def test_niblack_sauvola_pathological_image():
# For certain values, floating point error can cause
# E(X^2) - (E(X))^2 to be negative, and taking the square root of this
# resulted in NaNs. Here we check that these are safely caught.
# see https://github.com/scikit-image/scikit-image/issues/3007
value = 0.03082192 + 2.19178082e-09
src_img = np.full((4, 4), value).astype(np.float64)
assert not np.any(np.isnan(threshold_niblack(src_img)))
def test_niblack_sauvola_pathological_image():
# For certain values, floating point error can cause
# E(X^2) - (E(X))^2 to be negative, and taking the square root of this
# resulted in NaNs. Here we check that these are safely caught.
# see https://github.com/scikit-image/scikit-image/issues/3007
value = 0.03082192 + 2.19178082e-09
src_img = np.full((4, 4), value).astype(np.float64)
assert not np.any(np.isnan(threshold_niblack(src_img)))
def test_threshold_niblack_iterable_window_size(self):
ref = np.array([[False, False, False, True, True],
[False, False, True, True, True],
[False, True, True, True, False],
[False, True, True, True, False],
[True, True, False, False, False]])
thres = threshold_niblack(self.image, window_size=[3, 5], k=0.5)
out = self.image > thres
assert_array_equal(ref, out)
def test_threshold_niblack(self):
ref = np.array(
[[False, False, False, True, True],
[False, True, True, True, True],
[False, True, True, True, False],
[False, True, True, True, True],
[True, True, False, False, False]]
)
thres = threshold_niblack(self.image, window_size=3, k=0.5)
out = self.image > thres
assert_equal(ref, out)
