Image Quality Measures      
VistaQualityTools 1.0+

The (VI (S) TA) Image Quality Toolbox ( VistaQualityTools ) is a Matlab Toolbox for full reference color (and also achromatic) image quality assessment based on divisive normalization models in DCT and wavelet domains.

The general idea to assess the perceptual distance between two images is to compute the q-norm Euclidean distance in the image representation at the V1 visual cortex, as suggested in [Teo&Heeger, IEEE ICIP 1994]. This markedly differs from the Mean Square Error (2-norm Euclidean measure in the spatial domain), as shown in [Pons99, Epifanio03]

NEW! These ideas have been implemented in the wavelet domain in the new code associated to the paper [Laparra10a], to be included in a future version of VistaQualityTools. Results using the wavelet based measure outperform SSIM and VIF and are intuitively interpretable in a linear way. The updated version of the wavelet image quality measure is not included in the general toolbox, but can be downloaded from HERE.


We have decided to make the library available to the research community free of charge. If you use VistaQualityTools in your research, we kindly ask that you reference this website: 

J. Malo, J. Gutiérrez, J. Muñoz and M. Simón. "VistaQualityTools: an image quality assessment toolbox for Matlab",


and the paper(s) associated to each algorithm.


Note that the package also contains some previously released public domain wavelet software authored by Eero P. Simoncelli, belonging to his MatlabPyrTools toolbox ( When using wavelet-based functions in VistaQualityTools you shall aknowledge the author of MatlabPyrTools as well!.



  • Download the file (16 MBytes)
  • Decompress at your computer and set the Matlab path accordingly.
  • Look at the help of the functions below for instructions on how to use each algorithm.
  • Warning!: VistaQualityTools1.0 has been tested on Matlab 7.2 (Matlab 2006a).  Posterior Matlab versions for windows may need recompilation of some mex files of MatlabPyrTools.

Basic features:
  • DCT distance: The original ideas for the DCT-based distance presented here date back to the late 90's [Malo97, Pons99]. The use of the V1 representation with different summation norms gives rise to the Maximum Perceptual Error (MPE) concept, as used in a number of our DCT image/video coding algorithms [  Malo99 ,  Malo00 , Malo01 , Epifanio03 , Gomez05 , Malo06 , Camps08 ].

The current version of the algorithm uses parameters to reproduce achromatic contrast incremental thresholds. Extension to color channels was done by using the same parameters in the non-linearity but using the Mullen chromatic CSFs in the blue-yellow (U) and red-green (V) channels [Mullen, J.Physiol.1985] instead of the achromatic band-pass sensitivity. Achromatic and chromatic CSFs have been relatively scaled using an appropriate chromatic contrast measure. No additional fit to any subjectivelly rated image data base has been done. Next version of the toolbox will include optimized parameters.

  • Wavelet distance: The wavelet-based distance measure only differs from the DCT one
    in the following aspects:

    • The initial linear transform is an orthogonal QMF wavelet instead of block DCT

    • As a result, the kernel in the divisive normalization may also include spatial masking (beyond the frequency masking considered in the DCT case).

    • Parameters were optimized to maximize the correlation among the predictions of the model and the Mean Opinion Score (MOS) on a subjectively rated data base: the JPEG2000 subset of the LIVE database (

  • Matlab functions for image quality assessment:
    • distance_DCT_color.m 
2-norm Euclidean measure in a DCT-based divisive normalized
domain using psychophysical parameters.

    • distance_wav_color.m

2-norm Euclidean measure in a Wavelet-based divisive normalized domain using optimized parameters to reproduce subjectively rated distortions.

    • image_quality_demo.m
This demo applies different degradations to some particular image leading to distorted images with the same MSE but quite
different visual quality. In this demo the proposed measures are
applied leading to numerical distortions more correlated to visual

Download   VistaQualityTools !

Some Results:

The suitability of a distortion measure can be seen by looking at the correlation between subjective distortions (e.g. measured by Mean Opinion Score -MOS-) and the values predicted by the distortion measure. In the examples below we compare this agreement for different distortion measures.

RMSE and our DCT-based measure for images corrupted by white noise


MPE DCT-based measure

RMSE, Structural Similarity Index SSIM (Wang et al. IEEE Tr.Im.Proc. 2004) and our Wavelet-based measure for images degraded by JPEG, JPEG2000 and fast fading distortion



MPE Wavelet-Based Measure

Download   VistaQualityTools !