Table of Contents

• What does generating point clouds with drones mean?

• Benefits of drone point cloud surveying

• Advantages of generating point clouds in the cloud

• Steps to start drone point cloud generation in the cloud

• Key points to succeed in drone point cloud generation

• Main applications of drone point cloud data

• Recommendation for simple surveying with LRTK

• FAQ

In recent years, with advances in drones and 3D technologies, you increasingly hear the term “point cloud data” on construction and surveying sites. Photographs taken by drones can be used to obtain terrain and structures as three-dimensional point cloud data, and this is being applied to construction measurements and as-built management. However, beginners may feel uncertain, asking “How do I get started?” or “Don’t I need expensive equipment or difficult software?” In fact, these days, by leveraging cloud services, you can start generating drone point clouds easily without specialized knowledge or a high-performance PC. In this article, we clearly explain the basics of generating point cloud data using drones, the benefits of using the cloud, concrete steps to get started, and points for success. At the end of the article, we also introduce an attractive tool for beginners to easily experience 3D surveying: “LRTK.” If you’re interested in drone point cloud measurement but have not yet taken the first step, please refer to this guide.

What does generating point clouds with drones mean?

First, point cloud data refers to three-dimensional data that represents real-world objects with countless points. Each point contains X, Y, and Z coordinates (position), and when the collection of points is displayed on a computer, the object’s shape can be reproduced in three dimensions.

How, then, can point clouds be generated with drones? The common method uses a technique called photogrammetry. A drone-mounted camera takes many photos of terrain and buildings from above at various angles, and by analyzing those photos with dedicated software or cloud services, high-density 3D point cloud data is generated. The process matches common feature points that appear in multiple images and calculates each point’s position from differences in camera viewpoints.

There is also another method: mounting a lightweight laser scanner (LiDAR) on a drone to obtain point clouds by direct laser measurement. Laser measurement has advantages, such as being able to capture the ground beneath tree canopy, but equipment costs and operational difficulty are higher, making it a steeper hurdle for beginners to adopt immediately. In contrast, photogrammetry can be undertaken with commercially available camera-equipped drones and analysis software (or cloud services), so it is an ideal method for first-time point cloud generation.

Benefits of drone point cloud surveying

• Efficiency and labor savings: Drone surveying enables one person to measure areas that previously required many people and days, in a short time. Because a drone can capture large amounts of data at once even on extensive sites, “missed spots” are reduced and the need for additional surveys is minimized. This major improvement in efficiency and labor reduction is highly attractive in the construction industry, where labor shortages are severe.

• High-precision and comprehensive measurement: Point cloud data records objects with countless points, so it provides more precise and comprehensive information compared to traditional point-by-point surveying. Since shapes can be digitally preserved down to millimeter-level detail, it is easy to remeasure any location later. Even without drawings, accurate 3D models can be produced, aiding quality control and improving design and construction accuracy.

• Improved safety: Laser measurement and drone photogrammetry are non-contact methods, enabling safe data acquisition in hazardous areas where people cannot enter. High-altitude or steep-slope surveys can be conducted remotely, reducing the risk of accidents during work. Tasks that previously required working at height or traffic regulation can often be minimized, contributing to on-site safety management.

• Smoother consensus-building through data sharing: Acquired point cloud data can be shared with stakeholders as 3D models on the cloud. Supervisors or clients in remote locations can check the site three-dimensionally on their PCs or tablets, improving communication. Information that was difficult to convey with drawings or photos becomes immediately clear in 3D, enabling persuasive explanations during as-built inspections or design-change discussions.

Because of these benefits, the Ministry of Land, Infrastructure, Transport and Tourism is promoting the use of ICT such as drone surveying under “i-Construction,” and point cloud technology has become a symbol of DX (digital transformation) in the construction industry.

Advantages of generating point clouds in the cloud

• No need for a high-performance PC: Generating point clouds normally requires a powerful computer to process many images, but with cloud services, the heavy computations are executed on the server side. You can operate from a standard laptop or tablet in the field without preparing an expensive workstation.

• Easy to adopt: There is no need to install dedicated software or perform complex setup. A simple workflow is provided where you upload photos in a web browser and start analysis with a button. User interfaces designed for beginners greatly reduce the effort of learning how to use software from scratch.

• Always access the latest technology: Because cloud services update algorithms and features on their side, users always benefit from the latest versions. You don’t need to worry about updating software or compatibility. New point cloud processing technologies and AI-based automatic classification features can be used immediately once added to the cloud.

• Easy for team use: Having data in the cloud makes it simple to share point cloud results with multiple people. Invite stakeholders to the same project, and they can view and measure 3D data from a browser without installing anything. Members in remote locations can share the situation in real time, making collaborative work smoother.

