MTLearn Documentation

MTLearn is a C++/Python research library for learnable connected operators based on morphological trees.

This site combines:

• Python API reference generated from docstrings.

• User guides for building experiments with the public Python API.

Concepts

• Morphology Concepts
  • Morphological Trees
  • Max-Tree and Min-Tree
  • Tree of Shapes
  • Node Ids and Node Slots
  • Proper Parts
  • Scalar Attributes and Groups
• Attribute Catalog
  • Terminology
  • Attribute Groups
  • Contracts
  • Gray-Level Attributes
  • Size and Bounding-Box Attributes
  • Distance-Transform Attributes
  • Filled-Shape Attributes
  • Moment Attributes
  • Boundary and Bitquad Attributes
  • Tree-Topology Attributes
  • Choosing Attributes

Guides

• Guides
  • Getting Started
  • Morphology User Guide
  • Connected Filter Preprocessing
  • Custom Attributes and Filter Specs
  • PyTorch Integration
  • Reading Order

API Reference

• Python API
  • Morphology Facade
  • Layers
  • CFP Preparation and Storage
  • Native Binding Classes
  • Data Helpers
  • Datasets