RESOURCE

Machine Vision Toolbox

Toolbox versions

Many videos in the Academy make use of MATLAB® examples, and many of those require extra (free) software such as the Machine Vision Toolbox for MATLAB, details below.

The videos were made in the period 2015-16 and are consistent with:

You can install the toolbox following the instructions above.

If you have MATLAB19a and a MATLAB Drive account or MATLAB Online you can access the older version of the Toolbox without needing to download anything by simply linking to a MATLAB Drive shared folder:

  1. Click this link to accept the invitation to share.
  2. A folder named RVC1 will appear in your MATLAB drive
  3. Using MATLAB file browser navigate to RVC1/rvctools and double-click the script named startup_rvc.m

The most recent versions of this book and Toolbox are:

About the toolbox MVTB3.4

The third release of the Toolbox represents over two decades of development. The last release was in 2005 and this version captures a large number of changes and extensions generated over the last two years that support Peter Corke’s new book Robotics, Vision & Control.

The Machine Vision Toolbox (MVTB) provides many functions that are useful in machine vision and vision-based control. It is a somewhat eclectic collection reflecting his personal interest in areas of photometry, photogrammetry, and colorimetry. It includes over 100 functions spanning operations such as image file reading and writing, acquisition, display, filtering, blob, point and line feature extraction, mathematical morphology, homographies, visual Jacobians, camera calibration and colour space conversion.

The Toolbox, combined with MATLAB® and a modern workstation computer, is a useful and convenient environment for investigating machine vision algorithms. For modest image sizes, the processing rate can be sufficiently “real-time” to allow for closed-loop control. Focus of attention methods such as dynamic windowing (not provided) can be used to increase the processing rate. With input from a firewire or web camera (support provided) and output to a robot (not provided), it would be possible to implement a visual servo system entirely in MATLAB.

An image is usually treated as a rectangular array of scalar values representing intensity or perhaps range. The matrix is the natural datatype for MATLAB and thus makes the manipulation of images easily expressible in terms of arithmetic statements in MATLAB® language. Many image operations such as thresholding, filtering and statistics can be achieved with existing MATLAB® functions.

The Toolbox extends this core functionality with M-files that implement functions and classes and mex-files for some compute intensive operations. It is possible to use mex-files to interface with image acquisition hardware ranging from simple framegrabbers to robots. Examples for firewire cameras under Linux are provided.

The routines are written in a straightforward manner that allows for easy understanding. MATLAB® vectorization has been used as much as possible to improve efficiency, however some algorithms are not amenable to vectorization. If you have the MATLAB® compiler available then this can be used to compile bottleneck functions. Some particularly compute-intensive functions are provided as mex-files and may need to be compiled for the particular platform. This Toolbox considers images generally as arrays of double precision numbers. This is extravagant on storage, though this is much less significant today than it was in the past.

This Toolbox is not a clone of the Mathwork® Image Processing Toolbox (IPT), although there are many functions in common. This toolbox predates IPT by many years, is open-source and contains many functions that are useful for image feature extraction and control.

This third release of the Toolbox has been significantly extended to include classes to represent different types of cameras (perspective, fisheye, catadioptric and spherical), pose estimation, visual Jacobians and advanced segmentation techniques such as MSER and graph-based. The Toolbox also includes Simulink® models for PBVS and IBVS visual servoing systems for arm-type, mobile and flying robots. 

Advantages of the Toolbox

  • The code is quite mature and provides a point of comparison for other implementations of the same algorithms.
  • The routines are generally written in a straightforward manner, which allows for easy understanding, perhaps at the expense of computational efficiency. If you feel strongly about computational efficiency then you can always rewrite the function to be more efficient, compile the M-file using the Matlab compiler, or create a MEX version.
  • Since source code is available there is a benefit for understanding and teaching.

Downloading the Toolbox

  • Download the Machine Vision Toolbox in zip format (.zip).
  • The Toolbox is tested with MATLAB® R2011a.
  • To install the Toolbox simply unpack the archive, which will create the directories (folders): vision, simulink, and common.
  • Adjust your MATLABPATH to include this directory.
  • Execute the startup file rvctools/startup_rvc.m and this will place the correct directories in your MATLAB path.

Documentation

  • Robotics, Vision & Control (Corke, 2011) is a detailed introduction to colour, image geometry, image processing, feature extraction, multi-view geometry and vision-based control, all illustrated using the Machine Vision Toolbox for MATLAB.
  • The manual Machine Vision Toolbox for MATLAB Release 3 [PDF] s a printable document (around 200 pages). It is auto-generated from the comments in the MATLAB code and is fully hyperlinked to external websites, the table of content to functions, and the “See also” functions to each other.
  • The Toolbox documentation also appears in the MATLAB help browser.

Related publications

If you like the Toolbox and want to cite it please reference it as:

  • P.I. Corke, “Robotics, Vision & Control”, Springer 2011, ISBN 978-3-642-20143-1. [bibtex]

The following are now quite old publications about the Toolbox and the syntax has changed considerably over time:

  • P. Corke. Machine vision toolbox. IEEE Robotics and Automation Magazine, 12(4):16–25, Nov. 2005. [PDF]
  • P.I. Corke, “MATLAB toolboxes: robotics and vision for students and teachers”, IEEE Robotics and Automation Magazine, Volume 14(4), December 2007, pp. 16-17 [PDF]

Support

There is no support! This software is made freely available in the hope that you find it useful in solving whatever problems you have to hand. Professor Corke is happy to correspond with people who have found genuine bugs or deficiencies but response time can be long and there is no guarantee that Professor Corke will respond to your email. Contributions for inclusion in future versions of the toolbox will be considered and you will be suitably acknowledged.

Professor Corke specifically states: ‘I can guarantee that I will not respond to any requests for help with assignments or homework, no matter how urgent or important they might be to you. That’s what your teachers, tutors, lecturers and professors are paid to do’.

You might instead like to communicate with other users via the Google Group discussion. You need to sign up in order to post, and this is moderated by Professor Corke, allow a few days for this to happen. You will need to write a few words about why you want to join the list so you can be distinguished from a spammer or a web-bot.

There is also a frequently asked questions (FAQ) wiki page.

The Toolbox will not work with Octave. Octave is quite sophisticated, but there are just too many differences compared to vanilla MATLAB®. You’re on your own with this.

Other vision related software on the web

History

The Machine Vision Toolbox developed slowly during the 1990s to assist in research related to visual servoing. The first release was in 1999, and the second in 2005 to coincide with a paper.

During the writing of the book, the code grew in size and sophistication and made strong use of classes for cameras and image features.