def denoise(self, var_in, var_out, niter, W):
arr = np.array(self.vars[var_in], dtype=np.float32)
img = tvdenoise(arr, niter=niter, W=W)
self.vars[var_out] = img
if len(sys.argv) > 1:
parser = argparse.ArgumentParser(description='Total-variation denoising')
parser.add_argument('filename_in', metavar='in', help='the input file')
parser.add_argument('-W', default=50.0, type=float,
help='weight on regularization term')
parser.add_argument('-n', default=10, type=int,
help='number of iterations')
args = parser.parse_args()
filename = args.filename_in
n = args.n
W = args.W
else:
filename = os.path.join(data_dir, 'lena256.tif')
n = 10
W = 50.0
im = imread(filename)
imshow(im)
im = np.array(im, dtype=np.float32)
im2 = tvdenoise(im, n, W)
def view_float(im):
m = np.min(im)
M = np.max(im)
scale = float(M - m)
rescaled = np.rint(255.0 * (im - m) / scale)
imshow(np.array(rescaled, dtype=np.uint8))
view_float(im2)
