.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "examples/morpho/example_watershed.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_examples_morpho_example_watershed.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_examples_morpho_example_watershed.py:


Watershed Segmentation Example
==============================

This example shows how to use the watershed on an image to make a segmentation.

The watershed segmentation is based on the usage of an image as a landscape, where each minimum
is a basin of a lake. It simulates the filling of these basins and where two basins meet, it make a
watershed line.

Meyer's watershed from local minima
***********************************

First, we import the modules needed to use it

.. GENERATED FROM PYTHON SOURCE LINES 16-22

.. code-block:: Python


    from pylena import morpho
    from skimage.data import camera
    import matplotlib.pyplot as plt
    import numpy as np








.. GENERATED FROM PYTHON SOURCE LINES 23-24

We first get an image

.. GENERATED FROM PYTHON SOURCE LINES 24-26

.. code-block:: Python

    img = camera()








.. GENERATED FROM PYTHON SOURCE LINES 27-28

We compute the gradient

.. GENERATED FROM PYTHON SOURCE LINES 28-38

.. code-block:: Python


    se = morpho.make_structuring_element_2d("disc", 2)
    grad = morpho.gradient(img, se)

    plt.figure(figsize=(8, 8))
    plt.imshow(grad, cmap="gray")
    plt.xticks([])
    plt.yticks([])
    plt.show()




.. image-sg:: /examples/morpho/images/sphx_glr_example_watershed_001.png
   :alt: example watershed
   :srcset: /examples/morpho/images/sphx_glr_example_watershed_001.png
   :class: sphx-glr-single-img





.. GENERATED FROM PYTHON SOURCE LINES 39-40

Then, we compute the watershed on this gradient.

.. GENERATED FROM PYTHON SOURCE LINES 40-44

.. code-block:: Python


    ws, nlabel = morpho.watershed(img, 8)
    print(f"Number of label: {nlabel}")





.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    Number of label: 13563




.. GENERATED FROM PYTHON SOURCE LINES 45-46

We finally illustrate the resulting segmentation

.. GENERATED FROM PYTHON SOURCE LINES 46-58

.. code-block:: Python


    np.random.seed(0)
    color = np.random.randint(low=0, high=255, size=(np.max(ws) + 1, 3))
    color[0, :] = 0  # Watershed lines are illustrated in black
    colorized = color[ws]

    plt.figure(figsize=(8, 8))
    plt.imshow(colorized)
    plt.xticks([])
    plt.yticks([])
    plt.show()




.. image-sg:: /examples/morpho/images/sphx_glr_example_watershed_002.png
   :alt: example watershed
   :srcset: /examples/morpho/images/sphx_glr_example_watershed_002.png
   :class: sphx-glr-single-img





.. GENERATED FROM PYTHON SOURCE LINES 59-67

The disadvantage of the watershed from local minima is the fact it over-segments the image.
To solve this issue, it may be useful to provide some markers from which the watershed will
start the flooding

Watershed segmentation from user markers
****************************************

First, we provide the input markers.

.. GENERATED FROM PYTHON SOURCE LINES 68-74

.. code-block:: Python


    markers = np.zeros(img.shape, dtype=np.int16)
    markers[200:400, 100] = 1
    markers[250:480, 450] = 1
    markers[50, 50:450] = 1








.. GENERATED FROM PYTHON SOURCE LINES 75-76

Then, we compute the watershed from markers.

.. GENERATED FROM PYTHON SOURCE LINES 76-79

.. code-block:: Python


    ws2, nlabel2 = morpho.watershed(img, 8, markers=markers)








.. GENERATED FROM PYTHON SOURCE LINES 80-81

Finally we illustrate

.. GENERATED FROM PYTHON SOURCE LINES 81-99

.. code-block:: Python


    color2 = np.random.randint(low=0, high=255, size=(np.max(ws2) + 1, 3))
    color2[0, :] = 0  # Watershed lines are illustrated in black
    colorized2 = color2[ws2]

    markers = morpho.dilation(markers.astype(np.uint8), se)
    markers_illustration = np.zeros((img.shape[0], img.shape[1], 3), dtype=np.uint8)
    markers_illustration[markers > 0, 0] = 255

    plt.figure(figsize=(12, 6))
    plt.subplot(121)
    plt.title("Markers")
    plt.imshow(markers_illustration)
    plt.subplot(122)
    plt.title("Watershed by markers")
    plt.imshow(colorized2)
    plt.tight_layout()
    plt.show()



.. image-sg:: /examples/morpho/images/sphx_glr_example_watershed_003.png
   :alt: Markers, Watershed by markers
   :srcset: /examples/morpho/images/sphx_glr_example_watershed_003.png
   :class: sphx-glr-single-img






.. rst-class:: sphx-glr-timing

   **Total running time of the script:** (0 minutes 0.565 seconds)


.. _sphx_glr_download_examples_morpho_example_watershed.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: example_watershed.ipynb <example_watershed.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: example_watershed.py <example_watershed.py>`

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: example_watershed.zip <example_watershed.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_