⚡ electrical engineering

Assessment of Geometric Deviations in a Complex Concrete Dome Using UAV Photogrammetry and Integrated Geodetic Techniques

This study evaluates the geometric deviations of a complex concrete dome at St. Mary and St. Joseph Church in Alexandria, Egypt, by comparing UAV photogrammetry, laser scanning, and total station surveying, ultimately demonstrating that UAV and laser scanning are effective for detecting irregularities with deviations up to 45 mm.

Original authors: RAMY K. Kelliny, Ashraf A. A. Beshr, magdy Israil, Magda H. Farahan

Published 2026-07-23

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: RAMY K. Kelliny, Ashraf A. A. Beshr, magdy Israil, Magda H. Farahan

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are an architect trying to build a perfect, giant eggshell out of concrete. You want it to be smooth, symmetrical, and strong enough to hold up a roof over a huge space without any pillars in the middle. This is the job of a "dome." For centuries, humans have built these structures for mosques, churches, and museums because they are beautiful and efficient. But here's the tricky part: concrete is heavy, and over time, or even just during construction, these giant shells can get a little bit crooked. They might lean to one side, bulge out, or shrink in a weird spot. If you don't catch these tiny mistakes early, the whole thing could become unsafe.

To find these hidden flaws, scientists used to have to climb up the structure with tape measures and rulers, checking one tiny spot at a time. It was slow, and you might miss a problem hiding between your measurement points. But now, we have a new set of super-powered eyes. Think of them as digital detectives. One is a "Terrestrial Laser Scanner" (TLS), which is like a super-fast, high-tech flashlight that shoots millions of invisible laser beams to map every bump and dip of a building in 3D. The other is a "Drone" (UAV), a flying robot camera that takes hundreds of overlapping photos to stitch together a perfect 3D picture of the building from the sky. By combining these two high-tech methods with old-school surveying tools, engineers can now see the "shape" of a building with incredible precision, spotting errors as small as a few millimeters—about the width of a pencil eraser.

This paper tells the story of a team of engineers who used these digital detectives to inspect a very specific, complex concrete dome: the one sitting on top of the St. Mary and St. Joseph Church in Alexandria, Egypt. Built in 1949, this dome is a historical treasure, but no one had checked its shape since it was built. The researchers wanted to know: Is this 70-year-old dome still holding its perfect curve, or has it started to warp? They didn't just guess; they used a laser scanner, a drone, and a traditional total station (a high-precision theodolite) to measure the dome from every angle.

The team found that the dome wasn't perfectly perfect. While the bottom part of the dome was built with amazing accuracy, almost like a flawless circle, things started to get a little wobbly as they went higher up. By the time they reached the top, the dome had drifted away from its ideal shape by about 45 millimeters along its central axis. That's roughly the thickness of two stacked coins. When they looked at the horizontal rings of the dome, they found deviations of up to 30 millimeters, meaning some parts were pushed out or pulled in compared to the original blueprints. The total "wobble" of the surface was about 25 millimeters.

Interestingly, the researchers discovered that these mistakes weren't random. The dome seemed to lean and twist in specific directions, with the biggest "push out" happening at one side and the biggest "pull in" at another. It turns out that building a dome is like stacking a tower of Jenga blocks: the lower levels are easy to control, but as you go higher, it gets harder to keep the formwork (the mold for the concrete) perfectly aligned, and small mistakes add up. The study confirmed that the laser scanner and the drone were both excellent at finding these problems, and when they compared their results, they matched up almost perfectly. This gives engineers a lot of confidence that these high-tech tools can be trusted to keep our historic buildings safe. The paper concludes that while the dome is still standing strong, it has some geometric "scars" from its construction, and using these advanced scanning methods is the best way to monitor such structures in the future to ensure they don't get any worse.

Technical Summary: Assessment of Geometric Deviations in a Complex Concrete Dome Using UAV Photogrammetry and Integrated Geodetic Techniques

Problem Statement
Monitoring the structural health of complex architectural geometries, particularly historical domes, presents significant challenges. These structures, such as the concrete dome of St. Mary and St. Joseph Church in Alexandria, Egypt (built in 1949), are susceptible to geometric deviations caused by construction imperfections, material changes, and environmental factors like coastal humidity. Traditional single-point monitoring methods often fail to capture the dense, continuous surface data required to detect subtle deformations in non-standard shapes. There is a need for non-contact, high-precision techniques capable of assessing both horizontal and vertical deviations across the entire surface of such structures to ensure their preservation and safety.

