Photogrammetry at the Cutting Edge: Recent Cases in Construction, Surveying, and Infrastructure Management

In recent years, photogrammetry using drones and smartphones has begun to spread rapidly on construction and civil engineering sites. Photogrammetry is a method for reconstructing three-dimensional models (point cloud data or mesh models) of objects from multiple photographic images. This makes it possible to record and reproduce entire on-site structures and terrain as digital twins, and to efficiently acquire detailed data that conventional cross-section surveys or visual records could not obtain. With aerial photography by drones equipped with high-performance cameras and LiDAR, and mobile scanning using a smartphone plus RTK positioning, anyone can increasingly perform high-precision 3D measurements easily. This article highlights three fields where photogrammetry is active: structural inspection, earthwork quantity management, and disaster recovery. For each, it explains how photogrammetry is used and what advantages it brings in terms of accuracy and efficiency, comparing it to conventional methods in a technical but easy-to-understand way. At the end of the article, we also introduce a simple surveying solution using LRTK to support on-site adoption.

Using Photogrammetry for Structural Inspections: Safely Creating Detailed 3D Models of High and Confined Spaces

Photogrammetry is bringing a fresh approach to regular inspections of social infrastructure such as bridges, tunnels, and dams. Traditional bridge inspections have centered on manual work: deploying aerial work platforms or erecting scaffolding to access the underside of girders and bearing parts, visually searching for cracks and measuring widths with a scale. These methods require many personnel and much time, and they involve unavoidable hazardous tasks at height. But the introduction of photogrammetry is beginning to transform this process.

For example, in bridge inspections, a drone can photograph the entire structure from below down to the smallest details, and hundreds of photos can be used to generate a precise 3D model of the bridge. Alternatively, a mobile scan method using a smartphone fitted with a small RTK-capable GPS antenna is also effective. Workers can simply record video with a smartphone pointed at the structure from a walkway or riverbank, and obtain point cloud data of the entire bridge, including hard-to-approach areas such as the undersides of girders and bearings (the joints that support the bridge). Because the generated 3D point cloud model also has photographic textures applied to each part, detailed inspection and analysis can be performed on the digital twin from the office. The location and shape of cracks or concrete spalling can be accurately confirmed on the model, and actual measurements can be used to objectively assess the extent of deterioration. Photogrammetry-derived point cloud models record the entire structure comprehensively, so even minute cracks that might be overlooked by manual inspection are not missed. By accumulating 3D models from regular inspections and comparing them, tracking of aging changes becomes easy. Changes such as crack progression or component deformation compared to the previous inspection can be quantitatively detected on the order of millimeters, aiding long-term infrastructure health assessments.

The benefits of using photogrammetry for structural inspections can be summarized as follows.

• Non-contact, safer high-elevation inspections: Inspections of high or confined areas such as bearings and girder undersides can be completed by photographing from the ground. This greatly reduces the use of aerial work platforms and temporary scaffolding, lowering the risk to workers.

• Increased inspection efficiency: Drones and smartphones can scan wide areas in a short time, dramatically shortening site survey durations. Some cases report labor reductions of about 30–40% compared to conventional methods, and reduced night work or traffic restrictions can also lower costs.

• Precise deterioration diagnostic data: Crack widths and component dimensions can be accurately measured on the point cloud plus photo 3D model, providing objective evidence to support repair decisions. Because even tiny damages that are hard to find manually can be preserved as digital data, the reliability of diagnoses is improved.

• Data accumulation and utilization: By storing inspection results as 3D data in the cloud, they can be used for future repair planning and durability assessments. Unlike paper records, a digital archive preserves deterioration conditions semi-permanently, contributing to more advanced infrastructure maintenance management.

