How to 3D Measure Inscriptions | 6 Steps to Improve Recording Accuracy

Recording inscriptions is not merely an exercise in reading the characters. Objects that bear inscriptions—such as stone monuments, memorials, gravestones, boundary markers, votive offerings, and architectural elements with nameplates or engravings—often carry connections to why they were erected, their dates, the names of people involved, records of construction or donation, and the historical context of the area.

Therefore, it is valuable to record not only whether the inscription is legible, but also where it is located, how deeply and with what relief it is carved, and what kinds of damage or wear are present in the surrounding area.

Traditional recording methods centered on photography, rubbings, manual measurement, sketches, and drafting. Of course, these remain useful today. However, when dealing with shallow characters worn by abrasion, incisions whose appearance changes with shading conditions, inscriptions carved on complex curved surfaces, or objects with losses and chips, planar records alone may not fully convey the condition. 3D measurement is effective in such situations. By using 3D measurement, the line width and depth of characters, the curvature of the monument surface, the location of chips, and how the piece fits into its surroundings can be treated as three-dimensional information, allowing the creation of records that are easy to re-examine later.

That said, 3D measurement of inscriptions is not a task you finish by bringing equipment and tracing it once. The appropriate approach varies depending on the material of the object, the degree of weathering, whether it is outdoors or indoors, the strength of reflections, the required recording accuracy, and future uses. Creating a visually pleasing 3D model is not the same as leaving a record that is usable in practical work. What is needed is to design the entire workflow—on-site assessment, establishing standards, acquisition, processing, and preservation—tailored to the objectives.

This article organizes and explains a method for 3D measurement of inscriptions in six steps so that practitioners are less likely to be confused on site. It summarizes an approach that can be applied to a wide range of purposes, including cultural heritage surveys, preservation records, before-and-after comparisons of repairs, facility asset management, historic site maintenance, and digitization of local materials.

Table of Contents

• Reasons Why 3D Measurement of Inscriptions Is Required

• Step 1: First determine the purpose and required accuracy

• Step 2: Check the material and the state of deterioration of the target.

• Step 3: Set up the reference position and measurement conditions

• Step 4 Ensure complete capture to preserve character shapes

• Step 5 Improve readability in data processing

• Step 6: Set up file formats and sharing methods

• Common Pitfalls in 3D Measurement of Inscriptions

• Summary

Reasons Why 3D Measurement of Inscriptions Is Required

The primary reason 3D measurement is required for recording inscriptions is that it preserves textual information and shape information together, without separating them. An inscription is not only important for what is written. Information such as how deep the carving is, how much the letter contours have been worn, whether the stone surface is flat or curved, and which parts of the characters are affected by loss is also important in conservation and research. With 2D photographs, shadows can make inscriptions appear readable, but actual depth and deformation are difficult to grasp, and conversely, poor lighting alone can make them hard to read.

Especially on weathered stone or metal surfaces, the boundaries of characters can be difficult to discern with the naked eye. Even in such cases, if fine surface undulations are captured by 3D measurement, post-processing can be used to enhance height variations or to simulate oblique lighting. This makes it easier to closely examine information that was hard to see on site after returning. Being able to minimize time on site while allowing for multifaceted review afterward is a considerable practical advantage.

Also, 3D measurement is a method that is strong for comparisons. For example, if you want to see in chronological order differences in condition before and after repairs, the progression of weathering over several years, the expansion of surface delamination, or deformation around cracks, having 3D data acquired to the same reference makes it easier to track the amount of change. Changes can also be seen by comparing photographs, but the accuracy of such comparisons drops if the shooting position or lens conditions change even slightly. With 3D data, it is easy to overlay and compare on the same reference, making it more manageable for record-keeping tasks that require accountability.

Furthermore, inscriptions do not necessarily exist in isolation. They can carry meaning together with their surrounding environment, such as their pedestal, nearby stonework, differences in ground elevation, and their positional relationship to approach paths and structures. It is important to capture inscriptions alone in high detail, but if you can also document where and how they are installed, that information can more readily support maintenance and relocation planning, post-disaster restoration, and the improvement of wayfinding signage. In practice, it is important to regard 3D measurement not simply as a record of shape but as a current-condition record that includes position and relationships.

