What should the cross-sectional slice thickness be in mm (in)? Explaining optimal values for point cloud cross-sections with examples

Table of Contents

• Introduction

• What is slice thickness

• Examples of terrain measurement

• Examples of building measurement

• Examples of bridge measurement

• Examples of road pavement surfaces

• Procedure for determining slice thickness

• Relationship between slice thickness and point cloud density

• Mixing point clouds of different densities

• Relationship between noise and slice thickness

• Dealing with complex shapes

• Recording and standardization of parameters

• Slice thickness and final quality

• Best practices for recording, managing, and standardizing slice thickness

• Impact of changes in the measurement environment on slice thickness

• Total management of slice thickness

Introduction

One of the most important setting parameters when creating cross-sections from point clouds is the "slice thickness." Slice thickness is the value that specifies how far to the left and right of the measurement line point cloud data should be included in that section. If this value is too small, the included data will be insufficient and the cross-section line will become jagged. If it is too large, information that should not be included will be mixed in, preventing an accurate representation of the actual terrain.

However, there is no clear answer to the question, "What is the optimal slice thickness in millimeters?" This is because the optimal value depends on multiple factors, such as the characteristics of the object being measured, the density of the point cloud, and the required level of accuracy. In this article, we explain these factors in detail and provide recommended values for various practical scenarios, along with concrete examples.

What is slice thickness

It is important to first understand the concept of slice thickness accurately. Slice thickness is a parameter that defines how much width to allow around the measurement line as the central axis. For example, if the slice thickness is 1 m (3.3 ft), all points within 0.5 m (1.6 ft) to the left and right of the measurement line (a band-shaped region totaling 1 m (3.3 ft)) will be extracted.

Slice thickness is not merely a mathematical parameter; it defines the "representativeness" of a real measurement target. If the road width is 5 m (16.4 ft), a slice thickness of 1 m (3.3 ft) will not cover the entire road. Conversely, if set to a slice thickness of 10 m (32.8 ft), adjacent terrain will also be included. The optimal slice thickness should be a parameter determined by both the geometric characteristics of the target and the measurement objectives.

Determining slice thickness follows the same fundamental principles as sample survey design. The essence of choosing slice thickness is the problem of how to extract a representative sample from the population. The key to determining an optimal slice thickness is balancing the conflicting demands that slices that are too thick compromise representativeness, while slices that are too thin reduce statistical reliability.

The slice thickness setting can be understood as a statistical sampling problem. It is rare for data points along a measurement line to match exactly; typically, the point cloud is dispersed around the measurement line. Slice thickness determines how much of this dispersion range is allowed.

Slice thickness also has a physical meaning. For example, in road measurement, slice thickness means "how much of the road's width is included in a cross section." If the pavement is not perfectly straight and has slight gradients or curves, a small slice thickness will not accurately capture the actual shape of the road surface.

Examples of Topographic Surveying

This explains examples of selecting slice thickness for measurements of natural terrain and road topography.

For measurements of flat terrain, a slice thickness of 0.3–0.5 m (1.0–1.6 ft) is common. With this thickness, even at a typical point cloud density (10–50 points per square meter (10.8 ft²)), sufficient data points can be obtained. On flat terrain, because vertical variation is small, even if the slice thickness is set smaller, the cross-section lines remain relatively smooth.

In hilly and mountainous terrain with undulating topography, it is recommended to increase the slice thickness to 0.5-1.0 m (1.6-3.3 ft). More data points are required to accurately represent complex undulations. Increasing the slice thickness increases the number of valid data points per cross section, yielding statistically more reliable cross-section lines.

In river measurements, slice thickness tends to be relatively large because it must include both the water surface and the riverbed. Typically, 1.0–2.0 m (3.3–6.6 ft) is used. A certain thickness is necessary to represent river meanders and complex topography.

Examples of Building Measurement

Measurements of buildings and structures require considerations different from those for topographic surveying.