Photogrammetry for Earthwork Quantity Management: Improved As-Built Measurement Accuracy and Instant Analysis

Photogrammetry is also proving powerful for as-built quality control (inspection of the finished shape after construction) in earthworks and paving. Conventional as-built measurements involved survey staff walking the site and measuring heights and thicknesses at various points with leveling rods and levels. For example, in road construction, the thickness of a roadbed or pavement would be sampled at dozens of locations after completion to confirm that height and slope meet design. This method inherently limits the number of measurement points, creating a risk of failing to capture overall slope changes or local irregularities. Manual surveying is time- and labor-intensive and imposes a significant burden for verifying large areas.

Introducing photogrammetry into as-built management solves these problems at once. If the entire construction surface is photographed by drone or smartphone scan immediately after completion, high-density point cloud data covering the entire pavement or embankment can be obtained in just a few minutes. For example, a site supervisor scanning the shoulder with a smartphone right after paving can capture a 3D point cloud that records the pavement surface across the entire width and length of the road without omission. Overlaying this on the design finish model in the cloud allows immediate generation of a heat map that color-codes deviations from the design. Because the deviation from the design height can be visualized in centimeters, locations that are too high or too low according to standards become immediately apparent. Subtle surface unevenness in slope or insufficient pavement thickness that previously could only be detected by cross-sections or leveling surveys will not be missed on this heat map. Point cloud data also allow extraction of cross-sections or heights at arbitrary locations later, so longitudinal slope and cross-sectional shape can be analyzed without additional measurements. Even checks for localized puddles that went unnoticed on drawings can be completed on the 3D data.

Furthermore, 3D surveying by photogrammetry is highly effective for earthwork quantity management (calculating cut and fill volumes). For example, after embankment construction, a drone photogrammetry survey can capture the entire site terrain model, and the volume of imported soil can be automatically calculated from the difference with pre-construction terrain data. What used to require multiple manual cross-section measurements and volume estimation can now be completed accurately in minutes by differencing the point clouds. One site reported that soil volume measurement that previously took four people several days could be done by one person scanning for a few tens of minutes, greatly reducing time and labor costs. Volume calculations from point clouds have smaller errors than manual measurements and help ensure accurate accounting of as-built quantities. In addition, photogrammetry data are surface-wide and comprehensive, preventing measurement omissions that might later reveal localized shortages in embankment thickness. Data captured on-site can be immediately uploaded to the cloud and shared with stakeholders, enabling client inspections to be handled on the spot.

The main benefits of adopting photogrammetry for as-built and earthwork quantity management are:

• Improved accuracy and quality: Because the construction surface can be measured comprehensively with point cloud data, deviations from design can be checked to the millimeter level, enabling 100% verification of heights and thicknesses over wide areas and reliably detecting previously overlooked irregularities or localized nonconformities.

• Speed and labor savings: A single scan acquires many points at once, greatly reducing the labor of measuring each point manually. Survey work is accelerated and results can be obtained and shared in real time. Reduced night work and traffic restrictions also lower safety and cost burdens.

• Improved safety: Time spent surveying on roadways or slopes is minimized, enhancing worker safety. Measuring with tape on traffic lanes or entering steep slopes becomes unnecessary, reducing accident and fall risks.

• Immediate data utilization: Volume calculations and cross-section creation can be performed from the on-site point cloud, enabling as-built inspection and quantity confirmation to be completed simultaneously. Cloud integration allows supervisors and clients in the office to view data in real time, shortening the time lag to inspection approval.

Photogrammetry for Disaster Recovery: Rapid Damage Assessment and Measurement

In disasters such as earthquakes and heavy rain that cause landslides and river floods, photogrammetry contributes greatly to quick damage assessment and recovery planning. Traditionally, immediately after a disaster, limited personnel would visually inspect damaged areas and use surveying instruments to measure some cross-sections to estimate collapsed soil volumes. However, entering sites at risk of secondary disasters is difficult, and measuring vast areas quickly is not easy.

