Table of Contents

• Relationship between volume calculation and point cloud data

• Characteristics and challenges of on-site photogrammetry

• End-to-end flow from point cloud generation to volume calculation and required tools

• Comparison with conventional methods (total station (TS) / GNSS-only) (accuracy, personnel, time, reporting)

• Time-series comparison of volume changes (visualizing fills, excavations, and design differences)

• Smartphone-based point cloud acquisition with LRTK and its benefits (capture assistance, positioning, cloud sync)

• Use cases on construction sites and integration with daily reports and reports

• FAQ

Relationship between volume calculation and point cloud data

In civil engineering sites, excavation and embankment volume calculation is a critical task for as-built management. Traditionally, cross-sections were created from surveyed terrain and volumes were calculated using methods such as the average-end-area method or grid method. Recently, however, by utilizing point cloud data composed of countless coordinate points, these volume calculations can be performed more efficiently and with higher accuracy than before. Point clouds obtained by 3D laser scanners or photogrammetry can record fine terrain undulations, enabling near-real reproduction of the ground surface. By comparing point cloud data captured before and after construction, the volumes of fill and excavation can be directly computed.

The principle of volume calculation using point clouds is simple: determine the volume difference between the surface models before and after construction. For example, in excavation works, the point cloud of the pre-excavation surface is overlaid with the post-excavation surface to compute the volume of soil that used to occupy the space between them. Because point clouds reconstruct the terrain as a surface from numerous measured points, there is no need to interpolate between sparse survey points as in traditional methods, allowing accurate quantity calculation that fully reflects terrain undulations. Also, once a point cloud dataset is obtained, you can change the calculation area or reference plane and recompute volumes as many times as needed, enabling flexible re-calculations or scenario analyses without additional surveying. These advantages make point-cloud-based volume calculation superior in both accuracy and efficiency, and it is becoming a foundational technology supporting the digitalization of construction management.

Of course, ensuring the quality of the point cloud itself is important for high-accuracy volume calculation. Acquiring a point cloud with no coverage gaps and sufficient density, aligning coordinates correctly to the reference coordinate system, and properly removing extraneous objects (machinery, trees, etc.) from the point cloud — meeting these conditions enables volume calculations with small errors. Field validations have reported cases where the as-built quantities calculated from point clouds differed from conventional manual survey results by only about 1%, confirming that under proper operation point-cloud-based volume calculation is sufficiently reliable for site use.

Characteristics and challenges of on-site photogrammetry

Now that point cloud data are known to be useful for volume management, the challenge at sites becomes how to obtain those point clouds easily. Traditionally, this required specialized equipment such as terrestrial laser scanners or survey drones and a survey team, but advances in photogrammetry have increased cases where site staff themselves acquire point clouds using smartphones or drones. On-site photogrammetry refers to the method of conveniently taking photos on site and generating 3D models (point clouds) from those images.

The greatest feature of photogrammetry is the ease with which wide areas can be measured using familiar devices. For example, aerial photography with a drone can capture photographic data of a large site from above in a short time, allowing surveying of steep or hazardous terrain without personnel entering the area. With smartphone photography, if you walk around the subject and capture a sufficient number of overlapping photos, specialized software can generate a high-density point cloud model. The ability to generate point clouds with an ordinary camera without expensive laser scanners is a major advantage and has attracted attention as a method to promote on-site DX (digital transformation).

However, there are several challenges in acquiring point clouds via photogrammetry. First, measurement accuracy depends on capture conditions. If photo resolution or exposure is insufficient, or if there are strong shadows or reflections on the subject, errors or data loss may occur during point cloud generation. If the capture area includes vegetation or debris, additional effort is required to extract only the ground surface data from the generated point cloud. Moreover, some processes are difficult to complete entirely on-site. Even if capture itself takes little time, subsequent image analysis (point cloud generation) often requires high-performance PCs or cloud services and several hours, causing a time lag from surveying to data processing to volume calculation. In conventional photogrammetry operations, it has been difficult to compute volumes in real time on site.

Additionally, using drones requires flight permission and skilled operators, and smartphone-only photogrammetry risks model position and scale being offset from real-world coordinate systems due to GPS location errors. Even if a point cloud can be generated from smartphone photos, if its vertical datum and position are unclear it cannot be used for as-built quantity calculations. Therefore, on-site photogrammetry typically required extra steps to ensure accuracy, such as installing ground control points (GCPs) or ensuring sufficient photo overlap. Although on-site photogrammetry has high potential, traditional techniques alone left gaps in achieving the ideal of “anyone can easily complete high-accuracy point cloud measurement on site.”

