Development of a Lightweight UAV-based Photogrammetric System for Rapid 3D Civil Infrastructure Mapping and Change Detection in Post-Disaster Scenarios

Project Abstract

This study presents the development and evaluation of a lightweight UAV-based photogrammetric system designed for rapid 3D civil infrastructure mapping and change detection in post-disaster scenarios. The system integrates a compact multi-sensor payload, including a high-resolution consumer-grade camera, an inertial measurement unit, a lightweight GNSS receiver, and an onboard computer, to deliver centimeter-level 3D models with minimal flight weight and power consumption. A modular workflow was implemented to streamline data acquisition, processing, and analysis in austere post-disaster environments. The photogrammetric pipeline leverages dense image matching, multi-view stereo, and structure-from-motion algorithms optimized for low-overhead hardware, enabling near-real-time 3D reconstruction. A robust calibration and georeferencing framework was developed to mitigate lens distortion, GPS-denied operation, and atmospheric effects, ensuring consistent multi-temporal datasets for change detection. The system supports rapid mission planning through autonomous flight path generation, safety-aware obstacle avoidance, and offline/backhaul data synchronization to field-deployable servers. To address the exigent needs of disaster response, we introduce a novel change-detection methodology that combines voxel-based volumetric differencing, surface deformability metrics, and semantic segmentation to classify structural damage, material loss, and subsidence. The methodology is validated across a spectrum of post-disaster scenarios, including earthquakes and floods, using both synthetic and real-world datasets. Ground-truth data from terrestrial surveys and high-precision TLS are employed to quantify positional accuracy, completeness, and thematic correctness of the generated point clouds and 3D meshes. Results demonstrate that the lightweight system achieves centimeter-level elevation accuracy (with horizontal accuracy within a similar order) at flight heights suitable for rapid deployment, while delivering complete 3D models of large civil structures within minutes to hours depending on site complexity. Change-detection performance shows robust identification of critical damage patterns such as cracking, spalling, deformations, and collapse risk indicators, with statistically significant improvements over conventional image-based assessments. The study also discusses operational considerations, including flight authorization, battery management, data integrity under compromised communication links, and the ethical implications of rapid-mapping in disaster zones. A cost-performance analysis compares the proposed system against larger, heavier UAV platforms and traditional ground-based surveying methods, highlighting reductions in deployment time, PPE requirements, and data-interpretation workload. The research concludes with guidelines for field deployment, recommended refinements to sensor fusion and processing pipelines, and prospects for integration with command-and-control systems for disaster management, urban resilience planning, and post-disaster reconstruction prioritization.

Project Overview

What This Project Is About

---

A plain-language overview of the topic and what the project investigates.

---

---

The Problem It Addresses

---

What problem or gap this project tackles and why it matters to the field or society.

---

---

Objectives of the Project

---

1. Understand how lightweight drones can collect 3D data quickly after disasters.
2. Build a simple photogrammetry workflow to turn photos into 3D models.
3. Detect changes in civil infrastructure over time using updated 3D models.
4. Evaluate system performance in terms of accuracy, speed, and usability.
5. Provide practical guidelines for field deployment in emergency settings.

---

---

What You Will Do Step by Step

---

1. Learn basic drone operation and data collection procedures.

2. Plan missions to capture essential views of structures after a disaster.

3. Process images to create 3D reconstructions and identify changes.

4. Compare new models with baseline data to quantify damage and recovery needs.

5. Assess the strengths and limitations of the lightweight system in real scenarios.

---

---

Expected Outcome

A ready-to-use, lightweight drone-based workflow that produces accurate 3D models and change detection results, suitable for rapid assessment and decision-making after disasters.