skimage.future#

Functionality with an experimental API.

Warning

Although you can count on the functions in this package being around in the future, the API may change with any version update and will not follow the skimage two-version deprecation path. Therefore, use the functions herein with care, and do not use them in production code that will depend on updated skimage versions.

skimage.future.fit_segmenter

Segmentation using labeled parts of the image and a classifier.

skimage.future.manual_lasso_segmentation

Return a label image based on freeform selections made with the mouse.

skimage.future.manual_polygon_segmentation

Return a label image based on polygon selections made with the mouse.

skimage.future.predict_segmenter

Segmentation of images using a pretrained classifier.

skimage.future.TrainableSegmenter

Estimator for classifying pixels.


skimage.future.fit_segmenter(labels, features, clf)[source]#

Segmentation using labeled parts of the image and a classifier.

Parameters:
labelsndarray of ints

Image of labels. Labels >= 1 correspond to the training set and label 0 to unlabeled pixels to be segmented.

featuresndarray

Array of features, with the first dimension corresponding to the number of features, and the other dimensions correspond to labels.shape.

clfclassifier object

classifier object, exposing a fit and a predict method as in scikit-learn’s API, for example an instance of RandomForestClassifier or LogisticRegression classifier.

Returns:
clfclassifier object

classifier trained on labels

Raises:
NotFittedError if self.clf has not been fitted yet (use self.fit).

Trainable segmentation using local features and random forests

Trainable segmentation using local features and random forests

skimage.future.manual_lasso_segmentation(image, alpha=0.4, return_all=False)[source]#

Return a label image based on freeform selections made with the mouse.

Parameters:
image(M, N[, 3]) array

Grayscale or RGB image.

alphafloat, optional

Transparency value for polygons drawn over the image.

return_allbool, optional

If True, an array containing each separate polygon drawn is returned. (The polygons may overlap.) If False (default), latter polygons “overwrite” earlier ones where they overlap.

Returns:
labelsarray of int, shape ([Q, ]M, N)

The segmented regions. If mode is ‘separate’, the leading dimension of the array corresponds to the number of regions that the user drew.

Notes

Press and hold the left mouse button to draw around each object.

Examples

>>> from skimage import data, future, io
>>> camera = data.camera()
>>> mask = future.manual_lasso_segmentation(camera)  
>>> io.imshow(mask)  
>>> io.show()  

skimage.future.manual_polygon_segmentation(image, alpha=0.4, return_all=False)[source]#

Return a label image based on polygon selections made with the mouse.

Parameters:
image(M, N[, 3]) array

Grayscale or RGB image.

alphafloat, optional

Transparency value for polygons drawn over the image.

return_allbool, optional

If True, an array containing each separate polygon drawn is returned. (The polygons may overlap.) If False (default), latter polygons “overwrite” earlier ones where they overlap.

Returns:
labelsarray of int, shape ([Q, ]M, N)

The segmented regions. If mode is ‘separate’, the leading dimension of the array corresponds to the number of regions that the user drew.

Notes

Use left click to select the vertices of the polygon and right click to confirm the selection once all vertices are selected.

Examples

>>> from skimage import data, future, io
>>> camera = data.camera()
>>> mask = future.manual_polygon_segmentation(camera)  
>>> io.imshow(mask)  
>>> io.show()  

skimage.future.predict_segmenter(features, clf)[source]#

Segmentation of images using a pretrained classifier.

Parameters:
featuresndarray

Array of features, with the last dimension corresponding to the number of features, and the other dimensions are compatible with the shape of the image to segment, or a flattened image.

clfclassifier object

trained classifier object, exposing a predict method as in scikit-learn’s API, for example an instance of RandomForestClassifier or LogisticRegression classifier. The classifier must be already trained, for example with skimage.segmentation.fit_segmenter().

Returns:
outputndarray

Labeled array, built from the prediction of the classifier.

Trainable segmentation using local features and random forests

Trainable segmentation using local features and random forests
class skimage.future.TrainableSegmenter(clf=None, features_func=None)[source]#

Bases: object

Estimator for classifying pixels.

Parameters:
clfclassifier object, optional

classifier object, exposing a fit and a predict method as in scikit-learn’s API, for example an instance of RandomForestClassifier or LogisticRegression classifier.

features_funcfunction, optional

function computing features on all pixels of the image, to be passed to the classifier. The output should be of shape (m_features, *labels.shape). If None, skimage.feature.multiscale_basic_features() is used.

Methods

fit(image, labels)

Train classifier using partially labeled (annotated) image.

predict(image)

Segment new image using trained internal classifier.

compute_features

__init__(clf=None, features_func=None)[source]#
compute_features(image)[source]#
fit(image, labels)[source]#

Train classifier using partially labeled (annotated) image.

Parameters:
imagendarray

Input image, which can be grayscale or multichannel, and must have a number of dimensions compatible with self.features_func.

labelsndarray of ints

Labeled array of shape compatible with image (same shape for a single-channel image). Labels >= 1 correspond to the training set and label 0 to unlabeled pixels to be segmented.

predict(image)[source]#

Segment new image using trained internal classifier.

Parameters:
imagendarray

Input image, which can be grayscale or multichannel, and must have a number of dimensions compatible with self.features_func.

Raises:
NotFittedError if self.clf has not been fitted yet (use self.fit).