Step 1: Determine the purpose and required accuracy first

The first thing to do is clarify what the 3D measurement is for. If you proceed while this is ambiguous, you may collect more detailed data than necessary and only increase processing load, or conversely fail to achieve the required accuracy and end up with unusable records. For 3D measurement of inscriptions, the content of the data you need will vary depending on whether you want to capture the character relief to assist reading, preserve the overall shape as a conservation record, provide material for repair planning, or aim toward drafting and exhibition use.

For example, if you want to record the location and installation condition of an entire inscription including its pedestal, the overall shape and spatial relationships of the whole subject become important. On the other hand, if you want to check the contours of each individual character that has been worn down, the priority is how well the surface’s fine irregularities can be represented. These two may appear similar, but they require different resolutions and different capture methods. Data collected under conditions suitable for overall recording can make character details appear blurred, and capturing only enlarged images of the characters will lose their relationship to the whole. Therefore, it is important to plan separately for overall documentation and for detailed recording.

When considering required accuracy, simply aiming for higher precision is not necessarily the right approach. What’s important is the perspective of what minimum line width or depth you need to distinguish, which units you will want to use for comparisons in the future, and who will use the final deliverable and in what situations. Whether site personnel want to review conditions later, researchers want to consider the readability of inscriptions, or contractors want to reference it as a pre-repair baseline will change what level can be considered sufficient. Sharing the purpose of the 3D measurement in advance also makes it easier to realistically estimate the number of photos or scans required, scan density, on-site work time, and data volume.

At this stage, it makes things easier later if you roughly decide which approach to center on. Photographic methods tend to preserve surface patterns and color information, and by adjusting oblique lighting conditions you can help enhance the visibility of inscriptions. Laser-based and similar methods have strengths in the stability of shape capture. Because suitability depends on the object's size, surface reflectivity, and site constraints, you should choose a method by working backward from the desired outcomes rather than by equipment names. In practice, rather than insisting on a single method, the approach of dividing roles between overall documentation and detailed documentation is less likely to fail.

Step 2 Confirm the material of the target and its deterioration condition

Next, it is important to closely observe the object on which the inscription is carved. Even with the same inscription, measurement-related considerations change depending on the material—stone, metal, wood, concrete, brick, and so on. Even among stones, a rough surface and a polished one differ in how they reflect light and in how easy they are to measure. Metal plates tend to be highly reflective, and on wooden parts the grain direction or chips can be mistaken for the boundaries of characters. You need to grasp the properties of the material itself, not just the characters.

Checking the condition of deterioration is also essential. Weathering, dirt, moss and lichen attachment, residual coatings, rain streaks, efflorescence, cracks, chipping, and surface flaking or delamination can affect not only the legibility of the characters but also the stability of 3D acquisition. What should be noted here is not to touch the surface carelessly to make recording easier. Especially for inscriptions treated as objects for preservation, cleaning or wiping with water can cause changes in condition. Pre-measurement handling must always be reconciled with the principles of object conservation. If making something easier to see becomes the goal and the object is damaged as a result, it defeats the purpose.

Checking the installation environment is also extremely important in practical work. Outdoors, conditions such as the direction of sunlight, the time of day, whether surfaces are still wet after rain, whether shadows from surrounding trees will fall, whether there is pedestrian or vehicle traffic, and whether scaffolding can be secured directly affect work quality. Even indoors, factors that actually make measurements difficult include flickering lighting, maneuvering in confined spaces, display cases with strong reflections, and the distance to nearby exhibits. To avoid encountering problems for the first time on site, it is important to identify potential impediments through pre-checks or an initial observation on the day.

Also decide at this stage how much of the object's shape to include. Whether it is sufficient to photograph only the inscribed face, or whether you need the entire monument, the base, the back, the sides, and the surrounding installation context will greatly change the scope of measurement. It is not uncommon to later find yourself saying, "We also needed the names of the people who erected it on the back" or "We wish we had checked the base for settlement." When you are on site, your attention tends to focus on the characters in front of you, but because inscriptions are often part of a larger installed object, being conscious of how much to treat as a single recording unit can reduce the risk of having to re-measure.