When measuring the exterior walls of a building, slice thickness is generally 0.1–0.3 meters (0.3–1.0 ft). The surface of a building is usually relatively flat and regular in shape, so a small slice thickness is sufficient. This thickness allows subtle irregularities and damage in the wall to be captured in detail.

For buildings with complex shapes (for example, buildings with multiple protruding sections), a slice thickness of 0.2–0.5 m (0.7–1.6 ft) is appropriate. In complex areas, more detailed information is required, so setting a smaller slice thickness makes it possible to accurately capture the shape.

Slice thickness of 0.3-0.5 m (1.0-1.6 ft) is recommended for measurements of interior spaces (corridors, rooms, etc.). Cross-sectional drawings of interiors are used directly for architectural design information and renovation planning, so a certain level of accuracy is required.

Examples of Bridge Measurements

The measurement of bridge structures requires careful selection of slice thickness due to high accuracy requirements.

In measurements to ascertain the thickness of the bridge deck and the condition of surface deterioration, slice thicknesses of 0.1-0.2 m (0.3-0.7 ft) are used. The bridge deck surface is relatively flat, but because fine damage and settlement must be detected, the slice thickness is kept small.

For measurements of bridge sides and shoring, a slice thickness of 0.2–0.5 m (0.7–1.6 ft) is appropriate. A moderate thickness is necessary to accurately represent complex arrangements of members.

In measurements intended to capture the entire bridge’s three-dimensional shape, slice thicknesses of 0.5-1.0 m (1.6-3.3 ft) may also be used. In such cases, understanding the overall structure is prioritized over fine details.

Examples of road pavement surfaces

In measurements for diagnosing the condition of road pavement surfaces, the setting of slice thickness is particularly important.

For standard pavement surface measurements, a slice thickness of 0.5-0.7 m (1.6-2.3 ft) is standard. This thickness allows appropriate capture of longitudinal surface undulations (waviness, cracking). If the slice thickness is too small, the influence of measurement noise increases and the profile lines tend to become jagged.

In quality control of newly constructed pavement, setting a relatively small slice thickness of 0.3～0.5 m (1.0～1.6 ft) is effective for detecting finer irregularities.

When measuring large-scale pavement settlement or surface steps, increasing the slice thickness to 1.0–1.5 m (3.3–4.9 ft) allows you to more clearly grasp the overall trend of deformation.

Procedure for determining slice thickness

I will describe a systematic procedure for determining the optimal slice thickness in a real project.

First, check the point cloud density. Calculate the number of points per 1 square meter (10.8 ft^2), or the average point spacing. The higher the density, the smaller you can set the slice thickness.

Next, evaluate the shape complexity of the object to be measured. If complex shapes or fine variations are important, set the slice thickness smaller. If you prioritize capturing overall trends, set it larger.

After that, cross-sectional diagrams are created experimentally using multiple slice thicknesses. Typically, experiments are conducted with 3–5 different values (for example, 0.2 m (0.7 ft), 0.5 m (1.6 ft), 1.0 m (3.3 ft), 1.5 m (4.9 ft)).

We visually inspect each test result and evaluate them from the following perspectives. We compare each result in terms of the smoothness of cross-section lines, the completeness of the data, and the degree to which important shape features are represented.

Finally, choose the slice thickness that best meets the requirements. If multiple candidates have equivalent quality, choosing a smaller slice thickness will preserve more detailed information.

Relationship Between Slice Thickness and Point Cloud Density

Quantitatively understanding the relationship between slice thickness and point cloud density is important for determining the optimal value.

The expected number of points contained in each cross-section can be estimated by the following simple formula: Expected number of points = point cloud density (points/m² (points/ft²)) × slice thickness (m (ft)) × cross-section length (m (ft)).

For example, if the point cloud density is 30 points/m² (2.8 points/ft²), the slice thickness is 0.5 m (1.6 ft), and the measurement line length is 100 m (328.1 ft), the expected number of points is approximately 1500. With this number of points, you can generate statistically reliable cross-section lines.

