Habitat-Mapper

Habitat Segmentation with AI

(The artist formerly known as "Kelp-O-Matic")

Get Started View on GitHub

---

What is Habitat-Mapper?

Habitat-Mapper is an open-source tool for segmenting geospatial imagery. It automates the detection of coastal species using deep learning, streamlining analysis for drone (RPAS), aerial, and satellite imagery.

• Kelp Detection

  ---

  Detects Macrocystis and Nereocystis canopies in RGB and RGB+NIR fixed-wing and drone imagery.

• Intertidal Life

  ---

  Identifies Mussels and Gooseneck Barnacles in high-resolution RGB drone imagery.

• Satellite Scale

  ---

  Analyzes 8-band PlanetLabs satellite imagery for broad-scale kelp presence/absence.

Quick Start

Install and run your first segmentation in three steps.

1. Install via pip

   pip install habitat-mapper
   

2. Check available models

   hab models
   

Model Name	Revision	Description
kelp-ps8b	20250818	Kelp segmentation model for 8-band PlanetScope imagery.
kelp-rgb	20240722	Kelp segmentation model for RGB drone imagery.
kelp-rgbi	20231214	Kelp segmentation model for 4-band RGB+NIR drone imagery.
mussel-gooseneck-rgb	20250725	Mussel and gooseneck barnacle segmentation model for RGB drone imagery.
mussel-rgb	20250711	Mussel segmentation model for RGB drone imagery.

1. Segment an image

   hab segment kelp-rgb input_image.tif output_mask.tif
   

New to the Command Line?

Don't worry! We have a dedicated Beginner's Guide that walks you through setting up your terminal, installing the software, and processing your first image.

---

Choose Your Path

New to Command Line ToolsExperienced Python Users

Never used a terminal before? Start here:

1. Terminal Crash Course - Learn basic terminal navigation
2. Installation Guide - Set up the software step-by-step
3. Your First Segmentation - Process your first image
4. Post-Processing Results - Clean and analyze your data

Jump straight to what you need:

• Installation - Quick pip install habitat-mapper
• CLI Reference - All command options
• Python API - Use as a library in your scripts
• Input Requirements - Data format specifications

---

Project & Community

Contribute

We welcome feedback! If you encounter bugs or have feature requests, please use our Issue Tracker.

Team

This project was co-created at the Hakai Institute by:

Name	Role
Taylor Denouden	Machine Learning & Software Development
Luba Reshitnyk	Scientific Application (Kelp), Data Collation
Will McInnes	Data Collation & Labelling (Mussels)
Alyssa Gehman	Scientific Application (Mussels)

License

Habitat-Mapper is released under the MIT license.