Photogrammetry in the Era of On-site DX: Easy, High-Precision 3D Measurement for Anyone with LRTK

In the construction and civil engineering industries, on-site digital transformation—commonly called “on-site DX”—has been accelerating in recent years. The background to this is a situation where chronic labor shortages and demands for reforming work styles require improvements in productivity, and movements to streamline and enhance on-site operations with digital technologies are becoming active. One of the technologies attracting attention in this context is “photogrammetry.” Photogrammetry is a method for creating three-dimensional surveying data (such as point cloud models) from site photographs, and compared with traditional surveying using heavy machinery or surveying instruments, it has the advantage of easily digitizing detailed as-built conditions. This article explains the role photogrammetry plays in the on-site DX era and how it can be used for point cloud measurement and as-built management. We also introduce practical uses of the latest solution “LRTK,” which enables anyone to perform simple, high-precision 3D measurement, and consider the benefits of introducing it on-site. Now, let’s take a closer look at the basics of photogrammetry and the use of LRTK.

What is photogrammetry? 3D measurement technology evolving with on-site DX

Photogrammetry is a technique that derives the position and shape of objects from multiple photographic images to generate 3D models or surveying data. By photographing a site from various angles with digital cameras or drones and analyzing those images on a computer, you can create high-density point clouds and terrain models. Traditionally, aerial photogrammetry using aircraft or drones was the focus, but in recent years, improvements in camera performance and advances in software technology have made it possible to obtain high-precision 3D models easily even from ground-based photography.

The principle behind photogrammetry is to detect corresponding points (feature points) among multiple images and calculate 3D coordinates by triangulation from the parallax of those points. This method is also called “Structure from Motion (SfM),” and with recent improvements in computer performance, it has become possible to reconstruct point clouds from large numbers of photos at high speed.

Photogrammetry is beginning to play a major role in construction and civil engineering as well. For example, it is increasingly common to aerially photograph a construction site with a drone to generate a terrain point cloud model and use it to calculate earthwork volumes (fill and excavation) and verify as-built conditions. Also, in initiatives such as *i-Construction* and ICT construction promoted by the Ministry of Land, Infrastructure, Transport and Tourism, the use of 3D data generated by photogrammetry is recommended, making it one of the trump cards of on-site DX. Photogrammetry is an advanced technology that can digitally measure wide areas in a short time compared with traditional manual surveying, contributing to site digitization and operational efficiency.

Point cloud data obtained by photogrammetry and its use in as-built management

A representative deliverable generated by photogrammetry is “point cloud data.” Point cloud data are collections of many points that make up an object or terrain, each point having X, Y, Z coordinates in real space (and sometimes color information). Point clouds obtained by photogrammetry form precise 3D models as if the site were scanned in place, recording details from terrain folds to structural features.

Such high-density point cloud data become powerful tools in as-built management. As-built management is a quality control process that verifies whether completed structures and ground surfaces match the design in shape and dimensions. Traditionally, a few survey points were extracted and measured for height or distance to make judgments, but using point cloud data allows comprehensive checks of the entire site. Concrete use cases include the following:

• As-built verification and quality inspection: By overlaying the acquired point cloud model with the design 3D data (BIM/CAD models or reference planes), you can inspect the post-construction shape down to the smallest detail. Generating a heat map that color-codes errors makes differences from the design (areas where fill is lacking and lower than intended, or areas overfilled and higher than intended) immediately apparent. You can calculate missing fill volumes on-site and instruct additional filling, or immediately consider reducing overfilled areas, greatly reducing rework after inspection.

• Earthwork volume calculation: You can calculate the volume of any area from point cloud data. If you scan before and after excavation or filling, you can precisely calculate the as-built fill and excavation volumes, which is useful for as-built documentation and quantity management. Tasks that used to be done manually by surveying cross-sections and estimating volumes can now be performed quickly and with high accuracy using point clouds from photogrammetry.

• Disaster situation assessment: Photogrammetry is also powerful for rapidly understanding conditions immediately after a large-scale disaster. By quickly photographing affected areas from the air or ground and generating point cloud models, you can record the extent of damage as 3D data over wide areas. Using those data to plan restoration work or identify hazardous locations can be done rapidly, making photogrammetry valuable from a disaster prevention and mitigation perspective.