With photogrammetry, detailed 3D data of affected areas can be obtained safely and remotely even with a small team. For instance, at a large landslide site, aerial drone imagery can be used to create a point cloud model of the collapsed terrain, enabling precise measurement of the collapse extent and the volume of displaced soil. If the slope had been scanned in normal times and baseline data saved, photographing the same area after the disaster allows automatic calculation of how much and where collapse occurred by differencing the before-and-after point clouds. In some cases, soil volume calculations that used to take several days have been completed in minutes, dramatically improving the speed of initial response. The resulting numerical data serve as evidence for selecting recovery methods, arranging construction machinery, and planning soil disposal. Measuring dimensions and slopes on the point cloud model also enables accurate route selection for temporary roads and risk assessment for secondary collapse.

The advantages of photogrammetry in disaster response are safety and immediacy. Dangerous areas where people cannot enter can be assessed by photographing them from a distance with drones or high-zoom cameras. This reduces secondary disaster risk for rescue teams and engineers while collecting the necessary information in a short time. If a 3D terrain map or orthophoto of the site can be obtained within hours of the disaster, agencies can share and visualize the full extent of damage three-dimensionally, enabling rapid decision-making. In fact, at a large debris flow in 2021, comparing pre-existing baseline terrain data published by the government with post-disaster photogrammetry data allowed accurate calculation of collapsed soil volumes and identification of the embankment location that was a causal factor. In this way, comparing pre- and post-disaster digital twins can advance damage analysis.

Additionally, 3D models created by photogrammetry can be used for subsequent recovery work and safety management. For example, unstable areas can be extracted from the point cloud of a collapsed slope and those areas can be displayed on-site via AR to visually indicate restricted zones. If a dangerous area is overlaid in red on a smartphone screen against the actual scenery, workers can intuitively understand how close is too close. AR can also project planned reinforcement placements or retaining wall positions from recovery designs onto the site, aiding precision control during construction. Traditional site judgments for disaster recovery that relied on experience are evolving into scientifically grounded measures through visualization with 3D data and AR.

Simple Surveying Tools to Support On-Site Adoption: Easy Photogrammetry with LRTK

Even if one understands the benefits of photogrammetry, some may feel that adoption is difficult because it seems a sophisticated technology. However, simple surveying tools that anyone can use have emerged, greatly lowering the barrier to photogrammetry. A representative example is LRTK, developed by Refixia, a startup from Tokyo Institute of Technology. LRTK is a pocket-sized RTK-GNSS receiver that can be attached to a smartphone, turning the phone itself into a centimeter-level accuracy (half-inch accuracy) surveying instrument. It also includes a dedicated app that links with the smartphone camera and LiDAR sensor to perform photo capture → point cloud generation → positioning in a one-stop workflow. With this single device you can obtain position coordinates, perform 3D point cloud scans, and even project models using AR, and the acquired data can be shared to the cloud in real time. The equipment price is set far more affordably than conventional surveying instruments, attracting attention as a portable surveying tool for “one device per person.”

With LRTK, photogrammetry surveying that previously required expensive laser scanners and skilled technicians can be practiced easily by anyone on site. For example, as-built measurements that used to take half a day with a total station can be completed by one person in a matter of tens of minutes using LRTK, with immediate cloud-based point cloud analysis. This dramatically streamlines surveying work and allows on-site confirmation of point cloud data so any missing areas can be re-scanned immediately. The intuitive smartphone app interface makes it usable even without specialized knowledge, enabling field staff to handle everything from measurement to data analysis and reducing outsourcing costs and alleviating staffing shortages.

LRTK also conforms to ICT construction and i-Construction initiatives promoted by the Ministry of Land, Infrastructure, Transport and Tourism, making it an easy-to-adopt solution for local governments and small-to-medium construction firms. It is quietly becoming a nationwide trend and is highly regarded as “easy 3D surveying with a smartphone.” Now that the technologies underpinning photogrammetry are this accessible, it may be a good time to consider introducing them into your company’s operations or regional infrastructure management. Adopting cutting-edge measurement technology can be an opportunity to dramatically improve both safety and productivity.