Generally, each cross-section should ideally contain at least 100–200 points. Fewer than this will reduce the reliability of the cross-section line. If the point cloud density is low (for example, 5 points / m² (0.5 points / ft²)), it is necessary to increase the slice thickness to ensure a sufficient number of valid points.

When mixing point clouds of different densities

When integrating data acquired from multiple measurement sessions, point cloud density may vary. Describe the considerations for determining slice thickness in this case.

It is recommended to determine the slice thickness based on the dataset with the lowest density. This ensures that even low-density regions contain a sufficient number of points.

Alternatively, it is possible to use different slice thicknesses for each region. Use smaller slice thicknesses in high-density regions and larger slice thicknesses in low-density regions. This method allows you to achieve uniform quality across all regions.

However, when using multiple slice thicknesses, it is important to note that the appearance and characteristics of cross-sectional images may differ by region. If necessary, homogenization processing of the results should be performed.

Relationship Between Noise Effects and Slice Thickness

Noise in point cloud data also affects the choice of slice thickness.

The smaller the slice thickness, the greater the relative impact of noise. If the number of points contained within a slice thickness is small, a single outlier (noise) can significantly change the statistical results.

Conversely, increasing the slice thickness can relatively reduce the impact of noise. Because many points are included, outliers are statistically averaged.

Therefore, with noisy measurement data, setting a larger slice thickness yields smoother and more reliable cross-sectional lines.

However, combining noise processing and slice thickness adjustment is the most effective approach. By applying a noise-reduction filter beforehand and then setting an appropriate slice thickness, optimal results can be achieved in terms of both quality and accuracy.

Handling Complex Shapes

When measuring complex geometric shapes, a single slice thickness may not be sufficient.

For example, in V-shaped valleys or complex curved shapes, if the slice thickness is too small there will be insufficient data, and if it is too large different terrain will be mixed. In such cases, it is effective to locally adjust the slice thickness according to the characteristics of the shape. Some software offers the ability to change settings on a per-segment basis, making it easier to handle complex terrain.

In addition, it is possible to divide the measurement line into multiple short segments and use different slice thicknesses for each segment. Use smaller slice thicknesses in areas of higher complexity and larger slice thicknesses in simpler areas. These local adjustments allow you to ensure quality in complex parts without compromising overall efficiency.

Development of automatic adjustment functions that leverage machine learning and AI is also advancing. These technologies could enable systems that automatically recognize the complexity of data and set the optimal slice thickness. Such automation would free users from complex parameter tuning and allow them to spend more time on creative tasks.

Recording and standardization of parameters

It is important to record all processing parameters, including the determined slice thickness.

Please include the following information in the project document: slice thickness (m (ft) or mm (in)), point cloud density (points/m² (points/ft²)), software used and version, processing date and time, author name, etc.

This information is important for maintaining consistency if additional work on the same project becomes necessary later. It also facilitates quality verification by other stakeholders.

When carrying out multiple projects, it is efficient to standardize the recommended slice thickness for each type of measurement target. Using the same slice thickness for the same type of measurement ensures consistent quality.

Slice Thickness and Final Quality

Slice thickness is the most important parameter that directly affects the quality of the final cross-sectional image.

When an appropriate slice thickness is selected, cross-sectional images with the following characteristics are obtained: smooth, natural contours, accurate reflection of the trends in the original data, clear representation of the shape of the measured object, and statistically reliable results.

If the slice thickness is inappropriate, problems such as jagged lines, scattered data, loss of important information, or the mixing in of unnecessary information can occur.

Spending time optimizing slice thickness is the most efficient investment to ensure the quality of the final deliverable. Combining this with high-precision positioning devices also aids in determining slice thickness. For example, by using an iPhone-mounted GNSS high-precision positioning device (LRTK) to measure multiple reference points, you can precisely define the positions of survey lines. Accurate definition of the survey lines makes processing results at the optimal slice thickness more reliable.