Step 3 Prepare the reference position and measurement conditions

To stabilize the accuracy of 3D measurements, it is important to set up reference positions and conditions before the acquisition itself. By "reference" here we mean references for dimensions, position, and orientation. No matter how detailed the acquisition, if the scale is ambiguous or the handling of coordinates is unclear, it becomes difficult to use later for comparison or drafting. When recording inscriptions, one tends to focus only on the fine details of the characters, but considering downstream processes, establishing references can actually be more important.

First, what you must secure on site is a system that can guarantee the subject's measurements. If there are markers showing a known reference length or reference points for alignment, it becomes easier to accurately match the scale during post-processing. Even when capturing localized details, you should at least make clear which part of the whole was recorded. If you capture overall records and detailed records separately, you need common elements to link them. Without this, even highly detailed data will be ambiguous as to where it belonged within the whole.

For outdoor inscriptions, if you want to manage them including the surrounding terrain and installation location, capturing highly accurate on-site positional information will pay off later. For example, for monuments within historic sites, boundary markers, memorials, bridge nameplates, and inscriptions of civil engineering heritage, not only the object itself but also where it is located is important. In such cases, in addition to local dimensional references, assigning absolute position coordinates makes future re-measurement and overlay with other data easier. It is important to think of 3D data not as isolated files but as positioned within the actual site space.

Organizing measurement conditions is just as important. In photographic methods, ensuring sufficient overlap, minimizing blur, stabilizing exposure, and managing focal length all affect accuracy. Even with methods like lasers, you must be aware of the distance to the subject, the angle, occlusions, and the effects of reflections. Furthermore, when dealing with shallow relief such as inscriptions, you need to separate the lighting conditions that improve legibility from the conditions that stabilize shape capture. Even if oblique lighting is effective for reading, it is not necessarily optimal for direct shape acquisition. Planning separate records for reading assistance and for stabilizing geometric information will make the results markedly more consistent.

Organizing the on-site workflow is something you should prepare as well. Practical arrangements—such as where to approach from, whether ladders or scaffolding are required, whether the movement of worshippers or passersby will be obstructed, and which side to capture before the sun starts to set—determine the quality. 3D measurement is often associated with data processing, but if on-site preparations are poor, even the most advanced methods will result in missing data or blurring. Especially in locations that can be accessed only once, organizing the measurement conditions directly translates into the success rate.

Step 4 Preserve character shapes by capturing them without omission

In actual acquisition, simply capturing the inscribed surface once from the front is insufficient. If you want to accurately preserve the contours and depths of the characters, the way chips have formed, and the boundaries with surrounding surfaces, you need to treat the subject not as a plane but as a three-dimensional object. Even if the monument face appears flat, it actually has subtle warps and distortions, and the engraving sits on top of that. Relying solely on frontal capture tends to soften the rise of the side faces and the edges of chips, and weakens information about the depth dimension of the characters.

Therefore, captures should be made from the front, at an angle, from the side, and, when necessary, from above and below, supplying information about the grooves and edges of the characters from multiple directions. In particular, inscriptions that have been worn shallow may have only parts of their contours remaining. Such subtle undulations can suddenly appear or disappear when the viewing direction changes. Do not assume on site that “if it’s visible, it must have been captured”; it is important to obtain overlapping coverage from multiple directions.

What you should be mindful of here is separating the whole from the details. First capture the entire subject at a manageable density, then capture the text content and important areas up close at high density; doing so makes it easier to reconcile the overall context with the readability of the details. If you pursue only the details from the start, it becomes difficult later to understand positional relationships and to connect them to the whole. Conversely, if you only cover the whole, you tend to lack depth in the details. Time allocation on site is limited, but adopting a two-layer approach of whole and detail expands the possible uses of the deliverables.

For shallow incisions, the way light is used is also important. However, it should not be misunderstood that lighting that makes the inscription easy to see and conditions that make it easy to measure necessarily coincide. Oblique (raking) light is effective at bringing out the relief of characters, but overly strong shadows can obscure some information. Therefore, in practice it is useful to separate tasks: retain raking-light records to assist reading, while performing the actual shape acquisition under conditions where shadows are not too disruptive so the results are stable. What is needed is not to try to complete everything in a single capture.

