerged, so outsourcing to experts is another viable solution. In any case, to fully leverage point clouds, it is necessary to build an environment that includes both software and skilled personnel. It is recommended to start with small-scale pilot implementations and gradually build up the internal skills for using point cloud data by providing more opportunities for field technicians to interact with the technology.

• Initial Implementation Costs:
  High-performance laser scanners, drones, and point cloud processing software can be expensive, and some companies may hesitate to invest in them. A typical ground-based laser scanner costs several million yen to tens of millions of yen, and drone LiDAR systems also require a significant investment, which can be a high barrier for small and medium-sized businesses. Additionally, training personnel to use these tools incurs both time and costs. A promising solution to this issue is the use of recently introduced low-cost equipment. For example, there are efforts to use LiDAR scanners integrated into iPhones and iPads for simpler point cloud data acquisition, and affordable survey gadgets like the smartphone-mounted RTK device "LRTK Phone" have also emerged. Renting equipment to reduce initial investment costs is another approach. The Ministry of Land, Infrastructure, Transport, and Tourism (MLIT) is also providing subsidies for introducing new technologies in i-Construction model projects, so considering the use of grants could be beneficial. As a result, affordable solutions are gradually becoming available, making it easier to move towards an era where each worker has their own device, even in sites that previously avoided point cloud usage due to high costs.

While the challenges mentioned above exist, technological innovations are providing solutions. The next tool, LRTK, is gaining attention as a new solution capable of addressing these challenges all at once.​

​

Introduction to LRTK

(A New Solution for Point Cloud Measurement)

​

Finally, we introduce the groundbreaking tool "LRTK" that makes the use of point cloud data even more accessible. The LRTK Phone is a small device that attaches to a smartphone, designed to allow anyone to easily acquire high-precision point clouds with absolute coordinates.

​

By simply attaching the LRTK Phone device to a smartphone, RTK positioning and 3D scanning become possible. Its pocket-sized form makes it convenient for on-site use.

​

Streamlining Point Cloud Acquisition with LRTK Phone:

​

By simply attaching the LRTK Phone to a smartphone, the functions of traditional surveying instruments and scanners are integrated into one device. Specifically, while scanning the surroundings with the smartphone's built-in LiDAR scanner and camera, the LRTK device's RTK-GNSS receiver adds centimeter-level position coordinates in real-time.

​

This allows for the acquisition of 3D point cloud data with absolute coordinates on-site. Traditionally, assigning survey coordinates to point clouds required post-processing to match reference points, but with LRTK, point clouds in the global coordinate system can be obtained during the field measurements, eliminating the need for post-measurement alignment. Additionally, with the dedicated app, it is possible to immediately perform analyses on the smartphone, such as measuring the distance between two points or calculating volume. For example, by scanning an embankment, the calculated volume can be instantly displayed on-site, allowing for quick verification of as-built conditions and quantity estimation.

​

The One and Only Tool for Acquiring Point Clouds with Absolute Coordinates:

​

What sets LRTK apart is that it is the only tool that allows for easy acquisition of point clouds with absolute coordinates. While many smartphones and tablets are equipped with LiDAR, these devices alone can only capture relative shape data and cannot perform precise positioning. By combining high-precision positioning with the LiDAR functionality, LRTK Phone makes it possible for anyone to handle point cloud measurement with absolute coordinates, a task that traditionally required specialized surveying equipment and advanced skills.

Additionally, LRTK is compatible with Japan's satellite positioning service "Michibiki (QZSS)" and its CLAS signal, allowing RTK positioning even in areas outside of mobile signal coverage by receiving correction information from satellites. It also offers the flexibility to maintain high-precision positioning simply by changing antennas, making it effective in mountainous areas or tunnels. These features are unique to LRTK, making it truly a "one-of-a-kind" tool.

Streamlining Data Processing with Cloud Integration:

LRTK is not only a hardware tool but also integrates with cloud services. The measured point cloud data and high-precision images are automatically uploaded to the LRTK cloud, where they can be plotted and viewed on a map through a browser. Data sharing on the cloud can also be done with a single click, and by issuing a URL, 3D data and position information can be instantly shared with stakeholders. Additionally, LRTK cloud allows for the automatic processing and downloading of 3D models created through photogrammetry, leaving heavy processing tasks to the cloud while the user only needs to wait for the results. This makes it possible to process large volumes of point cloud data without needing a high-performance PC, and centralizes data management on the cloud. Security features are also considered, allowing for the addition of viewing expiration dates or password protection to shared links, ensuring the system can be used safely. The seamless flow from on-site acquisition -> cloud storage -> office use significantly improves the efficiency of point cloud data utilization.

With the introduction of LRTK Phone, "one device per person" high-precision point cloud measurement is becoming a reality. In practice, construction site managers and workers are now able to carry the LRTK in their pockets and quickly perform surveying or scanning whenever needed, creating a new style of operation that is already boosting productivity. The price is also very affordable compared to traditional surveying equipment, making it an attractive solution with a low implementation barrier. By providing a solution to the challenges while offering the benefits of point cloud data utilization, LRTK is becoming a powerful tool for a wide range of users, from large general contractors and small to medium-sized construction companies to surveying firms and infrastructure managers.

​

Conclusion:

​

Point cloud data, composed of countless points captured by laser scanners and drones, is a 3D data format that is revolutionizing design, construction, and maintenance in the civil engineering industry. The use cases for point cloud data are vast, including the enhancement of design drawing creation, the streamlining of as-built management, the advancement of infrastructure inspections, and the acceleration of volume calculations. By utilizing point cloud data, surveying tasks become safer and faster, design accuracy improves, construction errors decrease, and data can be leveraged as a long-term asset. On the other hand, challenges such as handling large data volumes and mastering specialized software remain. However, technological innovations like LRTK, the latest solution, are addressing these challenges, and the era where anyone can easily handle high-precision point clouds is fast approaching. Whether you are a large general contractor, a local construction company, or an infrastructure manager working on railways or highways, the fundamental knowledge and applications of point cloud data, as well as information on innovative tools introduced in this article, can be a valuable resource for advancing your on-site digital transformation (DX).