End-to-end flow from point cloud generation to volume calculation and required tools

So what procedures and tools are needed to actually generate point cloud data and calculate volumes? Below is a typical workflow.

• Data Acquisition (Measurement): First, capture the site terrain and soil conditions with a smartphone camera, LiDAR scanner, or drone-mounted camera. In photogrammetry, it is important to take sufficient photos from all directions of the subject and ensure overlapping areas between photos. With a drone, use an automated flight plan to take photos at set intervals, and with a smartphone, walk around the subject to photograph it from all angles. On the other hand, devices like the latest iPhones with built-in LiDAR can acquire point clouds in real time, in some cases completing point cloud generation on site.

• Point Cloud Generation (Processing): Next, generate a point cloud model from the captured photos. When photos are input into dedicated photogrammetry processing software or cloud services, feature matching and image analysis reconstruct camera positions and 3D point clouds in space, producing point cloud data on the order of millions to tens of millions of points. This processing requires computing resources and time, but with high-performance PCs or GPUs, it can be done accurately and quickly. Recently, cloud services that automatically generate point clouds simply by uploading photos have appeared, so if you send data via the internet from the site, point cloud generation may be finished by the time you return to the office.

• Georeferencing: To use the generated point cloud data for volume calculation, it must be positioned properly in a real-world survey coordinate system. This step gives the model scale, orientation, and vertical reference so it aligns with physical coordinates. For drone photos, pre-installed ground control points (GCPs) are included in photos and the model is aligned to real coordinates using their known coordinates. For smartphone photos, you can photograph known points on site and match them in post-processing, or — as described later — acquire point clouds with absolute coordinates from the start using high-precision positioning features. In any case, to compare point clouds from different dates or to overlay design data, all data must be on the same coordinate datum. Proper georeferencing makes point clouds useful measurement data that can be handled in GIS or CAD.

• Volume Computation: With aligned point cloud data, specify the area for which you want volume and perform calculations. Typically, volumes are calculated using a TIN (triangulated irregular network) terrain model generated from the point cloud. Concretely, you can integrate the volumes of cut and fill relative to a reference elevation, or overlay two terrain point clouds at different times and compute the volume difference. Civil-engineering 3D software or point cloud processing tools are used for these calculations, and once the workflow is set the software can automatically compute volume values. For example, to determine the fill volume within a specific area, simply specify the polygon and the software will quantify the fill volume from the points inside. Open-source point cloud processing software increasingly includes volume calculation features, and cloud-based point cloud viewers that load point cloud models in a web browser and display volumes have also emerged. Even without dedicated software, environments for checking as-built quantities via a browser are becoming available.

• Sharing and Reporting Results: Finally, share the calculated volumes with stakeholders and compile them into reports. Traditionally, results were organized in Excel and cross-sections drawn on drawings with color coding to explain quantities. Using point clouds enables more advanced sharing. You can attach screen captures of colorized difference volumes from a 3D point cloud viewer to reports or send cloud links to stakeholders so they can interactively inspect the site. Sharing the entire 3D dataset that underpins the results is a major advantage unique to point cloud utilization. Tools that support this workflow include capture devices (smartphone, drone), photogrammetry conversion software, point cloud processing software or cloud services, and viewer/share platforms. Whereas these components were once combined from separate tools, integrated solutions that provide them as a package have appeared in recent years.

Comparison with conventional methods (TS / GNSS-only) (accuracy, personnel, time, reporting)

Approaches using point clouds differ greatly from conventional methods that rely only on total stations (TS) or high-precision GNSS survey equipment. Let’s compare them in terms of accuracy, required personnel, time, and reporting.

First, on accuracy: single-point positioning accuracy itself is generally better with TS or high-precision GNSS. TS can obtain point coordinates with millimeter-level accuracy (0.04 in) using prism distance measurement, and RTK-GNSS with a reference station can limit horizontal errors to within a few centimeters (a few in). However, “accuracy” in volume computation is not just about small point errors but about how well the overall terrain shape is captured. When surveying with TS/GNSS, measurements are typically taken on a grid at intervals of several meters (several ft) or along selected cross-section lines, sampling the terrain. In this method, although each measured point may be accurate, small undulations between measured points may be missed and must be interpolated. In contrast, point clouds are a vast collection of points that continuously cover the terrain as a surface and capture the ground in detail. Therefore, small depressions or bulges can be detected with point clouds but might be overlooked by coarser traditional meshes. Overall volume computation error depends on undetected parts like these, so under identical conditions point-cloud-based volume computation can reach accuracy comparable to conventional methods. In fact, for large embankment works, reports indicate differences of about 1% between point-cloud-based volume calculations and those from the average-end-area method, confirming that point clouds can provide sufficient accuracy if proper procedures are followed.

