🏙️ Point Cloud Processing And Spatial Analysis In Enschede, The Netherlands

🚨Task Force

Demonstrate proficiency in processing Airborne Laser Scanning (ALS) and aerial imagery point cloud data. Applying filtering techniques and quality checks to generate Digital Terrain Models (DTMs) and Digital Surface Models (DSMs). The derived data is then applied for flood simulation and optimising surveillance camera placement within Enschede, the Netherlands.

🎯Aims

Propose a workflow in applying photogrammetry, aerial laser scanning, and point cloud processing techniques to generate DEMs for the applications to simulate floods and make informed decisions on data acquisition.

1. To apply advanced filtering and cleaning techniques for preparing ALS and aerial images point cloud data.

2. To perform rigorous quality checks, including RMSE computation between both point clouds and internal quality checks.

3. To simulate flood events for identifying flood-prone areas.

4. To optimise camera positions with consideration of visibility and coverage factors.

🛠️ Approach

• Cloth Simulation Filtering (CSF) and Progressive TIN Densification (PTD) applied to ALS point cloud data to obtain detailed terrain analysis, including DTM, Hillshade, Slope, Aspect, etc.

• Triangulation performed on data generated from UAV aerial imagery to derive orthomosaics and DEMs.

• Quality checks to assess DEMs qualitatively (visual control, hillshade) and quantitatively (histogram, standard & robust accuracy metrics).

• Flood simulation to identify areas at risk of flooding for improved flood risk management.

• Viewshed analysis to optimise camera placement with visibility, coverage, and strategic placement.

✅Success

• Point Cloud DEM has a more accurate terrain representation.

• Textureless elevation models are supplemented by orthomosaic.

• Viewshed analysis optimises visible surfaces from an observer point(s) over a DEM.

• Blender is powerful and versatile for 3D modelling, animation and rendering. A school of motion with a complex interface and steep learning curve for beginners.

🪞 Reflection

• Point cloud filtering and parameter threshold poses the greatest challenge for generating DTMs ➡️ Radius outlier filter (ROL), Statistical analysis theory.

• Flood simulation 3D visualisation integrating DSM, Orthomosaic and building model ➡️ Digital twin modelling.

• Application of visibility analysis and digital elevation model ➡️ Spatial planning, Anticipatory action.