概率图#

APOC 基于 pyclesperanto 和 sklearn。

让我们从加载一个示例图像和一些地面真相开始：

from skimage.io import imread, imshow, imsave
import matplotlib.pyplot as plt
import numpy as np
import apoc

image = imread('blobs.tif')
imshow(image)

<matplotlib.image.AxesImage at 0x25fc553f9a0>
../_images/5a4c39e0461e2e52e2033810b9b5414fac596269ce2d789b46600c150d286c9f.png 
do_manual_annotation = False

if do_manual_annotation: # you can use this to make manual annotations
    import napari

    # start napari
    viewer = napari.Viewer()
    napari.run()

    # add image
    viewer.add_image(image)

    # add an empty labels layer and keep it in a variable
    labels = np.zeros(image.shape).astype(int)
    viewer.add_labels(labels)
    
else:
    labels = imread('annotations_3class.tif')

manual_annotations = labels
if do_manual_annotation:
    imsave('annotations_3class.tif', manual_annotations)

from skimage.io import imshow
imshow(manual_annotations, vmin=0, vmax=3)

C:\Users\rober\miniconda3\envs\bio_38\lib\site-packages\skimage\io\_plugins\matplotlib_plugin.py:150: UserWarning: Low image data range; displaying image with stretched contrast.
  lo, hi, cmap = _get_display_range(image)

<matplotlib.image.AxesImage at 0x25fc565c5e0>
../_images/a01cbf11939ee92b88fe012daa0788b19e9cb563e2a4a0516873c46140fa1f49.png

训练#

现在我们训练一个PixelClassifier，它在底层是一个scikit-learn RandomForestClassifier。训练后，分类器将被转换为与clij兼容的OpenCL代码，并以给定的文件名保存到磁盘。

# define features: original image, a blurred version and an edge image
features = "original gaussian_blur=2 sobel_of_gaussian_blur=2"

# this is where the model will be saved
cl_filename = 'my_model.cl'

output_probability_of_class = 3

apoc.erase_classifier(cl_filename)
clf = apoc.ProbabilityMapper(opencl_filename=cl_filename, output_probability_of_class=output_probability_of_class)
clf.train(features, manual_annotations, image)

预测#

然后可以使用分类器对给定图像中的所有像素进行分类。起点仍然是特征堆栈。因此，用户必须确保在训练和预测时使用相同的特征。

result = clf.predict(image=image)
imshow(result)

<matplotlib.image.AxesImage at 0x25fc5770670>
../_images/58c1763290241911a3217d74a012f1e2b171748d445be2066bce602eaec3059a.png

对其他类别进行训练/预测#

我们现在将再次训练并输出另一个类别的概率

output_probability_of_class = 2

apoc.erase_classifier(cl_filename)
clf = apoc.ProbabilityMapper(opencl_filename=cl_filename, output_probability_of_class=output_probability_of_class)
clf.train(features, manual_annotations, image)

result = clf.predict(image=image)
imshow(result)

<matplotlib.image.AxesImage at 0x25fc57d9d30>
../_images/084b771ea6ee52d109f6b68e1029ad74a455ecde73273ce4ee1a782321d28ce6.png 
output_probability_of_class = 1

apoc.erase_classifier(cl_filename)
clf = apoc.ProbabilityMapper(opencl_filename=cl_filename, output_probability_of_class=output_probability_of_class)
clf.train(features, manual_annotations, image)

result = clf.predict(image=image)
imshow(result)

<matplotlib.image.AxesImage at 0x25fc58484c0>
../_images/9deedc1187c1b4a138f80d974a7cdf273a7cd6f6d5697cc8451d00fbfb3be8fd.png