Next, compare required personnel and time. Conventional methods required many manual steps from surveying to drawing creation and volume calculation. For example, a large development site might have taken a survey team of four working a full week (20–30 person-days in total) to perform terrain surveying, cross-section creation, and volume computation. Switching to photogrammetry plus point cloud processing in some cases reduced that to two people in one day (2 person-days). A drone may capture images from the air in about 15 minutes, and point cloud generation and volume calculation can be completed the same day. Thus, point cloud utilization dramatically reduces personnel and time for surveying. One operator can handle the smartphone or drone, freeing others for different tasks. Given the shortage of skilled surveyors, the ability of digital measurement tools that anyone can operate to mitigate labor shortages is significant.

Reporting efficiency also differs. Traditionally, compiling surveys into drawings and tables and preparing documentation to verify as-built quantities took time. Large sites involved internal checks and client inspections that complicated reporting. With point clouds, the 3D model itself can serve as evidence. Showing the point cloud allows stakeholders to intuitively understand the terrain and visually explain the basis for quantity calculations. Sharing a color-coded heatmap of differences or a point cloud view with volume values clearly communicates where and how much soil is in deficit or excess. Report creation is also eased because cross-sections and overview images extracted from the point cloud can be auto-generated and pasted into documents, reducing the workload on staff. In short, reporting that once required significant manual effort can be semi-automated with point cloud workflows. Point cloud use speeds up the entire cycle from surveying to reporting and accelerates on-site decision-making.

Time-series comparison of volume changes (visualizing fills, excavations, and design differences)

A strength of point-cloud-based volume calculation is the ability to grasp terrain changes over time. Construction sites change daily, but if you regularly acquire point clouds you can track fill and excavation progress quantitatively. For example, if a site captures drone imagery and creates point cloud models every weekend, you can graph weekly increases in fill volume or display color-coded maps highlighting where changes occurred compared to the previous week. Being able to compare and visualize volume changes over time allows objective management of construction progress. If progress is delayed, you can promptly arrange additional equipment, improving the PDCA cycle of construction management.

From an as-built management perspective, comparing differences from design data is also important. If you have point cloud data and the design surface model (planned ground surface), you can check overall whether the as-built conforms to the design. In excavation works, you can identify areas that have been excavated to the specified depth and areas where soil remains; in fill works, you can detect overfills exceeding the specified height, all from point cloud differences. Visualizing differences as a heatmap makes interpretation intuitive — for instance, coloring areas higher than the design surface red and lower areas blue instantly reveals where rework is needed. Showing spatial distributions of excess and deficit for cut and fill streamlines corrective instructions on site and helps prevent rework and material waste.

Moreover, accumulating terrain data is useful not only during construction but also for post-completion maintenance and disaster response. If you store the point cloud at completion, you can compare future surveys to quantify long-term changes. For example, if an embankment or fill structure subsides over years, the amount of settlement can be calculated by differencing past point clouds, and in the event of a large slope failure you can estimate collapsed soil volume from before-and-after point clouds. Traditionally, disaster response required manual post-disaster surveys to calculate damaged soil volumes, but remote measurements allow quick and accurate quantity assessment even in hazardous sites, aiding restoration planning.

Smartphone-based point cloud acquisition with LRTK and its benefits (capture assistance, positioning, cloud sync)

As noted, on-site photogrammetry faced obstacles, but new technologies have begun to overcome these and make it possible for “anyone to easily acquire high-precision point clouds.” A representative example is LRTK. LRTK is a smartphone-integrated high-precision positioning system provided by Reficsia Inc., where a small dedicated antenna is attached to an iPhone or similar smartphone and network RTK corrections are used to improve the phone’s position to centimeter-level accuracy (cm level accuracy (half-inch accuracy)). This attaches accurate coordinates to each point of the point cloud obtained by the phone’s camera or LiDAR, enabling smartphone measurements to achieve accuracy comparable to surveying instruments. Previously, high-accuracy 3D surveying required drone + GNSS base station setup or expensive laser scanners, but LRTK can replace those with a single handheld smartphone, which is revolutionary. No specialized equipment operation knowledge is required; site technicians can use it as an extension of their everyday workflow. Compared with other 3D measurement methods, LRTK’s lower introduction cost and lack of need for vehicles or power supplies mean it excels in agility — you can measure immediately when needed — making it suitable for frequent everyday measurements.

