Can UAV Photogrammetry Be Used in Cultural Heritage Surveys? Six Points on Accuracy and Cautions

In cultural heritage surveys, balancing recording accuracy and work efficiency is a constant challenge. The shapes and site conditions of targets—buildings, stone walls, archaeological features, gardens, old roads, and traces left on slopes—are diverse, and which method is chosen within limited time affects the quality of the deliverables. In this context, photogrammetry using UAVs is attracting attention. Because it can reconstruct terrain and structural shapes in three dimensions from a large number of photos continuously taken from above, it is a powerful option when efficient documentation of a wide area is desired.

However, for cultural heritage surveys, simply being able to produce a 3D model is not sufficient. If you do not understand in advance the level of accuracy that can be expected, whether fine-detail reproduction will be adequate, how much the surrounding environment and weather will affect the results, and what needs to be prepared on site, the deliverables may fail to meet their purpose and require re-surveying. In particular, because cultural heritage places importance on highly reproducible records, a perspective slightly different from that used in general civil engineering or facility management is required.

In this article, we examine whether UAV photogrammetry can actually be used for cultural heritage surveys, organizing the discussion into six points focused on accuracy and precautions. As information to aid decision-making before implementation, we summarize, from a practical viewpoint, the situations where it is suitable and those where complementary methods will be needed.

Table of Contents

• Reasons why UAV photogrammetry is attracting attention in cultural heritage surveys

• Item 1 Excels at recording shapes over wide areas

• Item 2: Assess accuracy based on the target and purpose

• Item 3 Reproducibility varies depending on surface condition and lighting environment

• Item 4: Calibration and coordinate management determine the quality of deliverables

• Item 5 Flight planning and safety considerations are indispensable

• Item 6: More valuable when combined with other methods

• How to Proceed with Cultural Property Surveys Without Failing

• Summary

Why UAV Photogrammetry Is Gaining Attention in Cultural Heritage Surveys

The biggest reason UAV photogrammetry is attracting attention in cultural heritage surveys is that it makes it easy to record large areas in a short time. Roof surfaces, upper decorations, the overall undulations of a site, and relationships with the surrounding terrain—features that are difficult to grasp from ground-level observation alone—can be captured from above, which is a major strength. In particular, in places that are difficult to approach from the ground or where access is restricted, the ability to acquire information without physical contact is valuable in itself.

Because processing is based on photographs, another feature is that it is easy to record not only the shape but also the appearance of surfaces. In cultural heritage surveys, visual information is often important—not only pure dimensional measurements but also weathering conditions, missing parts, discoloration, traces of repairs, and the relationship with surrounding vegetation. UAV photogrammetry allows the subject to be viewed both as a three-dimensional model and as high-resolution images, making it a method that is easy to use as report material and as data for comparative analysis.

Furthermore, it is well suited for comparing the past and the present. By repeating capture and processing under consistent conditions, it becomes easier to grasp changes in shape, signs of collapse, and deformations of the ground surface over time. Because cultural properties, once damaged, can suffer irreversible loss, the practice of periodically accumulating records of their current condition is important. In that respect, UAV photogrammetry, which can record broad surface areas, can be an effective means.

However, judging whether something can be used based solely on the impression that it seems convenient is insufficient. In cultural heritage surveys, you must consider the survey objectives, required accuracy, preservation environment, on-site constraints, work organization, and the long-term usability of the data. From the next chapter, we will examine, in order, the six items you should address before adoption.

Item 1 Strong at recording shapes over wide areas

The first point is that UAV photogrammetry is well suited to recording extensive areas. This is a very significant advantage in cultural heritage surveys. For example, when a site such as a castle ruin is not a single structure but exists as an areal space that includes earthen embankments, moats, slope faces, stone walls, traces of enclosures (kuruwa), and surrounding slopes, a method that records each point from the ground makes it difficult to grasp the overall picture. With a UAV, it is easier to capture the entire site and its relationship with the surrounding environment together, making it easier to document the topographic context that is important for preservation and maintenance.

