🛰️ Point Cloud Ingestion (LAZ / LAS)

MIRROR supports ingestion, querying, and experimentation with LiDAR-based point cloud data using an integrated Jupyter Notebook environment powered by PDAL and PostgreSQL.

This enables users to work with .laz / .las files interactively, visualize metadata, and validate storage inside the Postgres database using spatial extensions.

✅ What's Included in This Feature

📦 A custom Jupyter container (`citymap-jupyter-pdal`) with:

PDAL
PostgreSQL client libraries
Notebook support

🧠 PostgreSQL is extended with:

pointcloud
pointcloud_postgis
🔗 Full integration with your MIRROR database using the shared Docker network: citymap-network

📁 Mounted `notebooks/` folder where users can:

Upload .laz or .las files
Create PDAL ingestion pipelines
Write point clouds directly into PostgreSQL
Run SQL for validation

🧪 Accessing the Notebook

Once the setup is complete, go to:

👉 http://localhost:8888

You’ll land on the fully functional Jupyter Notebook environment pre-connected to the database.

Folder Structure

Make sure you have the following:

./db/
  └── notebooks/
      ├── example_pipeline.ipynb
      └── sample_data.laz

🧬 Reusable Ingestion Pipeline with PDAL

The Jupyter notebook uses pdal pipeline with the writers.pgpointcloud driver to ingest .laz files into PostgreSQL.

Example logic (within the notebook):

{
  "pipeline": [
    "sample_data.laz",
    {
      "type": "writers.pgpointcloud",
      "connection": "host='db-pg' dbname='citydb_v4' user='citymap' password='test'",
      "table": "lidar_building_patch",
      "srid": 4326
    }
  ]
}

Note

You can customize the table name, CRS (SRID), or add filters as needed.

🎥 Demo Video

Want to see it in action?

Click here for Demo Video

Volume and Storage

Point cloud ingestion can be disk intensive — ensure Docker has enough memory and volume capacity allocated.

Testing

Always start with a small .laz file to confirm pipeline logic before scaling to large datasets.