def main(argv=None):
if argv is None:
argv = sys.argv
if len(argv) != 5:
print "Usage: python %s <CheckerboardImage> <OutlierFraction> <CorruptedImage> <LowRankImage>" % sys.argv[0]
sys.exit(1)
# outlier fraction
p = float(sys.argv[2])
# read image
I = sitk.ReadImage(argv[1])
# data for processing
X = sitk.GetArrayFromImage(I)
# number of pixel
N = np.prod(X.shape)
eps = np.round(np.random.uniform(-10, 10, 100))
idx = np.random.random_integers(0, N-1, np.round(N*p))
X.ravel()[idx] = np.array(200+eps, dtype=np.uint8)
# write outlier image
J = sitk.GetImageFromArray(X)
sitk.WriteImage(J, sys.argv[3])
# decompose X into L+S
L, S, _ = ialm.recover(X)
C = sitk.GetImageFromArray(np.asarray(L, dtype=np.uint8))
sitk.WriteImage(C, sys.argv[4])
# compute mean-square error and Frobenius norm
print "MSE: %.4g" % np.sqrt(np.asmatrix((L-sitk.GetArrayFromImage(I))**2).sum())
print "Frobenius-Norm: %.4g" % np.linalg.norm(L-sitk.GetArrayFromImage(I),ord='fro')
def test_recover(self):
"""Test recovery from outliers.
"""
# run recovery
lr, sp, _ = ialm.recover(self._data, None)
# load baseline (no outliers)
baseline = np.genfromtxt("im_baseline.dat")
# Frobenius norm between recovered mat. and baseline
d = np.linalg.norm(np.round(lr)-baseline, ord='fro')
self.assertTrue(np.allclose(d,0.0))
def rpca(Y,lamda):
t_begin = time.clock()
gamma = lamda* np.sqrt(float(Y.shape[1])/Y.shape[0])
low_rank, sparse, n_iter,rank, sparsity, sumSparse= ialm.recover(Y,gamma)
gc.collect()
t_end = time.clock()
t_elapsed = t_end- t_begin
print 'RPCA takes:%f seconds'%t_elapsed
return (low_rank, sparse, n_iter,rank, sparsity, sumSparse)
def rpca(Y, lamda):
t_begin = time.clock()
gamma = lamda * np.sqrt(float(Y.shape[1]) / Y.shape[0])
low_rank, sparse, n_iter, rank, sparsity, sumSparse = ialm.recover(
Y, gamma)
gc.collect()
t_end = time.clock()
t_elapsed = t_end - t_begin
print 'RPCA takes:%f seconds' % t_elapsed
return (low_rank, sparse, n_iter, rank, sparsity, sumSparse)
def main(argv=None):
if argv is None:
argv = sys.argv
if len(argv) != 5:
print "Usage: python %s <CheckerboardImage> <OutlierFraction> <CorruptedImage> <LowRankImage>" % sys.argv[
0]
sys.exit(1)
# outlier fraction
p = float(sys.argv[2])
# read image
I = sitk.ReadImage(argv[1])
# data for processing
X = sitk.GetArrayFromImage(I)
# number of pixel
N = np.prod(X.shape)
eps = np.round(np.random.uniform(-10, 10, 100))
idx = np.random.random_integers(0, N - 1, np.round(N * p))
X.ravel()[idx] = np.array(200 + eps, dtype=np.uint8)
# write outlier image
J = sitk.GetImageFromArray(X)
sitk.WriteImage(J, sys.argv[3])
# decompose X into L+S
L, S, _ = ialm.recover(X)
C = sitk.GetImageFromArray(np.asarray(L, dtype=np.uint8))
sitk.WriteImage(C, sys.argv[4])
# compute mean-square error and Frobenius norm
print "MSE: %.4g" % np.sqrt(
np.asmatrix((L - sitk.GetArrayFromImage(I))**2).sum())
print "Frobenius-Norm: %.4g" % np.linalg.norm(
L - sitk.GetArrayFromImage(I), ord='fro')