This strength is also effective in surveys of individual buildings. Elements of cultural heritage buildings—such as roofs, upper levels, eaves, and their interference with surrounding trees—can be difficult to confirm by looking up from the ground alone. Using UAV photogrammetry makes it possible to photograph the subject from multiple directions around its perimeter, thereby supplementing information that planar records can easily miss. In particular, combining not only elevation views but also oblique photographs improves the stability of 3D reconstruction and makes it easier to grasp the subject’s contours and its external and internal corners.

Another practical advantage is that it makes it easier to reduce the time spent on site. In cultural heritage surveys, there are cases where work cannot be carried out freely for extended periods on site due to time constraints for stakeholder attendance, coordination with public opening hours, and consideration for nearby residents and worshippers. UAV photogrammetry, which makes it easier to gather the necessary photographs in a short time, is a method that can be planned more easily under such constraints.

However, the ability to cover large areas is not the same as being able to deliver adequate quality across every area. The wider the area, the more demanding become flight altitude, number of photographs, overlap rate, processing load, and the difficulty of coordinate management. Expanding the target area without planning just because you can capture widely can result in each image lacking sufficient information density and failing to meet the required level of detail. In cultural heritage surveys, it is important to first clarify what you want to record and not try to capture both overall context and fine-detail records under the same conditions.

In short, UAV photogrammetry is very well suited for wide-area documentation, but its strengths lie primarily in capturing surfaces and providing a bird’s-eye view. With this in mind, it is important to adopt an approach of supplementing it with different capture methods for areas that require detailed inspection.

Item 2: Assess accuracy based on the subject and the purpose

The second point is how to evaluate accuracy. When considering UAV photogrammetry for cultural heritage surveys, one of the most common misconceptions is trying to judge accuracy by a single number. In fact, the accuracy of UAV photogrammetry is determined by a combination of many factors—object size, flight altitude, pixel density, photo overlap, shooting angle, the presence or absence of calibration, processing conditions, ground surface conditions, and so on. Therefore, you must first be clear about what you intend to use it for.

For example, the accuracy required depends on whether you want to capture the overall shape of the archaeological remains as baseline material for a conservation plan, examine trends in the tilt or displacement of an elevation, compare differences before and after repairs, or create visualization data for public release. Even in cases where UAV photogrammetry is perfectly usable when the main purpose is to understand the overall layout, if you need to precisely track very fine cracks or slight surface irregularities, UAV photogrammetry alone may be insufficient.

What is important here is to consider relative accuracy and absolute accuracy separately. Relative accuracy refers to how correctly the positional relationships among points within the model are reproduced. Absolute accuracy, on the other hand, means how correctly the model is placed with respect to the site’s coordinate system and known reference points. In cultural heritage surveys, when considering future comparisons and overlays with other materials, not only relative accuracy but also absolute accuracy is important. Even if a visually appealing 3D model is produced on site, if its position is ambiguous it becomes difficult to use as a long-term management record.

It's important to understand not only the numerical accuracy but also what it struggles with. UAV photogrammetry is strong for targets with stable surface geometry and where feature points can be easily detected in photographs; however, it tends to produce errors on monotonous surfaces, highly reflective surfaces, areas with excessive shadows, and locations where thin elements are densely clustered. In other words, even if the overall average accuracy is good, it is insufficient in practice if details important as cultural heritage are weak.

Whether to use UAV-based photogrammetry in cultural heritage surveys should be determined not by whether its accuracy is high or low, but by whether it is appropriate for the required deliverables. Setting expectations appropriately and using it within its areas of strength is the quickest path to success.

Item 3: Reproducibility Varies with Surface Condition and Lighting Environment

The third factor is the influence of surface conditions and the lighting environment. UAV photogrammetry reconstructs shapes by matching features in photographs, so it is greatly affected by how the target surface appears. This is particularly important for cultural heritage, because the materials of the targets and their aging processes are diverse, and the parts that are most valuable for study are often those with the most challenging imaging conditions.

