Highly Nonlinear Approximations for Sparse Signal Representation


Encrypted Image Folding Software - EIFS (news: now extended to SCEIFS)

This software was developed by James Bowley to provide a MATLAB implementation of the Encrypted Image Folding (EIF) method proposed in the paper

[1] Sparsity and "Something Else": An Approach to Encrypted Image Folding
by James Bowley and Laura Rebollo-Neira.
Published in IEEE Signal Processing Letters, Vol. 8 No 3, 189--192, DOI: 10.1109/LSP.2011.2106496 (2011). Corrected version here.

All the source files for the routines are available in a zip archive (EIFS.zip), which contains a manual describing the main routines used, this manual can also be downloaded seperatley (EIFS_Manual.pdf).

Additionaly an example in html is located here.

To install the EIFS software:

  1. Download and extract the file EIFS.zip. This should give you a top directory EIFS and 7 sub directories listed below:
    • Approximation_Routines - required for approximating and folding the images.
    • Common_Routines - required for both folding and unfolding.
    • Example - contains an example showing how to use the EIF software.
    • Expanding_Routines - required for expanding folded images.
    • Folding_Routines - required for folding the image.
    • Images - location of test images.
    • Mex_Files - source code for the mex implementation of OMP2D.
  2. Add the directory EIFS and all subdirectories to your MATLAB path.
  3. Optional: To decrease the execution time when using the RDCTDB dictionary you can use the RDCTDBMex method which uses a C++ implementation of OMP2D. To do this
    • i) compile the OMP2D.cpp file,
    • ii) copy it to the Approximation_Routines subdirectory.
    Note 1: i) and ii) can be done simply running the script CompileOMP in the Mex_Files subdirectoy.
    Note 2: MATLAB has to be configured to compile mex files, please refer to the MATLAB documentation for details of how to do this.
The EIFS software has been tested on MATLAB versions 2009a and 2010a.