By combining a high-precision positioning device with an appropriate slice thickness setting, you can create precise cross-sectional diagrams with extremely high practical value.

Best Practices for Recording, Managing, and Standardizing Slice Thickness

Recording the slice thicknesses used in a project and the rationale for them is critically important from a quality management perspective. Keeping such records clarifies the basis for later review of results and enables rapid configuration decisions in similar projects.

The information to be recorded should include not only the value of the slice thickness used but also the rationale for selecting that value. For example, record the numbers together with their rationale, such as “Because the point cloud density was 5 points per square meter, a slice thickness of 1.0 m (3.3 ft) was chosen.” This allows for a quick assessment of the appropriateness of the settings when reviewing them later or when other team members refer to them.

Standardization within the organization should also be actively pursued. For similar measurement targets (road surveys, bridge surveys, building surveys, etc.), defining and sharing a standard slice thickness can help reduce variability in quality among operators. However, the standard value should be regarded only as a starting point and should be flexibly adjusted according to the actual point cloud density and measurement conditions.

Impact of Changes in Measurement Environment on Slice Thickness

Even for the same measurement target, point cloud density changes depending on the environmental conditions during measurement, so the optimal slice thickness also varies. By understanding these factors that cause variability, you can respond appropriately to environmental changes.

The impact of weather is particularly significant. Laser reflection characteristics differ greatly between clear conditions and rainy or foggy conditions, resulting in differences in point cloud density. In environments with rain or fog, scattering increases and the number of valid data points decreases, so the slice thickness needs to be increased. Conversely, under direct sunlight on clear days, reflections from certain materials can become excessively strong, which can also increase noise.

Measurement altitude and distance are also important factors. In drone surveys, point cloud density changes with flight altitude. Lower altitudes result in higher density, while higher altitudes result in lower density. When combining data from the same site collected at multiple altitudes, adjust the slice thickness to accommodate the lower-density regions, or use tools that can set adaptive slice thickness based on density.

Differences in the performance of measurement instruments must also be considered. When the same site is measured with different instruments, variations in beam divergence angle and measurement point spacing can result in different densities. In particular, when integrating data from older and newer instruments, it is important to understand each instrument’s density characteristics and set the slice thickness accordingly.

LRTK (an iPhone-mounted GNSS high-precision positioning device) is also effective for resetting reference points when measurement environments change. If re-surveying is required due to environmental changes, quickly acquiring high-precision reference coordinates with LRTK enables efficient reprocessing while maintaining the integration accuracy of new and old data. Establishing this kind of technical foundation is indispensable for maintaining high-quality cross-sections while flexibly responding to changes in measurement conditions.

Total Management of Slice Thickness

The setting of slice thickness is not merely a local parameter for the creation of cross-sectional drawings; it is an important design element that affects the entire measurement plan. By working backwards from the pre-measurement stage to determine how much point cloud density is required and thereby deciding scan conditions, setting an appropriate slice thickness during processing, and performing quality verification after output—being mindful of this cycle—quality control across the entire sequence of measurement tasks is improved.

Combining the establishment of high-precision control points using LRTK (an iPhone-mounted GNSS high-precision positioning device) with appropriate slice thickness management maintains high accuracy across the entire workflow from measurement to cross-section generation. A high-quality measurement foundation and appropriate parameter settings complement each other and maximize the reliability of the final deliverables.

The quality of point cloud processing can be greatly improved by understanding and properly managing various parameters, including slice thickness. Accumulating ongoing experience and deepening technical knowledge cultivates practical capabilities to handle a variety of measurement conditions.

Optimization of slice thickness is not a one-time task; it is a dynamic parameter that should be continually reviewed in response to changes in measurement conditions. Consistent management of settings with an emphasis on quality maintains the reliability of cross-sectional views.