Methodology
The study employed an integrated approach utilizing three distinct geodetic and photogrammetric techniques to assess the geometric deviations of the dome:

Terrestrial Laser Scanning (TLS): A TOPCON GLT-1003 scanner was used to acquire dense 3D point clouds from three different stations. The data was processed using MAGNET Collage software, utilizing Ground Control Points (GCPs) established via a Total Station (TS) for georeferencing.
UAV Photogrammetry: A DJI Air 3 Fly drone captured high-resolution imagery (172 photos with >70% overlap) using an orbit flight pattern. The images were processed using PhotoModeler software, which employed Direct Linear Transformation (DLT) for camera calibration and sub-pixel point extraction to generate a 3D photorealistic model.
Total Station (TS) Surveying: A Sokkia reflectorless TS served as the reference instrument to establish a local coordinate system, measure control points, and validate the accuracy of the TLS and UAV data.

Mathematical and Analytical Framework
The researchers developed a geometric modeling framework to analyze the dome's deformation:

Surface Representation: The dome was modeled as a sphere using the least squares adjustment technique to minimize residuals between observed points and the theoretical surface.
Sectional Analysis: Horizontal cross-sections were analyzed to determine the radius and center coordinates at various heights.
Deformation Indices: The study calculated axis deviations ( $\Delta X, \Delta Y$ ), tilt angles ( $\beta$ ), and radial/vertical deviations. A specific "deformation index" ( $\Delta i$ ) was formulated to measure inclination deviations relative to the ideal curvature trend.
Validation: The accuracy of the TLS was tested against the TS at varying distances (5.246 m to 13.066 m) and resolutions (3.1 mm to 12.5 mm). The UAV camera was calibrated using a 36x36 inch grid to correct lens distortions.

Key Results

Accuracy Validation: The study confirmed that TLS accuracy is highly dependent on distance and resolution. The highest precision (minimum error of ~2 mm) was achieved at the closest distance (5.246 m) with the finest resolution (3.1 mm). Errors increased significantly at longer distances (up to 31 mm at 13.066 m with 12.5 mm resolution).
Geometric Deviations:
- Horizontal Axis: A horizontal deviation of approximately 45 mm was detected along the central axis of the dome sections.
- Perimeter Deviations: Comparisons between the actual section perimeters and the ideal geometric model revealed deviations reaching up to 30 mm on the horizontal plane.
- Surface Variation: The combined effect of these deviations resulted in an overall surface variation of approximately 25 mm.
- Vertical Trends: The lower sections of the dome (up to +2.25 m) exhibited near-perfect circularity and stability. Deviations began to accumulate above +2.50 m, increasing progressively with height. The upper quarter of the dome showed the most significant irregularities, with cumulative vertical distortions reaching up to 100 mm at the apex.
Angular Anomalies: The analysis identified specific angular turning points (30°, 75°, 120°, 180°, 220°, 270°, 310°) where localized deformations occurred. The largest negative deviation (-36 mm) was found at 180°, while the largest positive deviation (+27 mm) occurred at 270°.
Method Consistency: The UAV photogrammetry results showed high consistency with the TLS data, validating the effectiveness of using drone-based close-range photogrammetry for detecting geometric irregularities in complex structures.

Significance and Claims
The paper claims that the integration of UAV close-range photogrammetry and Terrestrial Laser Scanning provides a robust, effective methodology for detecting geometric irregularities in spherical domes, offering a viable alternative or complement to traditional surveying techniques. The study demonstrates that:

Non-Contact Efficiency: These non-contact methods are essential for monitoring unstable or historical structures without physical interference.
High-Resolution Detection: The combined approach can detect millimeter-level deviations, providing a comprehensive understanding of structural behavior that single-point monitoring cannot achieve.
Construction Quality Insight: The observed pattern of increasing deviation with height suggests that construction precision (formwork, reinforcement placement, and concrete casting) decreases at higher elevations, highlighting the need for stricter quality control in upper structural levels.
Validation of Photogrammetry: The study confirms that UAV-based photogrammetry, when properly calibrated and processed, yields results consistent with high-precision TLS, making it a valuable tool for structural health monitoring and heritage conservation.

The authors conclude that while the current study focused on exterior scanning, future work should incorporate interior scanning and non-destructive testing (e.g., Ground-Penetrating Radar) to assess concrete thickness and reinforcement conditions, thereby enabling predictive modeling for long-term structural health monitoring.

More like this