Fetching latest commit…
Cannot retrieve the latest commit at this time.
|Failed to load latest commit information.|
.. -*- rst -*- ====================================================== Demo of Compressive Sampling in Image Reconstruction ====================================================== This code is essentially a Scientific Python port of Justin Romberg's Compressive Sampling (CS) demo, which accompanied his publication J. Romberg, "Imaging via Compressive Sampling," Signal Processing Magazine, March 2008. The original MATLAB code can be downloaded from this web site http://users.ece.gatech.edu/~justin/spmag/ Software Requirements --------------------- * Python * Numpy * Scipy * C compiler * Matplotlib (recommended) * Cython (optional) Building And Running -------------------- Typical Python setup:: python setup.py install --prefix=/path/to/install If Cython is installed, you may re-cythonize your .pyx files by clearing the .c files first:: make clean python setup.py install --prefix=/path/to/install Once installed into a location in your PYTHONPATH, you can run the demos from the python shell:: >>> from csdemo.demos import compsense_demo >>> compsense_demo.show_comparison(20000) To recreate Fig 2(a) from the "Imaging via Compressive Sampling", do this before going for a walk (plotting requires Matplotlib):: >>> coefs_trials = np.linspace(1000, 30000, 12) >>> dct, lptv, cs = compsense_demo.compare_at(coefs_trials, plot=True)