• Cost advantages: Cloud services are attractive because they let you start with low initial costs. Previously, software purchases or hardware investments could run into the hundreds of thousands of yen, but with cloud services you use only what you need when you need it. Monthly subscription or pay-as-you-go models are common, making it easy to manage costs per project. For small businesses that do not use services frequently, cloud services are an economically efficient option to take advantage of the latest technology without waste.

Steps to start drone point cloud generation in the cloud

• Prepare equipment and plan the flight: First, prepare the drone, camera, batteries, and other necessary equipment. For surveying, a drone equipped with a high-resolution camera is preferable. Plan the target survey area and flight course in advance, and set routes that efficiently cover the area. Aim for at least 70% front and side overlap between photos to obtain good results. Also check weather and wind conditions and choose a safe time to fly. Don’t forget to confirm whether flight in the airspace is legally permitted or whether prior notification is required, and check relevant drone regulations.

• Drone photography: Fly the drone according to the plan and take photos of the target area. Maintain the planned altitude and camera angle, and capture a series of highly overlapping images. Typically, an automated flight app flies in parallel, zigzag routes while photographing the ground. For buildings or structures, shooting from oblique angles as well reduces blind spots and yields a more complete point cloud. Monitor battery levels and aircraft condition at all times and prioritize safety during flights.

• Upload data to the cloud service: After shooting, access the cloud point cloud processing platform. Create a new project in the web browser and upload the multiple photos taken on site. If there are many photos, you will be transmitting a large volume of data at once, so it is best to use a high-speed, stable internet connection where possible. If photos were taken with a GPS-equipped drone, location information will be embedded in the image files and automatically mapped on the cloud. If you have known point coordinate data (such as GCPs), upload those at this stage to help improve accuracy during processing.

• Point cloud generation (analysis) in the cloud: After uploading the photos, start the point cloud generation process in the cloud. Follow the service prompts and click the button; the server will automatically perform image analysis (feature matching and 3D reconstruction). Even projects with tens to hundreds of photos can produce point clouds relatively quickly thanks to the cloud’s high-performance computing environment. Processing time can range from tens of minutes to several hours depending on data volume, but progress can be monitored in the browser. Users can proceed with other tasks while waiting, which also enhances productivity.

• Check and utilize the results: When processing is complete, review the generated point cloud data on the cloud. Use the platform’s 3D viewer to rotate and zoom the point cloud and check quality. Verify that the target objects are properly reproduced and that there are no missing areas. If there are unnatural noise points, some services offer filter functions to remove them. If everything looks good, download the point cloud data as needed to import into your CAD software, or perform volume calculations and cross-section creation on the cloud. The generated orthomosaic (composite nadir image) and 3D mesh models can also be used. Share completed deliverables via the cloud or attach them to reports to support on-site work.

Key points to succeed in drone point cloud generation

• Ensure sufficient photo overlap: The quality of point cloud generation is greatly affected by how much overlap the captured photos have. Ensure 70–80% overlap front-to-back and side-to-side so the same ground features are clearly captured in adjacent photos. Insufficient overlap can cause gaps or reduced accuracy in the point cloud.

• Consider weather and shooting conditions: For photogrammetry, it is important to capture clear, well-lit images. Choose daylight hours and avoid flights in rain or strong winds. Do not leave camera settings entirely to auto—fix focus and ensure a shutter speed that prevents blur. If exposure varies widely between photos, analysis accuracy will suffer, so shooting in manual mode to keep brightness consistent can be effective.

• Increase accuracy as needed: If elevation or coordinate accuracy is important, use RTK-capable drones or ground control points (GCPs) to provide georeference to the data. Using these methods allows you to attach precise latitude, longitude, and height information to the generated point cloud, achieving survey-grade accuracy. Although initial investment and effort increase, consider adoption for applications that require high precision such as public surveys or design work.

• Practice on small areas first: Instead of tackling a large site from the start, begin with a small area or a familiar structure. Experiencing the entire workflow with a small dataset and validating results will reveal areas for improvement for subsequent attempts. Gradually scaling up helps you gain skills while minimizing the impact of mistakes.

• Thoroughly follow regulations and safety management: Drone flights are governed by legal rules. Always check for no-fly zones and obtain necessary permissions in advance. Also perform battery and GPS checks and ensure monitoring to prevent third-party access—don’t neglect safety procedures. On site, be mindful of surroundings and operate cautiously; maintaining zero accidents is the most important priority.

Main applications of drone point cloud data

• Earthwork volume calculation and as-built management: From acquired point cloud data, you can calculate volumes of embankment and excavation and record finished terrain as-built diagrams. For example, by scanning the ground with a drone before and after construction and comparing, you can accurately determine the amount of earth moved. You can also cut cross-sections from point clouds to verify whether constructed elements match design dimensions and shapes. Point cloud data makes high-precision as-built management that was previously difficult much easier.

