🐀 MapMole

Raster change detection made easy — upload two GeoTIFF satellite images and instantly see what changed.

Overview

MapMole compares two raster images (GeoTIFF) and produces a binary change map highlighting areas of difference. It ships with both a Streamlit web UI and a command-line interface, so you can use whichever fits your workflow.

Use Cases

Domain	Example
HADR	Assess damage after natural disasters by comparing pre- and post-event satellite imagery
IMINT	Monitor activity patterns, infrastructure, and developments over time
Cartography	Detect deforestation, urban expansion, or shifting coastlines for mapsheet refreshes

How It Works

1. Read — extracts the first band of each GeoTIFF (single-band grayscale assumed).
2. Align — if dimensions differ, the second image is resampled to match the first using bilinear interpolation via Rasterio.
3. Diff — computes the absolute pixel-wise difference.
4. Enhance — normalises to 0–255, applies a log stretch (log(1 + x)), then thresholds with a configurable sensitivity factor to produce a binary mask.

Tip: Use images in the same projection system for best results. Colour-correcting the imagery beforehand also helps. Results may vary with cloud cover or georectification errors.

Quick Start

Prerequisites

• Python 3.9+

Install

pip install -r requirements.txt

Web UI (recommended)

streamlit run app.py

This opens an interactive dashboard where you can:

• Upload two .tif images
• Adjust the change-detection threshold with a slider
• Choose a colour map
• View side-by-side results with statistics
• Download the output change map

CLI

# Interactive prompts
python mapmole.py
# Or pass arguments directly
python mapmole.py --image1 before.tif --image2 after.tif --output changes.tif --threshold 0.7

Project Structure

mapmole/
├── app.py # Streamlit web UI
├── core.py # Shared change-detection engine
├── mapmole.py # Command-line interface
├── requirements.txt # Python dependencies
├── .gitignore
├── LICENSE # MIT
└── readme/ # Images for this README

---

CLI Reference

The CLI supports both interactive mode (guided prompts) and argument mode for scripting and automation.

Interactive Mode

python mapmole.py

You'll be prompted to enter:

1. Path to the first .tif image
2. Path to the second .tif image
3. Path for the output .tif file

A matplotlib window will display the before, after, and change map side by side.

Argument Mode

python mapmole.py --image1 <path> --image2 <path> --output <path> [--threshold <float>]

Flag	Required	Default	Description
--image1	Yes*	—	Path to the first (before) .tif image
--image2	Yes*	—	Path to the second (after) .tif image
--output	Yes*	—	Path for the output change map .tif
--threshold	No	0.75	Sensitivity factor (0.0–1.0). Lower = more changes detected

*If omitted, the CLI falls back to interactive prompts.

Examples

# Full argument mode — no prompts
python mapmole.py --image1 before.tif --image2 after.tif --output changes.tif
# High sensitivity (detect subtle changes)
python mapmole.py --image1 before.tif --image2 after.tif --output changes.tif --threshold 0.5
# Low sensitivity (only major changes)
python mapmole.py --image1 before.tif --image2 after.tif --output changes.tif --threshold 0.9
# Mixed — supply images via args, get prompted for output path
python mapmole.py --image1 before.tif --image2 after.tif

License

MIT — made by kevanwee.