Organized by “capture assistance,” “high-precision positioning,” and “cloud sync,” the benefits of the LRTK solution are:

• Capture Assistance: The LRTK smartphone app generates point clouds in real time and displays them on the screen while capturing, allowing the operator to proceed without missing areas. For example, if scanning a slope misses blind spots in the point cloud, additional capture on the spot to fill gaps is easy. Data captured in multiple sessions are automatically aligned, so high-quality point clouds can be obtained without specialized post-processing. The app also displays capture guidance to navigate optimal capture routes and includes functions to retake images from the same position and angle for fixed-point monitoring, enabling anyone on site to measure without mistakes.

• High-Precision Positioning: LRTK’s main feature is improving smartphone GNSS positioning. With a network-RTK-capable LRTK antenna, smartphone GPS, which normally has errors of several meters, becomes accurate to within a few centimeters (a few in). All acquired point cloud data and photos receive global geodetic coordinates and can be used immediately for volume calculations and drawing comparisons without later correction by GCPs. Vertical accuracy is also high, so the data are directly suitable for reference-plane height comparisons and cross-section creation. For example, LiDAR scans from an iPhone Pro combined with LRTK yield point clouds with high-precision coordinates from the start, allowing measured data to be used as as-built deliverables. Removing the cumbersome georeferencing step that plagued conventional photogrammetry greatly simplifies the entire workflow.

• Cloud Sync: LRTK provides a cloud service integrated with the field app, enabling automatic data sharing and storage. Point clouds and coordinate-tagged photos captured by the app are immediately uploaded to the cloud after capture, eliminating the need to copy files via USB or perform file conversions back at the office. Uploaded point clouds can be displayed instantly in a 3D viewer in the cloud, and browser-based analysis features such as volume calculation and plotting are available. For example, the workflow “field smartphone scan → cloud automatic processing → immediate volume confirmation” can be realized as a one-stop process, reducing time lag between capture and analysis to nearly zero. As data accumulate in the cloud, you can centrally manage terrain changes from project start to finish and easily compare them in time series. Past point clouds can be retrieved for reference or shared with stakeholders for collaborative review. Immediate sharing of site information, difficult with paper reports, is now possible, strongly supporting faster and more efficient construction management.

LRTK thus provides an integrated system that enables “easy smartphone point cloud capture → on-the-spot high-precision volume calculation → cloud-based data sharing.” It solves the barriers to on-site photogrammetry — capture skill, positioning accuracy, and data processing environment — in one package and is a solution that directly contributes to labor-saving and efficiency on construction sites.

Use cases on construction sites and integration with daily reports and reports

Finally, let’s look at practical use cases of point-cloud volume measurement on construction sites and how to incorporate it into daily reports and as-built reporting.

• Progress Management and Use in Daily Reports: On one development site, a site supervisor scanned a pile of removed soil every evening with a smartphone + LRTK to instantly determine the day’s removed soil volume. The obtained numbers were recorded in the daily report and used to decide the number of machines and dump truck dispatches for the next day. Traditionally, daily soil quantities were approximated from the number of dumps and payload, but recording measured values has improved the reliability of daily reports and facilitated smooth information sharing among contractors. Reviewing accumulated weekly and monthly point clouds provides objective evidence of progress and as-built quantities for internal and external reporting, aiding plan revision and schedule control.

• As-built Inspection and Report Creation: When fill works are completed, the point clouds obtained with LRTK are compared with the design surface to verify as-built over/under volumes. Automatically computed cut and fill volumes can be used directly as as-built quantity tables, and cross-sections and 3D views extracted from the point cloud can be attached to reports for client explanation. Because the point cloud is solid evidence, clients can immediately check quantities without repeated on-site re-measurements under supervision, saving effort. LRTK cloud services often include one-button PDF report output in a prescribed format, enabling quick generation of measurement reports with photos, coordinates, and notes. Using this, you can quickly create original as-built documents that combine on-site photos and point clouds, improving reporting efficiency and quality.