• Progress management and recordkeeping: If you photogrammetrically survey the site at each stage of construction, the records accumulate as a 3D construction history. This allows accurate reproduction of the site condition at any given time for later verification and enables objective data-based understanding of construction progress. Point cloud data make quantitative comparisons easier than photos, so you can clearly show, for example, how much soil was removed in a week. Also, if as-built data are shared instantly via the cloud, the site and the office can share the situation in a timely manner, smoothing consensus-building among stakeholders.

In this way, 3D point cloud data obtained by photogrammetry can be used widely, from post-construction quality checks to progress management and report preparation, dramatically improving the efficiency and accuracy of on-site work. Visualizing the site with digital data allows you to move away from judgments based on experience and intuition and realize reliable as-built management grounded in evidence.

Comparison with traditional surveying: Ease and improved accuracy of photogrammetry

Digital 3D measurement methods such as photogrammetry are dramatically easier and more efficient than traditional surveying. Previously, one had to measure points one by one with a total station or carry heavy, expensive equipment like terrestrial laser scanners to obtain point clouds, requiring skilled technicians to operate them over long periods. Even aerial photogrammetry with drones is not something anyone can use anytime: you must consider battery limitations and regulations (flight permissions and qualifications), and operations are affected by weather and radio environment. Moreover, to correct GPS positioning errors (on the order of meters), control points (ground control points) had to be measured in advance, requiring significant effort before assigning accurate coordinates to the acquired point cloud.

In contrast, the emergence of mobile scanning technologies, including photogrammetry, has made on-site 3D measurement markedly easier. Particularly with smartphone-based methods, site personnel can complete surveying simply by walking the site with a smartphone in hand. In one case, a mobile scan combining a smartphone’s built-in LiDAR and photogrammetry reportedly completed as-built surveying that used to require 2–3 people and more than half a day in less than an hour. Not only can surveying time be drastically reduced, but because each team member can survey in parallel, “waiting for surveying” time drops to zero, improving the overall construction management cycle. Also, equipment costs are orders of magnitude lower than for dedicated devices, enabling each site staff member to carry high-precision surveying tools—a new era where “one device per person” is becoming realistic.

As for measurement accuracy, photogrammetry has also improved dramatically in recent years. Point clouds obtained by smartphone or drone photogrammetry can meet the Ministry of Land, Infrastructure, Transport and Tourism’s as-built management standards (typically around ±50 mm (±1.97 in)) if proper procedures are followed. In fact, in 2022 the ministry’s guidelines added “3D measurement using mobile devices,” and results with errors of about 5 cm (2.0 in) from smartphone photogrammetry were officially confirmed. Moreover, by combining positioning with RTK (high-precision GNSS augmentation), positional errors can be reduced to a few centimeters (a few inches). Although this is slightly inferior to millimeter-level accuracy achieved by traditional laser scanners or total stations, it is practically high precision for civil engineering and infrastructure maintenance. The ability to capture large amounts of point cloud data in a short time enables areal precision management of the entire site, delivering value beyond purely local millimeter-level precision.

Thus, with the rise of photogrammetry, the traditional notion that “high-precision surveying = expensive and specialized” is being overturned. Affordable, easy-to-use equipment and automated analysis software have created an environment where anyone can perform 3D surveying quickly when needed. The ease and sufficient accuracy of photogrammetry provide a significant advantage in promoting on-site DX.