For example, walls with uniformly colored surfaces, stone that has lost much of its pattern due to weathering, glossy roof tiles, wet paving stones, ruins covered with moss or fallen leaves, and archaeological sites in forests with strong shafts of sunlight can all make it difficult to obtain stable feature points across photos. If shadows change over time, the same location can appear like a different object, which can degrade reconstruction accuracy. Conversely, subjects with moderate texture that can be photographed under even, diffuse lighting are relatively stable and easier to model.

In cultural property surveys, because the object itself can often not be altered, it is essential to arrange conditions on the photographing side. Choosing the time of day for shooting and avoiding extreme backlighting and heavy shadows is fundamental. Also, when surface reflections are strong, you should not try to capture everything in one go but consider planning a reshoot at a different time of day. Especially for stone, metal, and plaster surfaces, differences between dry and wet states directly affect appearance, so it is safer to be aware of the weather conditions on the previous day and on the day of shooting.

In addition, attention must be paid to objects that move in the wind, such as plants or fabric. Even if they are not cultural heritage themselves, branches and leaves of nearby trees can block the view or cause moving shadows, which can negatively affect processing results. If the outlines of archaeological remains are hidden by grass, their boundaries may become ambiguous in photographs, and the shape of the ground surface may not be accurately reconstructed. In cultural heritage surveys, it is necessary to consider shooting conditions that include the surrounding environment, not just the object itself.

In this way, the quality of UAV photogrammetry is not determined by aircraft performance alone. It is necessary to consider how the subject appears, how light falls on it, and how it changes over time. Because each cultural heritage item is in a different condition, it is important not to rely only on generalities but to assess in advance how easy the specific subject will be to photograph.

Item 4 Calibration and Coordinate Management Determine the Quality of Results

The fourth item is calibration and coordinate management. If you use UAV photogrammetry as a professional tool in cultural heritage surveys, this topic cannot be avoided. When reconstructing shapes from photos alone, you may be able to create a visually tidy model, but its positional relationship to the actual site and its scale can become unstable. Therefore, calibration using known points and check points, unifying coordinate systems, and organizing vertical (height) reference standards are important.

The need for coordinates in cultural property surveys is clear. First, to compare data from multiple periods, the data must be alignable to almost the same reference each time, otherwise the comparison is meaningless. Next, to combine with other materials such as plan views, cross-sections, cadastral maps, repair records, and ground survey data, a common reference is necessary. Furthermore, when used for on-site repairs or conservation planning, it is required to be able to explain, with location information, where and by how much changes have occurred.

What is important here is that placing control points is not the end of the process. If the control points are unevenly distributed, even if part of the model looks good, distortions can occur in distant areas. If there is insufficient constraint in the vertical direction, elevation representation can become unstable even when horizontal positions are correct. Also, for cultural heritage sites there are restrictions on where control points can be installed, so placing them only in locations that are easy to install tends to result in a geometrically weak configuration. It is necessary to plan taking into account site-specific circumstances such as being unable to touch the object, not disturbing the landscape, and the need to remove installations immediately after placement.

How check points are handled is also important. If accuracy is judged only by the points used in the adjustment, there is a risk of overestimating actual reproducibility. On site, points used for processing should be separated from those used for validation so that the reliability of the final results can be evaluated objectively. In cultural heritage surveys, because data once acquired are often referenced in the future, it is also important to document the approach to accuracy verification at that time.

Additionally, if the coordinate system and vertical datum are left ambiguous, the model may not be able to be overlaid with other datasets later. Even if you deliver only a 3D model, insufficient reference information will reduce its reusability. In fields that assume long-term preservation and reuse—such as cultural heritage—the preparation of operational information about which references were used to acquire the model and how it was validated supports the quality of the deliverables more than the model’s appearance.