Also, something that must always be done on site is checking there and then. After capture, view a quick preview to confirm that important parts of the characters are not missing, that reflections or blur have not caused information to be lost, and that no blind spots remain. This is very basic, but the busier the site, the more likely it is to be skipped. If you only find out after returning to the office and processing that the crucial year/month/day portion is faint or that the angle was insufficient and the depth does not show, it can be fatal for jobs where revisiting is difficult. For inscription measurement, incorporating on-site verification as part of the workflow is indispensable.

Step 5: Improve Readability in Data Processing

Data acquired on site is often difficult to use in practice as-is. The value of 3D measurement is realized not simply by collecting data, but only once it has been organized into a form that facilitates necessary decision-making. In the case of inscriptions, there are two main objectives of post-processing. One is to correctly reconstruct the shape, and the other is to arrange it into a representation that makes reading and comparison easy. Treating these two separately stabilizes the quality of the deliverables.

First, in shape reconstruction we align the acquired data, remove unwanted parts, clean up noise, and, when necessary, generate point clouds and meshes. At this stage, preserving the original shape is more important than making it look tidy. Over-smoothing the surface can erase the fine incisions of inscriptions and traces of weathering. Conversely, too much noise makes it difficult to distinguish the contours of characters from surface roughness. How much cleaning to perform should be determined by the intended use, not by appearance.

Furthermore, we add representations to improve legibility. For example, displays that emphasize surface height differences, simulated raking light, shadow-suppressed shape enhancement, orthographic imaging, and cross-sectional depth checks are effective in the practical work on inscriptions. Boundaries of characters that were ambiguous to the naked eye may become easier to discern when viewed as differences in height. However, the enhanced displays produced at this stage are visual aids for interpretation and are not the original data themselves. In reports and shared materials, it is important to distinguish between the original data and the enhanced displays.

Also, the task of deciphering characters and the recording of 3D shapes are separate things. Having 3D data does not necessarily make all characters readable. If wear or loss has progressed, multiple interpretations may be possible. That is why, in data processing, it is necessary to separate and retain what parts of the shape are based on direct measurement and where human interpretation begins. Editing measured results to make them too easy to read makes later verification difficult. To maintain the reliability of records, it is effective to manage the raw data, the processed data, and decipherment notes or annotations separately.

Furthermore, if comparisons are intended, you should arrange data from multiple time points according to the same criteria. When alignment standards are consistent, it becomes easier to observe surface changes and the progression of damage. For confirming before-and-after repairs, verifying the effects of cleaning or conservation measures, and assessing conditions after disasters, retaining data in a comparable state is more valuable than merely having converted it to 3D. Post-processing should be regarded not as a cosmetic finish but as a process to enhance future reusability.

Step 6 Prepare the file format and sharing method

The final step is to save the deliverables and prepare them in a form usable on site. If you take this lightly, the high-quality data you acquired may, after a few months, become impossible to find, open, or understand. 3D measurement of inscriptions does not end at the point of acquisition; it only becomes meaningful when it can be used for rechecking, comparison, sharing, and handover. To make records useful in practice, rules are needed for how data are retained.

First, what should be preserved is not only the finished 3D model. The original photographs and raw data, the processed point clouds and meshes, orthophotos, cross-sections, measurement notes, site photographs, acquisition dates, responsible personnel, object names, location information, the reference standards used, processing parameters, and so on should also be kept together. If you want to reprocess later, you cannot redo it without the original data. In particular, inscriptions—characters or traces that were not noticed at first—may well become necessary for different purposes in later years. Leaving the possibility for future review is important in recordkeeping.

It's safer not to bias toward a single file format. Relying only on formats that can be handled in specific viewing environments makes long-term preservation and handover more prone to problems. Keeping lightweight viewing data separate from original data that can be used for analysis or reprocessing makes it easier to choose what to share depending on the recipient. Field staff, managers, researchers, and designers need different levels of granularity. Preparing deliverables while considering who will receive what reduces both excess and shortfall.