Simple surveying anyone can do with LRTK: Turning a smartphone into a high-precision 3D measurement tool

A trump card that has emerged to enable everyone to use photogrammetry on-site is the smartphone RTK positioning solution “LRTK.” Currently, this pocket-sized surveying instrument is quietly gaining popularity among construction management engineers and workers, and the idea of one device per person is becoming realistic. LRTK is an ultra-compact RTK-GNSS receiver that attaches to an iPhone or iPad, turning a smartphone into a centimeter-level (half-inch-level) surveying instrument. Its mechanism is simple: attach the LRTK device to the smartphone port and launch the dedicated app. The LRTK terminal also supports the Japan Quasi-Zenith Satellite System “Michibiki” centimeter-level positioning augmentation service (CLAS), enabling stable high-precision positioning even in mountainous areas outside of communication coverage. This gives high-precision latitude, longitude, and elevation information to each point scanned by the smartphone’s built-in camera or LiDAR in real time. In other words, if you take photos or scan with a smartphone, you can immediately obtain point cloud data or surveying coordinates that conform to public coordinate systems.

The actual measurement procedure is also simple. For example, photogrammetry using LRTK follows a flow like this:

• Attach the LRTK device to the smartphone and launch the dedicated app.

• Walk through the area you want to survey, scanning the target with the smartphone’s camera or LiDAR (take photos at key points as needed).

• Tap the positioning button in the app to record the coordinates of any desired survey point and the surrounding point cloud.

• After scanning, upload the acquired data to the cloud on the spot and automatically generate a 3D model.

• From an office PC, check the point cloud model on the cloud and perform volume calculations and comparison analysis with design data.

With LRTK, you can perform “simple surveying” with intuitive operations even without specialized surveying knowledge. For example, just point the smartphone at the point you want to measure and press a button in the app to record the coordinates (latitude, longitude, elevation) at centimeter accuracy. The acquired point cloud data can be uploaded to the cloud as-is, allowing immediate checking from an office PC and overlaying with drawings for analysis. Even without dedicated CAD software, web-based functions to measure distances, areas, and volumes are available, enabling workflows such as creating as-built drawings and reports from data collected on-site on the same day. With only a smartphone and a pocket-sized receiver, tasks that used to require surveying sets costing millions of yen can be handled by a single person—making it a true on-site DX tool suited to the “one device per person” era. The dedicated app is designed with an intuitive UI, so even site workers unfamiliar with digital devices can master it with a short learning curve, adding to its ease of introduction.

Furthermore, LRTK has high affinity with AR (augmented reality) functions, allowing you to project acquired 3D data or design models onto the real world through the smartphone screen. For example, if you overlay reference points or the finished positions of structures from the design on the smartphone camera view, it becomes a powerful guide for on-site layout marking. Positioning tasks that previously relied on craftsmen’s intuition can be performed accurately by anyone using AR, improving not only surveying but also the precision and efficiency of construction itself.

By utilizing LRTK in this way, you can perform everything from point cloud acquisition by photogrammetry to various measurements and even construction support in one stop, significantly changing on-site operations. Without outsourcing to specialists or renting expensive equipment, your in-house engineers can carry out high-precision 3D measurements daily, directly reducing surveying costs and improving work efficiency. Real-time digital twinning of the site also enables early detection of quality problems and prevents rework, allowing safe and waste-free construction management.

In addition, integration with cloud services is another major strength of LRTK. Data measured and surveyed on-site can be uploaded to the cloud from the LRTK app with a single button, and office personnel can immediately check results on a web map or 3D view. The boundary between site and office disappears, enabling appropriate instructions and decisions even from remote locations. Because data are centrally managed in the cloud, they can be securely shared with clients or subcontractors via URLs, and past records can be easily reviewed in chronological order. In this way, cloud integration connects site and office through data and makes real-time surveying and as-built management increasingly feasible.

Conclusion

3D measurement technologies centered on photogrammetry have become indispensable in the on-site DX era. In fact, cases requiring submission of 3D as-built data in public works are increasing, and introducing photogrammetry helps respond to such demands. It is likely that these simple 3D surveying technologies will become a new standard in the future. Especially by introducing easy and high-precision solutions like LRTK, point cloud measurement and as-built management—which used to be left to specialists—can be carried out by site staff themselves. By leveraging digital technologies to dramatically improve productivity and accuracy and realize waste-free smart construction, consider adopting photogrammetry and LRTK. The new form of on-site measurement brought by photogrammetry and LRTK will concretely advance on-site DX and greatly contribute to productivity improvements in the future construction and civil engineering industries. Please experience their effects on-site.