%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
How to run the program?
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Step 0: Since you are reading this file, you have succesfuly completed Step 0.

Step 1: Open the file Expt_Setup.m

Step 2: Edit the file to set the self-explanatory parameters in the file to test with different images (eg. lena, boats, cameraman etc), blurs, noise levels etc and save it.

Step 3: Run the Expt_Setup file


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Contents of some of the files in the directory are listed below.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Expt_Setup.m : Chooses the inputs needed to set the deconvolution experiment.

WaRD_Setup.m : Sets up the experiment using the chosen parameter in Expt_Setup

Run_WaRD.m : Runs the WaRD algorithm on the setup experiment

Disp_results.m : Displays the results

Estimate_reg_param.m : In case you want to tailor the regularization parameter to your specific problem, then you could use this program to help you choose a near-optimal regularization parameter.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Functions used in the program
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Functions such as discrete wavelet transforms (mdwt.m) and redundant
wavelet transforms (mrdwt.m) from the Rice Wavelet toolbox are used in
this implementation of the WaRD algorithm.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Notes: 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1) The package is designed to run on Sun Solaris, Linux, or Windows environment. If you have any other platform, then you would need to download the Rice Wavelet tool box, compile the C files for your platform, and then run the programs.

2) The program takes less than a minute for 256x256 images on a
300MHz, 128 MB, Pentium 3 machine with Linux running on it.

3) Currently, the software is suited to run square images only. The
program uses around 60M of space for 256x256. Decreasing the number of
decomposition levels would help reduce the memory required.