Image Quality Assessment (IQA) Documentation

Overview

The Image Quality Assessment (IQA) library implements popular algorithms for generating image/video quality metrics. There are several principles that guide IQA's development:

• It must be easy to use (cover the common use cases... a.k.a. the 90% rule).
• It must be accurate (results are checked against provided Matlab files in Octave).
• It must be fast.
• It must be portable.
• It must be reliable (unit tested, Valgrind clean).

The following algorithms are implemented in the current release:

• MSE
• PSNR
• SSIM
• MS-SSIM
• MS-SSIM*

Check out the algorithms page for information about each one.

Download

Source: http://sourceforge.net/projects/iqa/files/

Browse the source tree: http://iqa.svn.sourceforge.net/viewvc/iqa/trunk

SVN Checkout:

 svn co https://iqa.svn.sourceforge.net/svnroot/iqa/trunk iqa 

Getting Started

• Algorithm Information
• Building IQA
• Sample Code

Having Problems?

Submit a question or file a bug report at https://sourceforge.net/projects/iqa/support

What's New in 1.1.2

• Fixed Windows linker warnings for debug builds.
• Fixed seg-fault in MS-SSIM algorithm when height is greater than width (#3349231).

See CHANGELOG.txt for a detailed list of changes in each version.