• GEOJSON
  • CSV
  • MANUAL
  • JSON

How to extract aerial feature geometries with your AI

Connect Nearmap to your AI client to turn high-res imagery into spatial data. Once connected, your agent handles the heavy lifting of identifying objects in the field.

Ask AI about this page

Short answer

How can I use my AI to map features from aerial imagery?

Your AI uses Nearmap's detection tools to find specific objects and return their geometric data. You get precise coordinates and shapes without manual digitizing. This process turns raw pixels into structured spatial layers.

Where your data lands.

The results of your extraction.

The AI processes the imagery and organizes the results into these specific outputs.

GEOJSON

Spatial files

The AI converts detected shapes into standard geometry formats. You can drop these directly into your GIS software.

CSV

Attribute tables

Get a list of all detected features with their properties. This works well for bulk analysis of solar panel counts or pool locations.

MANUAL

Human review

The AI flags low-confidence detections for you. You decide which geometries are accurate enough to keep.

JSON

API payloads

Receive structured data objects containing the exact bounding boxes and polygons found in the imagery.

The workflow

What your AI does when the imagery arrives.

The AI moves from identifying what is possible to pulling the actual shapes from the map.

  1. Identify available classes

    Your agent first checks which types of features Nearmap can actually see, such as buildings or vegetation.

    list_ai_feature_classes

  2. Locate specific features

    The AI scans the requested imagery area to find the specific objects you asked for.

    get_ai_detected_features

  3. Extract geometry

    Once the objects are found, the AI pulls the polygon and point data for each detection.

    get_ai_detected_features

  4. Format for export

    The agent organizes the raw detection data into a clean structure for your use.

    get_ai_detected_features

Try it

Copy these to start.

Use these prompts to trigger the extraction process.

Starting points

Swap out the bracketed text for your specific target, like 'solar panels' or 'swimming pools'.

Nearmap Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_oViNLOTWVhYk2I41UhrykHsqYqXvqSGGO7q494hx/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Nearmap capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "nearmap-high-res-aerial-imagery-ai-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_oViNLOTWVhYk2I41UhrykHsqYqXvqSGGO7q494hx/mcp"
    }
  }
}
Copy into chat04

  • List all the AI feature classes available in Nearmap.

  • Find all the solar panels in this imagery and give me their geometries.

  • Extract the building footprints for this area as a list of polygons.

  • Identify all swimming pools in the current view and provide their locations.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Nearmap once, then ask.

You only need to link your account once. Your credentials stay encrypted, and you can immediately start asking your agent to map features.

Connect Nearmap to your AI

FAQ

How this task behaves.

  • 01

    Can the AI change the actual imagery?

    No. The AI only reads the imagery to identify features; it cannot edit or modify the Nearmap maps.

  • 02

    Does the AI automatically save these to my GIS?

    No. The AI provides the data to you in your chat interface. You must then save or import that data into your own tools.

  • 03

    What happens if the AI isn't sure about a feature?

    The AI returns the detection, but you should always perform a visual check on the results to ensure accuracy.

  • 04

    Can I ask for features that aren't in the feature classes?

    No. The AI can only detect what is explicitly listed in the available feature classes.

  • 05

    Will the AI delete any of my existing data?

    No. This MCP is read-only regarding your imagery and existing datasets.

  • More questions about Nearmap? The Connector page answers them. See everything the Nearmap Connector can do

Connect Nearmap to Claude, Cursor, ChatGPT & more