def svm_classifiy(img):
training_labels = np.zeros(img.shape[:2], dtype=np.uint8)
# training_labels[:10] = 1
training_labels[180:, 420:] = 1
training_labels[70:110, 360:390] = 2
training_labels[55:85, 310:320] = 2
training_labels[120:140, 220:240] = 3
# training_labels[150:200, 720:860] = 4
sigma_min = 1
sigma_max = 16
features_func = partial(feature.multiscale_basic_features,
intensity=True,
edges=False,
texture=True,
sigma_min=sigma_min,
sigma_max=sigma_max,
multichannel=True)
features = features_func(img)
clf = RandomForestClassifier(n_estimators=50,
n_jobs=-1,
max_depth=10,
max_samples=0.05)
clf = future.fit_segmenter(training_labels, features, clf)
result = future.predict_segmenter(features, clf)
fig, ax = plt.subplots(1, 2, sharex=True, sharey=True, figsize=(9, 4))
ax[0].imshow(segmentation.mark_boundaries(img, result, mode='thick'))
ax[0].contour(training_labels)
ax[0].set_title('Image, mask and segmentation boundaries')
ax[1].imshow(result)
ax[1].set_title('Segmentation')
fig.tight_layout()
plt.show()
def test_trainable_segmentation_multichannel():
img = np.zeros((20, 20, 3))
img[:10] = 1
img += 0.05 * np.random.randn(*img.shape)
labels = np.zeros_like(img[..., 0], dtype=np.uint8)
labels[:2] = 1
labels[-2:] = 2
clf = DummyNNClassifier()
features_func = partial(multiscale_basic_features,
edges=False,
texture=False,
sigma_min=0.5,
sigma_max=2,
multichannel=True)
features = features_func(img)
clf = fit_segmenter(labels, features, clf)
out = predict_segmenter(features, clf)
assert np.all(out[:10] == 1)
assert np.all(out[10:] == 2)
def test_trainable_segmentation_predict():
img = np.zeros((20, 20))
img[:10] = 1
img += 0.05 * np.random.randn(*img.shape)
labels = np.zeros_like(img, dtype=np.uint8)
labels[:2] = 1
labels[-2:] = 2
clf = DummyNNClassifier()
features_func = partial(
multiscale_basic_features,
edges=False,
texture=False,
sigma_min=0.5,
sigma_max=2,
)
features = features_func(img)
clf = fit_segmenter(labels, features, clf)
test_features = np.random.random((5, 20, 20))
with pytest.raises(ValueError) as err:
_ = predict_segmenter(test_features, clf)
assert 'type of features' in str(err.value)
training_labels[:170, :400] = 1
training_labels[600:900, 200:650] = 2
training_labels[330:430, 210:320] = 3
training_labels[260:340, 60:170] = 4
training_labels[150:200, 720:860] = 4
sigma_min = 1
sigma_max = 16
features_func = partial(feature.multiscale_basic_features,
intensity=True, edges=False, texture=True,
sigma_min=sigma_min, sigma_max=sigma_max,
channel_axis=-1)
features = features_func(img)
clf = RandomForestClassifier(n_estimators=50, n_jobs=-1,
max_depth=10, max_samples=0.05)
clf = future.fit_segmenter(training_labels, features, clf)
result = future.predict_segmenter(features, clf)
fig, ax = plt.subplots(1, 2, sharex=True, sharey=True, figsize=(9, 4))
ax[0].imshow(segmentation.mark_boundaries(img, result, mode='thick'))
ax[0].contour(training_labels)
ax[0].set_title('Image, mask and segmentation boundaries')
ax[1].imshow(result)
ax[1].set_title('Segmentation')
fig.tight_layout()
##############################################################################
# Feature importance
# ------------------
#
# We inspect below the importance of the different features, as computed by