• Safety Management and Special Applications: Point cloud technology is powerful for measuring hazardous areas where personnel cannot enter. On slopes at risk of collapse or disaster sites, quantities that previously could only be estimated remotely can now be accurately determined using drones or LRTK. There are actual cases where LRTK was used to estimate post-heavy-rain sediment accumulation to support prompt recovery planning. For post-completion structure management, repeated fixed-point point cloud captures can record long-term changes like an electronic medical record. Comparing 3D data captured from the same position and angle over time reveals trends in settlement or deformation numerically and visually, increasing the persuasiveness of maintenance reports. These are advanced applications that build on the availability of easy on-site 3D scanning.

By introducing LRTK-based simple surveying on site, you can instantly measure when needed and immediately share and report results. This greatly saves personnel and time while ensuring accuracy and reliability in as-built management, contributing to improved productivity, cost reduction, and enhanced safety across the project. Completing measurements in-house rather than outsourcing to specialists greatly accelerates construction management. Truly, “on-site completion of volume calculation and point cloud measurement” is the benefit. Smartphone-based point cloud measurement and real-time as-built management are becoming standard processes in future civil construction. Consider trying smartphone scans and LRTK-enabled labor-saving volume management on your site.

FAQ

Q1. Is the accuracy of volume calculation using point cloud data acceptable? A1. Yes. If point cloud data are acquired and processed using appropriate methods, accurate volume calculations are achievable. In general, point clouds from photogrammetry or laser scanners, when calibrated with reference points and measured at sufficient density, will produce volume computation errors within the same range as traditional survey calculations (within a few percent). Field validation has reported cases where the quantities calculated from point clouds differed from conventional methods by only about 1%. However, ensuring high accuracy requires no gaps in the acquisition area, removal of non-ground points, and correct coordinate alignment. If these conditions are met, point-cloud-based volume calculation is sufficiently robust for site use.

Q2. Are specialized skills required to handle photogrammetry or LRTK? A2. Traditional photogrammetry had aspects that required expertise and experience, but recent solutions have made operations simpler. Smartphone app scans are intuitive: follow on-screen instructions to capture photos and the app will automatically generate point clouds, so special photography skills are not necessary. For LRTK, simply attach the antenna to a compatible smartphone, launch the app, and follow the guide to perform positioning, point cloud capture, and volume calculation automatically. You don’t need to be familiar with technical terms or complex settings; the system is designed so site personnel can start using it after short training. Data processing and analysis are also automated in the cloud, so users mainly review results. Modern photogrammetry tools and LRTK are designed to be usable by anyone at the site level.

Q3. What equipment and environment are needed for on-site point cloud measurement? A3. Basically, devices with high-quality cameras (smartphone, tablet, drone, etc.) and supporting software/services are sufficient for point cloud measurement. For smartphones, LiDAR-equipped models like recent iPhones or iPads are preferred, but photogrammetry can work with ordinary cameras as well. With LRTK, you need a compatible smartphone, the LRTK antenna, and a mobile network environment to receive RTK correction information. For drone use, a GPS-equipped drone with a camera plus flight permissions and a trained operator are required. In all cases, cloud services or PC-based point cloud software are needed to process acquired data. However, all-in-one services like LRTK allow capture to cloud storage on-site with just a smartphone and antenna, eliminating the need for a dedicated PC. Environmentally, open sight lines and sufficient positions for photography are important for wide-area measurement. Also, photogrammetry is affected by weather and lighting, so measuring in sunny, front-lit conditions improves accuracy. For safety, avoid having personnel enter high or steep slopes; use drones, or extend the device with monopods or poles when the footing is poor.

Q4. Should I use smartphone scans or drone surveys for volume measurement? A4. It’s best to choose based on site scale and purpose. For large sites or those with many high areas, drone surveys efficiently capture overview point clouds in a short time. For small plots, interiors, or detailed measurements, smartphone scans offer agility. Smartphones don’t require takeoff/landing space and can be used in no-fly zones. For frequent routine measurements, handheld smartphones are easier to deploy. In practice, a hybrid approach is common: use a drone to survey the entire site and create a base map, and complement it with smartphone + LRTK scans for detailed areas or change points. Point clouds from both sources can be integrated into a common coordinate system, so choose the optimal method based on site conditions and required accuracy/frequency. In any case, both methods yield point clouds that can be used for volume calculations, so select according to your needs.