These codes are written or modifed by Mehdi Rezagholizadeh and James J. Clark, Department of Electrical and Computer Engineering, McGill University. 
A part of these codes are the modified version of Ayan Chakrabarti et. al's work on color constancy to fit our algorithm. The original code can be found on: "http://vision.seas.harvard.edu/colorconstancy/"
The rest is written by Mehdi Rezagholizadeh and James J. Clark with regard to our publication: "Edge-Based and Efficient Chromaticity Spatio-spectral Models for Color Constancy."

If you use these codes, you may refer to them as follows: 
M. Rezagholizadeh and J.J. Clark, Edge-Based and Efficient Chromaticity Spatio-spectral Models for Color Constancy, in International Conference on Computer and Robot Vision (CRV), 2013.
--------------------------------------------------------------------------------------------------------------------------------------------------------

1- To generate synthesized database, you should go to the "Synthesized Database_generator" and run "Generator.m." Just note that before running, the path that synnthesized images are being saved should be specified. 

2- The folder "code_Chakrabarti_synthesized" contains the modified version of codes written by Chakrabarti et. al to work under synthesized images and "code_mine_synthesized" can be used for implementing the "Edge-based and Efficient Chromaticity color constancy algorithms". 

3- Folders "code_Chakrabarti_natural" and "code_mine_natural" are the spatio-spectral implementation for original and modified spatio-spectral model that works with natural images. (Just note that the folder that contains the image database should be substitute in the m-files properly based on the path that contain your database. 

4- The folder "RUN_mine", give you the chance to obtain the final results given in our published paper.    
For any inquery with regard to these codes you may contact "mehdi.rezagholizadeh@gmail.com". 
 