Machine Vision Toolbox for Python

A unified, object-oriented Python API that bridges the gap between raw pixels and high-level spatial reasoning by harmonizing a ‘best-of-breed’ foundation of NumPy, OpenCV, and Open3D.

• Introduction
  • Rationale
  • Image objects
  • Getting started
  • Examples
• machinevisiontoolbox package
  • The Image object
  • The PointCloud object
  • The Camera object
  • Image and PointCloud sources
  • NumPy integration
• Installing the Toolbox
  • Support files
  • Installation environment
  • Installation extras
• Command line tools
  • MVTB tool
  • Image tool
  • Tag tool
  • OCR tool
  • ROS bag tool
• Jupyter notebooks
  • Locally from command line
  • Locally using Visual Studio Code
  • In the browser using Jupyter lite
  • In the cloud using Google Colab
  • Jupyter notebook tools
  • Notebooks by category
• ROS interfaces
  • Reading a ROS bag file
  • Subscribing/publishing to a ROS message stream
• PyTorch tensor interface
  • Image → tensor
  • tensor → Image
  • Image source → batch tensor
  • batch tensor → Image iterator
• machinevisiontoolbox.base subpackage
  • Image input/output operations
  • Image type conversions
  • Color operations
  • Graphics operations
  • Peak finding
  • Data files
  • Shape models
  • Coordinate matrices
• bdsim blocks
  • Camera geometry blocks

Indices and tables

• Index

• Module Index

• Search Page