• As-built mapping and design review: Drone point clouds can serve as base data for high-density as-built surveys. From acquired 3D point clouds, you can create arbitrary plan views and longitudinal/cross sections to help designers accurately understand site conditions. It is possible to overlay design models (CAD data) on point clouds to check for clashes or simulate the suitability of planned routes. Intuitive 3D data helps identify issues that might be overlooked on traditional 2D drawings.

• Progress management and stakeholder sharing: By regularly recording construction progress with drone point clouds, you can track site changes over time. Overlaying point clouds from different survey times shows at a glance where and how much change has occurred, making them useful for schedule management materials. Sharing these data via the cloud with clients and team members conveys real conditions to those who can’t visit the site. Discussing progress using 3D models in regular meetings accelerates consensus building.

• Infrastructure inspection and disaster response: Point clouds from drone-mounted cameras or LiDAR are used for inspecting infrastructure such as bridges, dams, and tunnels, and for rapid situational assessment at disaster sites. Since data collection can be done safely in areas people cannot enter, these methods are effective for monitoring slopes at risk of collapse or measuring terrain changes after floods. Even in these special cases, the basic workflow is just taking photos and converting them to point clouds, so skills learned during normal operations are useful in emergencies.

Recommendation for simple surveying with LRTK

LRTK is an innovative solution that enables beginners to easily start 3D surveying. Based on the concept of smartphone + RTK, you attach a compact positioning device called the “LRTK Phone” to a handheld iPhone and launch a dedicated app to perform high-precision point cloud measurement. It brings centimeter-level positioning (Real-Time Kinematic: RTK), which previously required specialized equipment, to smartphones, allowing accurate position information to be attached to captured photos and LiDAR scans. The convenience of one pocket-sized device enabling a single person to survey a site is truly deserving of the term simple surveying. Even those without surveying experience can obtain 3D point cloud data simply by following the app’s prompts and pointing the phone, greatly lowering the barrier to on-site adoption.

Furthermore, LRTK seamlessly integrates with cloud services for data utilization. For example, if you upload point clouds acquired with LRTK or photos taken by drones to the LRTK Cloud, coordinate-attached point cloud models are automatically generated and can be edited and shared in the browser. Because you can handle large datasets without a high-spec PC, LRTK supports everything from small sites to large-scale projects. LRTK is an ideal tool for promoting on-site DX with the approach of “start small and gradually expand,” letting you try the latest 3D technologies while keeping initial investment low. If you’re a beginner interested in point cloud utilization, using LRTK is a great way to take an easy first step.

FAQ

Q: What equipment and preparations are needed for drone surveying? A: Basically, you need a drone with a high-resolution camera, a controller to fly the drone safely, batteries, and similar equipment. Also prepare registration with a cloud processing service (or dedicated software) to process the photos. If you want to align survey results with map coordinates, using dedicated high-precision GNSS (RTK-equipped drones or ground control points) improves positional accuracy. In addition, check legal procedures in advance such as confirming the flight area, obtaining necessary permits, and registering the aircraft (registration is required for aircraft weighing 100 g or more).

Q: Can beginners handle drone point cloud generation? A: Yes. In recent years, automated flight apps and cloud service interfaces have become robust, and beginners can obtain point cloud data by following the basic steps. However, safe operation and compliance with regulations are required for drone flights, so it is recommended to practice and start with relatively simple projects to build experience. Even if things don’t go perfectly at first, you will steadily improve by trial and error on small sites.

Q: Are there any downsides to using cloud services? A: The main considerations are dependence on the internet environment. Because you must upload large volumes of photo data, an unstable connection can increase processing time and communication costs. Also, cloud service usage fees are ongoing. However, many services use pay-for-what-you-use models and require no initial investment, so they can be cheaper than purchasing on-premises software. If you are concerned about data confidentiality, choose a reputable service and check the terms of use for how data are handled.

Q: What level of accuracy can be achieved with point cloud data? A: It varies by shooting method and equipment, but with adequate overlap and image quality in photogrammetry, point clouds generally achieve horizontal errors on the order of a few centimeters to tens of centimeters, and vertical errors of around 10–20 cm. Using RTK-capable drones or GCPs for proper correction can achieve centimeter-level accuracy. However, accuracy decreases in areas with dense vegetation or featureless surfaces such as water. Depending on required accuracy, consider using RTK, GCPs, or, in some cases, combining LiDAR.

Q: Are qualifications or permits required to pilot a drone? A: In Japan, registration of drone aircraft weighing 100 g or more has been mandatory since 2022. Also, flights outside visual line of sight, at night, or over densely populated areas (DID) require permission or approval from the Ministry of Land, Infrastructure, Transport and Tourism. A national qualification (license) system for “unmanned aircraft pilots” has also been introduced for certain operations. However, basic flights within visual line of sight may still be conducted without a license in many cases. Regulations may change, so always check the Ministry of Land, Infrastructure, Transport and Tourism’s latest guidelines before flying and take the appropriate procedures.