When sharing, it is important not to provide only the 3D data. Without at least minimal explanations—such as the object name, installation location, acquisition range, acquisition date and time, orientation, reference position, cautions, and annotations for hard-to-read parts—a third party cannot handle it correctly. Even if you think it will be obvious from looking at it, you yourself may not understand it months later. Because records of inscriptions depend on contextual information, you should make a point of keeping the dataset and its explanatory information together.

Moreover, for outdoor inscriptions and facility-management assets, there is high value in storing them linked to location information. Even if an object has a 3D shape, if its exact on-site location is ambiguous, revisits, inspection planning, and organizing relationships with surrounding equipment become more burdensome. Conversely, when the location is clear, asset register management, routine inspections, integration with repair histories, and coordination with surrounding maintenance can proceed more smoothly. To prevent 3D survey results from ending up as standalone records and to bring them into the maintenance workflow, an approach that connects location with the deliverables is indispensable.

Common Pitfalls in 3D Measurement of Inscriptions

A common mistake in 3D measurement of inscriptions is focusing so much on selecting equipment and methods that deciding what you want to preserve is put off. If your goal is to read the characters but you end up capturing only the overall shape neatly, or conversely you capture only fine details at high resolution and lose their overall position, the deliverables will have only limited uses. 3D measurement is not a panacea; it becomes effective only when designed to match the intended purpose.

Another common mistake is stopping after capturing only the center of the front. Because inscriptions are written on the face of the monument, people tend to rely on information from directly in front, but the depth of the characters, the shapes of chips, and surface distortions cannot be understood without viewing them from multiple directions. The shallower the engravings, the more direction-dependent they are, and if you rely solely on the impression you had on site, you will encounter insufficient information during post-processing. Rather than approaching it as if photographing a flat plane, you need to capture it with a sense of enveloping the three-dimensional form.

There are also failures that neglect the raw data. If you leave only the easy-to-read outputs after processing and do not organize the original data, you will not be able to respond when you want to re-examine them later from a different perspective. Inscriptions can hinge on a single character of a date or a person’s name. Keeping the raw data and the processing history so that the basis for readings can be rechecked is also extremely important for reliability.

Furthermore, insufficient consideration of site conditions cannot be overlooked. Wet surfaces, strong reflections, deep shadows, impeding passage, and inadequate safety measures can lead not only to reduced quality but also to on-site incidents. If the object is to be preserved, it is necessary to balance protecting the subject and measurement efficiency. You must avoid excessively altering surfaces to make recording easier or approaching too closely. In practice, a balance is required between capturing good records and not taking undue risks.

Finally, cases where 3D data is created without a plan for its use also lead to failure. If the storage location is ambiguous, sharing methods are undecided, it is not linked to location information, or it has not been incorporated into reporting materials, it will not be reused in the field. The 3D measurement of inscriptions is evaluated not at the point of creation but on whether it could be used later. That is why it is necessary to consider storage and sharing from before acquisition.

Summary

In 3D measurement of inscriptions, what matters more than minor differences in equipment is structuring the workflow according to the purpose of the record. First determine why it is being preserved, assess the material and the state of deterioration of the subject, set reference positions and measurement conditions, then capture the overall form and the details without omission, process the data into a form that makes reading and comparison easy, and finally store it in a reusable format. By following these six steps, you will create a 3D record that is useful in practice, not just visually.

When recording inscriptions, attention tends to focus on the legibility of the characters, but the real value lies in being able to treat the characters, shape, damage, installation location, and relationship with the surroundings together. This is especially true for outdoor objects such as stone monuments, memorials, boundary markers, directional signs, and plaques on civil engineering structures, where positional information itself is important; capturing not only shape data but also precisely where an object is located greatly affects the ease of maintenance and re‑surveying.

If you want 3D measurement of inscriptions to be more than mere digitization and to serve as records usable on site, it is important to consider how to handle location information as well. On sites where you want to carry out everything from simple surveying of outdoor targets to high-precision recording of installation positions in an integrated workflow, using the LRTK function of an iPhone-mounted high-precision GNSS positioning device makes it easier to link the shape records of the inscription itself with accurate on-site position management. If you are looking ahead to preservation, comparison, registry maintenance, and understanding relationships with surrounding structures, incorporating operations that utilize LRTK in addition to 3D measurement makes it easier to balance recording accuracy and practical usability.