change history.txt

% This is a sample Matlab/Octave program which reconstructs an image sampled compressively 
% with binarized or continuous DCT functions using the FDRI method.
% 
% This code is part of the FDRI package
% https://www.igf.fuw.edu.pl/fdri
% Copyright (C) 2018/2019 K. M. Czajkowski, A. Pastuszczak and R. Kotyński
%


FDRI version history

1. v1.0, Aug 2018
 - initial release

2. v1.1, Feb 11, 2019
 - corrected svd-based calculation of pseudoinverse in the presence of very small eigenvalues (fdri.m)
 - rearanged folders 
 - changes in the example_fdri:  more test images are included, psnr is calculated, by default the bases functions are chosen deterministically (this improves PSNR by approx. 2 dB)

3. v1.2, Feb 14, 2019
  - added example 2 (example2_fdri_wh.m) with Walsh-Hadamard sampling and FDRI reconstruction
  - added example 3 (example3_fdri_morlet.m) with sampling based on binary or continuous Morlet wavelets convolved with white noise and with FDRI reconstruction
  - examples of sampling functions are also plotted (PlotSamplingFunctionExamples.m)

4. v1.3 Apr 19, 2019
  - added example 4 (example4_fdri_nsimplex.m) with DCT sampling coded on n-simplex
  - wrapped reconstruction plotting, printing of simulation information and image loading into separate functions
  - added dithering of sampling functions upscaled to a desired sized and calculation of downscaled sampling matrix (DitherMeasurementMatrix and Downsize functions)
  - added utility for saving sampling functions
  - code is now available at https://github.com/KMCzajkowski/FDRI-single-pixel-imaging