Item 5 Flight planning and safety considerations are essential

The fifth item is flight planning and safety considerations. When conducting UAV photogrammetry for cultural heritage surveys, it is precisely because the subjects are highly valuable that more cautious preparation is required than at ordinary sites. The surroundings of cultural properties can present a complex overlap of flight risks — areas with many visitors, proximity to residential neighborhoods, places with many trees or power lines, and elevated locations that are easily affected by wind. Furthermore, not only is the success of the imaging important, but it is also essential to avoid imposing any burden on the subject itself.

For example, in narrow shrine precincts or historic parks, securing takeoff and landing spots can be difficult. In places that act as wind corridors, the aircraft's attitude can become unstable, which affects image quality. Near old wooden buildings or delicate architectural details, planning to get too close may itself be undesirable. Whether you can fly safely depends not only on piloting skill but also greatly on the plan of where, in what order, and at what heights and distances you will fly.

In cultural property surveys, flight permits and coordination with stakeholders should be arranged from an early stage. There are parties involved that differ from those at ordinary construction sites—managers, owners, conservation staff, and local community members—and it is necessary to align understanding about the purpose and scope of imaging, whether it will be made public, and how acquired data will be handled. Consideration must also be given to visitors appearing in images and to images of adjacent private land. Because cultural properties attract high public interest, even if a flight is technically possible, it is necessary to separately consider whether it is appropriate operationally.

Moreover, a flight plan is, in effect, an accuracy plan. If the capture altitude is too high, resolution will be insufficient; if it is too low, efficiency will suffer and blind spots will increase. If image overlap is insufficient, reconstruction will be unstable; if it is too high, processing load and photo management become excessive. Because cultural heritage often has three-dimensional forms, it is frequently necessary to incorporate oblique shots as well as photographs taken directly overhead. As a result, flight routes cannot be limited to simple grid patterns and require three-dimensional arrangements tailored to the subject.

In short, UAV photogrammetry in cultural heritage surveys is not simply a task of flying and photographing; it is a planning activity to ensure the necessary quality without damaging the conservation target. Without a perspective that reconciles on-site conditions with considerations for the cultural property, even if data are acquired, it cannot be considered a good survey.

Item 6: More valuable when combined with other methods

The sixth item concerns combining it with other methods. In cultural heritage surveys, if you try to complete everything using only UAV photogrammetry, there will inevitably remain aspects that it handles poorly. However, by combining it with other measurement methods and ground observations, the value of UAV photogrammetry is significantly enhanced. This is important not only in the sense of compensating for weaknesses, but also in terms of increasing the overall reliability of the deliverables and their range of applications.

For example, UAVs excel at capturing wide-area terrain, roof surfaces, and upper structures. On the other hand, areas such as under eaves, interior spaces, surfaces hidden by trees, fine carvings, and sharply shadowed parts are better suited to ground-based photography and terrestrial measurements. If these are acquired by dividing the roles, you can create three-dimensional records with fewer gaps than with a single method alone. Cultural heritage objects have complex shapes, and missing parts affect evaluation, so it is important to design a combined acquisition approach based on an understanding of what can be captured and what is difficult to capture.

Also, combining this with coordinate management is important. Three-dimensional models created from photographs are visually excellent, but to make them useful for ongoing surveys and drafting, a connection to on-site coordinates is necessary. If you combine field positioning that accurately secures key points with management of known points, the usability in subsequent processes improves. In cultural heritage surveys, rather than completing everything in a single investigation, it is important to leave data that will withstand future repairs, public display, academic use, and post-disaster comparisons. In that sense, the combination of photogrammetry and positioning management is highly compatible.

Furthermore, depending on the recording objectives, the approach of acquiring the whole with a UAV and densely recording only the important parts using a different method can also be effective. This makes it easier to reconcile the overall context with local detail. In cultural heritage surveys, the demand to broadly document the entire subject and the demand to deeply record parts that are important for preservation often coexist, so it is more rational to combine methods according to purpose than to process everything uniformly with a single method.

