import numpy as np import pytest from numpy.testing import assert_array_equal, assert_equal from skimage._shared._warnings import expected_warnings from skimage._shared.utils import _supported_float_type from skimage.filters import difference_of_gaussians, gaussian def test_negative_sigma(): a = np.zeros((3, 3)) a[1, 1] = 1 with pytest.raises(ValueError): gaussian(a, sigma=-1.0) with pytest.raises(ValueError): gaussian(a, sigma=[-1.0, 1.0]) with pytest.raises(ValueError): gaussian(a, sigma=np.asarray([-1.0, 1.0])) def test_null_sigma(): a = np.zeros((3, 3)) a[1, 1] = 1. assert np.all(gaussian(a, 0, preserve_range=True) == a) def test_default_sigma(): a = np.zeros((3, 3)) a[1, 1] = 1. assert_array_equal( gaussian(a, preserve_range=True), gaussian(a, preserve_range=True, sigma=1) ) @pytest.mark.parametrize( 'dtype', [np.uint8, np.int32, np.float16, np.float32, np.float64] ) def test_image_dtype(dtype): a = np.zeros((3, 3), dtype=dtype) assert gaussian(a).dtype == _supported_float_type(a.dtype) def test_energy_decrease(): a = np.zeros((3, 3)) a[1, 1] = 1. gaussian_a = gaussian(a, preserve_range=True, sigma=1, mode='reflect') assert gaussian_a.std() < a.std() @pytest.mark.parametrize('channel_axis', [0, 1, -1]) def test_multichannel(channel_axis): a = np.zeros((5, 5, 3)) a[1, 1] = np.arange(1, 4) a = np.moveaxis(a, -1, channel_axis) gaussian_rgb_a = gaussian(a, sigma=1, mode='reflect', preserve_range=True, channel_axis=channel_axis) # Check that the mean value is conserved in each channel # (color channels are not mixed together) spatial_axes = tuple( [ax for ax in range(a.ndim) if ax != channel_axis % a.ndim] ) assert np.allclose(a.mean(axis=spatial_axes), gaussian_rgb_a.mean(axis=spatial_axes)) if channel_axis % a.ndim == 2: # Test legacy behavior equivalent to old (multichannel = None) with expected_warnings(['multichannel']): gaussian_rgb_a = gaussian(a, sigma=1, mode='reflect', preserve_range=True) # Check that the mean value is conserved in each channel # (color channels are not mixed together) assert np.allclose(a.mean(axis=spatial_axes), gaussian_rgb_a.mean(axis=spatial_axes)) # Iterable sigma gaussian_rgb_a = gaussian(a, sigma=[1, 2], mode='reflect', channel_axis=channel_axis, preserve_range=True) assert np.allclose(a.mean(axis=spatial_axes), gaussian_rgb_a.mean(axis=spatial_axes)) def test_deprecated_multichannel(): a = np.zeros((5, 5, 3)) a[1, 1] = np.arange(1, 4) with expected_warnings(["`multichannel` is a deprecated argument"]): gaussian_rgb_a = gaussian(a, sigma=1, mode='reflect', multichannel=True) # Check that the mean value is conserved in each channel # (color channels are not mixed together) assert np.allclose(a.mean(axis=(0, 1)), gaussian_rgb_a.mean(axis=(0, 1))) # check positional multichannel argument warning with expected_warnings(["Providing the `multichannel` argument"]): gaussian_rgb_a = gaussian(a, 1, None, 'reflect', 0, True) def test_preserve_range(): """Test preserve_range parameter.""" ones = np.ones((2, 2), dtype=np.int64) filtered_ones = gaussian(ones, preserve_range=False) assert np.all(filtered_ones == filtered_ones[0, 0]) assert filtered_ones[0, 0] < 1e-10 filtered_preserved = gaussian(ones, preserve_range=True) assert np.all(filtered_preserved == 1.) img = np.array([[10.0, -10.0], [-4, 3]], dtype=np.float32) gaussian(img, 1) def test_1d_ok(): """Testing Gaussian Filter for 1D array. With any array consisting of positive integers and only one zero - it should filter all values to be greater than 0.1 """ nums = np.arange(7) filtered = gaussian(nums, preserve_range=True) assert np.all(filtered > 0.1) def test_4d_ok(): img = np.zeros((5,) * 4) img[2, 2, 2, 2] = 1 res = gaussian(img, 1, mode='reflect', preserve_range=True) assert np.allclose(res.sum(), 1) @pytest.mark.parametrize( "dtype", [np.float32, np.float64] ) def test_preserve_output(dtype): image = np.arange(9, dtype=dtype).reshape((3, 3)) output = np.zeros_like(image, dtype=dtype) gaussian_image = gaussian(image, sigma=1, output=output, preserve_range=True) assert gaussian_image is output def test_output_error(): image = np.arange(9, dtype=np.float32).reshape((3, 3)) output = np.zeros_like(image, dtype=np.uint8) with pytest.raises(ValueError): gaussian(image, sigma=1, output=output, preserve_range=True) @pytest.mark.parametrize("s", [1, (2, 3)]) @pytest.mark.parametrize("s2", [4, (5, 6)]) @pytest.mark.parametrize("channel_axis", [None, 0, 1, -1]) def test_difference_of_gaussians(s, s2, channel_axis): image = np.random.rand(10, 10) if channel_axis is not None: n_channels = 5 image = np.stack((image,) * n_channels, channel_axis) im1 = gaussian(image, s, preserve_range=True, channel_axis=channel_axis) im2 = gaussian(image, s2, preserve_range=True, channel_axis=channel_axis) dog = im1 - im2 dog2 = difference_of_gaussians(image, s, s2, channel_axis=channel_axis) assert np.allclose(dog, dog2) @pytest.mark.parametrize("s", [1, (1, 2)]) def test_auto_sigma2(s): image = np.random.rand(10, 10) im1 = gaussian(image, s, preserve_range=True) s2 = 1.6 * np.array(s) im2 = gaussian(image, s2, preserve_range=True) dog = im1 - im2 dog2 = difference_of_gaussians(image, s, s2) assert np.allclose(dog, dog2) def test_dog_invalid_sigma_dims(): image = np.ones((5, 5, 3)) with pytest.raises(ValueError): difference_of_gaussians(image, (1, 2)) with pytest.raises(ValueError): difference_of_gaussians(image, 1, (3, 4)) with pytest.raises(ValueError): with expected_warnings(["`multichannel` is a deprecated argument"]): difference_of_gaussians(image, (1, 2, 3), multichannel=True) with pytest.raises(ValueError): difference_of_gaussians(image, (1, 2, 3), channel_axis=-1) def test_dog_invalid_sigma2(): image = np.ones((3, 3)) with pytest.raises(ValueError): difference_of_gaussians(image, 3, 2) with pytest.raises(ValueError): difference_of_gaussians(image, (1, 5), (2, 4))