What is Point Cloud Data?

Point cloud data is 3D data consisting of numerous points in space, each with coordinate information as well as attributes such as color and intensity. It is obtained through laser scanning or photogrammetry and can accurately represent the shape of objects or terrain. For example, buildings and terrain can be measured and recorded with millimeter precision, making point cloud data widely used in the civil engineering and construction industries as the foundation for digital 3D models.

​

The Importance of Point Clouds in the Civil Engineering and Construction Industries

In recent years, the importance of point cloud data has significantly increased in the civil engineering and construction sectors, driven by the trends of i-Construction and digital construction. By utilizing point cloud data, tasks such as surveying and as-built verification, which previously relied on manual labor, are being streamlined, leading to improved construction accuracy and safety. With rich, real-time data available, design and construction planning can be more accurately executed, contributing to the reduction of rework and ensuring quality. Point cloud data has now become an essential technology in modern construction.

​

Design Workflow Using Point Cloud Data

​

Methods of Acquiring Point Cloud Data

​

There are several methods for acquiring point cloud data, and they can be selected based on the site conditions. The most common methods include the following:

​

• Drone Photogrammetry (SfM) – This method involves taking multiple photos from a drone in the air and converting them into point clouds using image analysis software. It is suitable for large-scale terrain surveying and has become increasingly popular in recent years.

• 3D Laser Scanning – Using ground-based or mobile LiDAR devices, lasers are emitted and reflected points are captured to scan the target object. This method allows for direct and highly accurate measurement of structures and terrain, and it can capture complex shapes in great detail.

• Combination with RTK Positioning – This method uses GNSS RTK (Real-Time Kinematic) technology to assign centimeter-level coordinates to the positions of drones or LiDAR devices. This allows the acquired point cloud data to be assigned high-precision absolute coordinates, simplifying the alignment process in subsequent stages.

​​

Processing and Analyzing Point Cloud Data

​

Raw point cloud data often contains noise and unnecessary points, so proper processing and editing are essential for its use. First, noise from people, machinery, or reflections that occurred during measurement is filtered out, and unnecessary points that degrade accuracy are removed. If point clouds are obtained from multiple locations, registration is performed to align the data, unifying the coordinate system and merging them into a single point cloud model. If necessary, the point cloud is downsampled, or categorized and organized by type (e.g., terrain, structures), which improves the efficiency of later analysis. Through this processing, high-quality point cloud data is obtained, leading to improved accuracy in subsequent stages and more reliable results.

​

Processed point cloud data can be used for various types of analysis. For example, distances and areas can be measured directly from the point cloud, and cross-sectional shapes can be extracted for comparison with design values. By maintaining the entire site as a digital collection of points, areas that were missed during manual measurements can be checked later, allowing for the identification of interferences or inconsistencies that were overlooked during the design phase. As a result, construction management and as-built management can achieve accurate site assessments with less effort compared to traditional methods, contributing to shorter construction timelines and improved safety. Additionally, since measurements can be taken remotely without requiring personnel to approach dangerous areas, there are significant benefits from a safety management perspective.

​

Steps for Creating Drawings from Point Cloud Data

​

To use point cloud data for creating design drawings, it is necessary to extract the required information and convert it into CAD data. The general process involves first identifying key features of the buildings or terrain in the point cloud data, and then connecting these points to create line and surface data. For example, in an office interior renovation, the process might involve manually clicking on the positions of walls and columns in the point cloud and tracing them to create a floor plan. This manual tracing method requires skilled operators but has the advantage of selectively drafting only the necessary parts.

​

On the other hand, recent advancements have led to semi-automation using point cloud processing software. When point cloud data is imported into specialized software, algorithms can automatically recognize shapes such as planes or cylinders and convert them into vector data. For example, from a building’s scanned point cloud, a polygonal model of the walls can be generated, and from there, sectional or elevation views can be automatically created. Since point cloud data allows for the accurate capture of a building’s dimensions and shape, converting it into 2D drawings such as floor plans, elevations, and sections can result in higher-precision drawings than traditional methods. Finally, when converted into a format editable in CAD software, this allows for the use of 3D models in construction management or the output of 2D drawings as needed.

​

Use Cases in Construction Management and As-Built Management

​

Point cloud data is producing tangible results in the management of construction sites and as-built (progress) management. Here, we will introduce several representative use cases.

​

Volume Calculation and Construction Planning

​

In earthworks, accurate volume calculations can be performed using point cloud data. By acquiring point clouds of the terrain before and after construction and comparing them, the volumes of embankment and excavation can be automatically calculated. Traditionally, surveyors would create cross-sections by area and manually calculate the volume, but with point clouds, the entire area can be measured at once, making the process faster and more accurate. Additionally, the 3D point cloud model of the current site can be used for construction simulations, helping to plan the placement of heavy machinery and construction procedures. For example, by overlaying the site’s point cloud with the design 3D data of the structures to be constructed, potential interferences and workspaces can be checked in advance, allowing for the development of an efficient construction plan. Construction plans based on point cloud data contribute significantly to preventing issues that might not be noticed on paper and to shortening the construction schedule.
 

