Automated UAV-based LiDAR and Photogrammetry Fusion for Precision Cothic River Basin Monitoring and 3D Cadastre Generation

Project Abstract

Automated UAV-based LiDAR and Photogrammetry Fusion for Precision Cothic River Basin Monitoring and 3D Cadastre Generation presents a novel integration of airborne LiDAR, high-resolution photogrammetry, and advanced data fusion techniques to deliver high-precision, temporally consistent 3D representations of complex riverine environments. The study addresses critical challenges in flood risk assessment, sediment transport analysis, land tenure monitoring, and ecohydrological modeling by generating accurate 3D cadastral models and detailed morphometric datasets for the Cothic River Basin. A multi-sensor workflow is implemented, beginning with UAV-mounted LiDAR scanning to capture canopy structure, shorelines, banks, and channel thalwegs under varying hydrological conditions. Concurrent high-resolution RGB imagery is acquired to augment textures and support semantic classification. A robust calibration and co-registration framework ensures centimeter-level spatial alignment between LiDAR point clouds and photogrammetric reconstructions, enabling seamless fusion of geometric and spectral information. The core methodology employs a Bayesian fusion architecture and deep learning-assisted point cloud segmentation to produce unified 3D surface models and vectorized cadastral features, including parcel boundaries, riverbank delineations, building footprints, utility corridors, and linear infrastructure along floodplains. Temporal datasets are captured across multiple seasons and flow regimes to quantify morphodynamic changes, vegetation dynamics, and sediment deposition patterns, with change detection implemented through multi-temporal point cloud differencing and surface-based metrics. The research introduces an enhanced 3D cadastre generation pipeline that integrates legal boundary metadata, elevation/landform attributes, and ownership records into an interoperable geospatial database, enabling precise parcel management and rapid legal adjudication in riverine zones. Evaluation is conducted against terrestrial LiDAR surveys, traditional photogrammetric models, and ground-truth measurements from fixed reference targets to assess accuracy, completeness, and reliability. Results demonstrate substantial improvements in vertical accuracy (sub-15 cm RMSZ in exposed zones and sub-25 cm in dense vegetation) and horizontal precision (sub-10 cm RMSX/Y) relative to baseline photogrammetry alone, particularly in shadowed or occluded river corridors where LiDAR contributes essential structural details. The fused models significantly enhance floodplain mapping, habitat assessment, and infrastructure resilience planning by providing actionable 3D city- and river-scape representations with attribution-rich cadastre layers. Computational performance analyses reveal scalable processing times through parallelization and cloud-based storage, supporting near-real-time monitoring capabilities for emergency response. Sensitivity analyses explore the impacts of flight altitude, point density, and texture resolution on fusion quality, while uncertainty quantification identifies optimal sensor configurations for diverse basin conditions. The study concludes with a set of best-practice guidelines for UAV-based LiDAR and photogrammetry campaigns in hydrologically dynamic environments and outlines future enhancements, including real-time data streaming, integration with hydroinformatic models, and policy-oriented cadastre modernization to support sustainable river basin management.

Project Overview

What This Project Is About

---

A beginner-friendly look at using drones to collect landscape data and combine two kinds of scans to map a river basin in 3D. The project explores how to measure terrain, water features, and land parcels accurately for better planning and records.

---

---

The Problem It Addresses

---

---

---

Objectives of the Project

---

1. Learn the basics of drone data collection and explain why accuracy matters.
2. Understand how LiDAR and photogrammetry work and what each adds to a map.
3. Develop a simple workflow to fuse LiDAR point clouds with 3D photos.
4. Create a 3D model and a basic cadastral record for a river basin area.
5. Assess data quality and identify common errors to avoid.

---

---

What You Will Do Step by Step

---

1) Review basics of UAV surveying and data types (LiDAR and photos). 2) Plan a small field data collection over a test basin. 3) Acquire LiDAR and photo data with a drone. 4) Process data to generate 3D surfaces. 5) Merge datasets into a unified model. 6) Create simple parcel boundaries and water features. 7) Check results against reference data. 8) Present findings and discuss limitations.

---

---

Expected Outcome

A clear, shareable 3D model of the river basin, including terrain, water features, and parcel boundaries, with a simple workflow that others can repeat for similar areas. The project should show improvements in accuracy over single-method approaches and provide a practical basis for small-scale cadastral generation and environmental planning.