OSIV Logo
Open Source Image Velocimetry
The Matlab toolbox provides access to all of the OSIV algorithms for performing cross-correlation. Unlike many Matlab toolboxes for performing PIV, this toolbox is implemented as a thin wrapper around the OSIV library, which is written in C. As such, the toolbox avoids the performance penalty usually associated with Matlab.


Instructions for installing the Matlab toolbox

  1. Download the toolbox appropriate to your operating system below
  2. Decompress the downloaded file to produce the directory 'osiv'
  3. Move the 'osiv' directory to a convienent location in your home directory
  4. Add the selected location to your Matlab path by selecting File:Set Path:Add Folder
  5. Test that installation was successful by typing 'help osivcorr' from Matlab prompt

Important: In previous versions of the OSIV Toolbox, the vector positions in the output were given using the same conventions as the grid parameters of the input. Specifically, positions were given in terms of the upper-left corner of the corresponding interrogation window. While this was the intended and documented behavior, it resulted in unnecessary confusion. Consequently, as of version 2.1.2, vector positions in the output are given in terms of the center of the corresponding interrogation window. The grid parameters continue to be specified in terms of the upper-left corner of the interrogation window. If you relied on the previous behavior, please update your scripts accordingly.