def test_normalize_norm_image():
pixels = np.ones((120, 120, 3))
pixels[..., 0] = 0.5
pixels[..., 1] = 0.2345
image = Image(pixels)
image.normalize_norm_inplace()
assert_allclose(np.mean(image.pixels), 0, atol=1e-10)
assert_allclose(np.linalg.norm(image.pixels), 1)
def test_normalize_norm_image():
pixels = np.ones((120, 120, 3))
pixels[..., 0] = 0.5
pixels[..., 1] = 0.2345
image = Image(pixels)
image.normalize_norm_inplace()
assert_allclose(np.mean(image.pixels), 0, atol=1e-10)
assert_allclose(np.linalg.norm(image.pixels), 1)
def test_normalize_norm_image_per_channel():
pixels = np.random.randn(120, 120, 3)
pixels[..., 1] *= 17
pixels[..., 0] += -114
pixels[..., 2] /= 30
image = Image(pixels)
image.normalize_norm_inplace(mode='per_channel')
assert_allclose(
np.mean(image.as_vector(keep_channels=True), axis=0), 0, atol=1e-10)
assert_allclose(
np.linalg.norm(image.as_vector(keep_channels=True), axis=0), 1)
def test_normalize_norm_image_per_channel():
pixels = np.random.randn(120, 120, 3)
pixels[..., 1] *= 17
pixels[..., 0] += -114
pixels[..., 2] /= 30
image = Image(pixels)
image.normalize_norm_inplace(mode='per_channel')
assert_allclose(np.mean(image.as_vector(keep_channels=True), axis=0),
0,
atol=1e-10)
assert_allclose(
np.linalg.norm(image.as_vector(keep_channels=True), axis=0), 1)
def build_parts_image(image, centres, parts_shape, offsets=np.array([[0, 0]]),
normalize_parts=False):
# extract patches
parts = image.extract_patches(PointCloud(np.round(centres.points)), np.array(parts_shape), offsets)
# build parts image
# img.pixels: n_centres x n_offsets x n_channels x height x width
img = Image(parts)
if normalize_parts:
# normalize parts if required
img.normalize_norm_inplace(mode='per_channel')
return img