When introducing UAV photogrammetry, rather than debating its superiority as a standalone technique, it is less likely to fail if you consider which part of the overall survey workflow it should handle. To avoid compromising the value of cultural heritage and to leave the necessary information for the future, it is important to regard methods as complementary rather than competitive.

How to Avoid Failure in Cultural Property Surveys

Taking the six points discussed so far into account, the key to successful UAV photogrammetry in cultural heritage surveys lies less in the choice of equipment than in advance preparation and role assignment. The first thing to do is to explicitly define in writing what you want to leave as a deliverable. Whether it is understanding the overall layout, documenting the current condition, comparing before-and-after repairs, or preparing material for public use, the shooting conditions and accuracy requirements will vary. If you shoot with an unclear purpose, you are likely to end up with a large volume of photos that still fail to produce the necessary deliverables.

Next, it is important to identify, from an on-site perspective, which parts of the subject are suitable for UAV photogrammetry and which are not. If you understand in advance the areas that are easy to see from above, those that are likely to be in shadow, those that are difficult to approach closely, those that are obscured by trees, and those with strong surface reflections, the accuracy of the survey plan will improve. Because re-shooting is often difficult for cultural heritage, assumptions made before entering the site are important.

Moreover, on site the approach to reference management is as important as the photography itself. Deciding from the outset which points will serve as references, where verification will take place, and what coordinate information to include in the deliverables stabilizes downstream processes. Because cultural heritage surveys often lead not only to investigation reports but also to repair planning, academic research, disaster response, and future comparisons, it is necessary to adopt a perspective that is not satisfied with models that can only be viewed on site.

And care is also required when checking after processing. Even if it looks good, you must verify whether there are holes or distortions in critical areas, whether the contours have been compromised, and whether you can derive the cross-sections or drawings you actually need. In cultural heritage surveys, it is essential to evaluate the results with the intended final use in mind. Do not treat the mere completion of the 3D model as the goal; it is important to confirm that it can be used for the necessary decisions and explanations.

UAV photogrammetry is a technique that, when used correctly, can greatly enhance the efficiency and documentation of cultural heritage surveys. However, it is not foolproof. That is why understanding its strengths, addressing its weaknesses, and planning according to site conditions and objectives are essential for success.

Summary

UAV photogrammetry is a fully usable technique for cultural heritage surveys. In particular, it is highly effective for purposes such as capturing wide areas, recording from an aerial viewpoint, rapid acquisition of areal information in a short time, and preserving the current condition for time-series comparison. On the other hand, there are also many practical points to be cautious about, including reproduction of fine details, areas in shadow, dealing with reflections and uniform surfaces, strict management of coordinates, and operations that require safety considerations.

Summarizing the six points discussed in this article, to make the most of UAV photogrammetry in cultural heritage surveys it is important to understand its strength in wide-area recording; determine the required accuracy according to the purpose; interpret the effects of lighting and material; carefully carry out calibration and coordinate management; thoroughly implement flight planning and consideration for stakeholders; and combine it with other methods as needed. In other words, the answer to whether it can be used is not a simple yes-or-no—based on practical experience, the conclusion is that it is highly effective when used with the correct premises.

In cultural heritage surveys, it is more important to preserve data in a form that can be used in the future than to simply create a 3D model. For that reason, planning needs to cover not only photography but also the organization of on-site reference points and the handling of positional information. If you want to make data acquired by UAV photogrammetry more usable and streamline the whole process up to on-site coordinate management and verification, combining it with a smartphone-mounted high-precision GNSS positioning device such as LRTK is an effective approach. When preparing records of the current condition of cultural heritage so they do not end up as merely attractive 3D data but become information that is easy to compare, explain, and connect to subsequent surveys, it is well worth considering methods that support simple on-site surveying.