Comparison with Design Drawings (As-Built Management)
 

In as-built management, point cloud data can be used to verify whether the completed structure has been constructed according to the design drawings. By obtaining point clouds of the structure or terrain after construction and comparing them with the design model or drawing data, the extent of discrepancies can be visually and quantitatively assessed. For example, in tunnel construction, if the excavation shape is measured with point clouds, the over-excavation or under-excavation can be immediately identified, allowing for decisions on whether additional excavation or backfilling is necessary. By overlaying point cloud data with the design data, misalignments in concrete thickness or steel frame positioning become immediately apparent, leading to early detection and correction of issues. This approach is more reliable than relying on visual inspections or scale measurements by site supervisors, significantly improving the accuracy of quality management. In fact, some sites have already integrated completed point cloud data with 3D design models for daily construction assessments. By incorporating point cloud "as-built checks," rework after construction can be reduced, and the quality of explanatory materials for clients can be enhanced.

​

Real-Time Progress Monitoring of Construction

​

Point cloud data is bringing innovation to construction progress management. By regularly scanning the entire site with drones or ground-based lasers, the progress can be digitally recorded over time, allowing real-time monitoring of the construction’s progress. For example, by point-clouding and overlaying weekly work areas, the progress of the work can be compared in 3D, making it easy to intuitively identify discrepancies from the plan. Uploading point cloud data to the cloud and sharing it with stakeholders allows for remote verification of the latest site conditions. In fact, major construction companies have started using point clouds, 360° photos, and BIM models to share data in the cloud, enabling "online site inspections" for construction management from their headquarters. This reduces the frequency of on-site visits, leading to reduced travel time and enabling immediate decision-making. By accumulating and visualizing the constantly changing site conditions as digital data, the use of point clouds significantly enhances the transparency and efficiency of construction management.

​

Utilization of LRTK and Its Benefits

​

What is LRTK? A New Approach to Point Cloud Acquisition

​

LRTK is an innovative solution that allows for the acquisition of point cloud data with high-precision positioning information using a smartphone. By attaching a compact RTK-GNSS receiver to a smartphone (iPhone/iPad) and integrating it with the device’s LiDAR scanner, it functions as a pocket-sized universal surveying tool. This enables point cloud measurement with centimeter-level accuracy without the need for specialized surveying equipment, allowing users to track their own position while scanning the surroundings. With LRTK, all acquired points are assigned high-precision global coordinates, solving the issue of point cloud distortion during scanning. Since anyone can handle coordinate-tagged point clouds without specialized knowledge, the barrier to on-site implementation is significantly lowered.

​

For example, the photo above shows the process of scanning the site with an LRTK device attached to a smartphone and immediately measuring the distance on the point cloud. With LRTK, high-precision point cloud data can be easily acquired through such simple operations. Additionally, the system allows for fast calculation of distances or volumes between two points right after data acquisition, enabling immediate access to necessary dimensions and quantities on-site. The ease of performing point cloud scanning and measurements with a single pocket-sized device aligns perfectly with the era where having one device per person is standard. There is no longer a need to carry heavy laser scanners or laptops around the site, and field workers can use the system flexibly for quick measurements whenever needed.

​

Benefits of Utilizing LRTK in Construction Management and As-Built Management

​

The introduction of LRTK makes the use of point cloud data more accessible and routine. For example, point cloud data measured with LRTK Phone can be directly uploaded to the cloud and instantly accessed via a browser from the on-site office or headquarters. Since no specialized software installation is required, point cloud data can easily be viewed and shared, allowing all stakeholders to share the latest data. In the cloud, not only are the coordinates of the acquired point cloud available, but there is also a viewer function that allows users to measure distances, areas, and volumes with a single click. This makes it easy to check as-built conditions using point cloud data scanned on-site, or to instantly calculate quantities and include them in reports.

​

Furthermore, using devices like LRTK 360, it is possible to integrate 360° camera images with point cloud data. By overlaying omni-directional photos with the point cloud data obtained from the smartphone’s LiDAR, clear records with real-life texture can be created. This allows users to "review the site" on the point cloud data as if viewing photographs, making it intuitive to understand the current conditions, even for non-experts. Time-sequenced 360° images are organized on a map and can be displayed side by side on the cloud, allowing for easy comparison of site changes like a continuous photo sequence. The ability to easily compare the changes at the same location makes it ideal for construction progress management and regular monitoring. For instance, by recording the site’s progress from foundation work to superstructure with LRTK, one can later retrieve point cloud data and photos from any specific date to verify the construction progress.

​

Another benefit of LRTK is real-time verification through integration with design data. Due to its high positioning accuracy, if the design BIM models or drawing data, which already have coordinates, are uploaded to the LRTK cloud, they can be automatically overlaid with the current point cloud for display. Advanced applications, such as displaying the design model in AR over the point cloud scanned on-site and immediately detecting discrepancies or construction errors, are becoming more realistic. This allows construction personnel to continuously check for differences between the design and the actual site during work, leading to early corrections of mistakes and improved construction quality.

As outlined above, LRTK is a solution that dramatically simplifies the acquisition, sharing, and utilization of point cloud data. Without relying on expensive specialized equipment or advanced expertise, it enables the digitalization of every corner of the site and supports design and construction management. To fully realize the value of point cloud data in transforming the design workflow, introducing tools like LRTK, which anyone can easily use, will be a key factor in the future of the construction industry.