:orphan:

Tutorials
=========

Here you can find a series of Tutorials that will guide you through the use
of Spyrit. It is recommended to follow them in order.

For each tutorial, please download the corresponding Python Script (*.py*) or
Jupyter notebook (*.ipynb*) file at the bottom of the page. The images used in
these tutorials can be found on `this page`_ of the Spyrit GitHub.

Below is a diagram of the entire image processing pipeline. Each tutorial
focuseson a specific part of the pipeline.

* :ref:`Tutorial 1 <sphx_glr_gallery_tuto_01_acquisition_operators.py>` focuses on the measurement operators, with or without noise

* :ref:`Tutorial 2 <sphx_glr_gallery_tuto_02_pseudoinverse_linear.py>` explains the pseudo-inverse reconstruction process from the (possibly noisy) measurements

* :ref:`Tutorial 3 <sphx_glr_gallery_tuto_03_pseudoinverse_cnn_linear.py>` uses a CNN to denoise the image if necessary

* :ref:`Tutorial 4 <sphx_glr_gallery_tuto_04_train_pseudoinverse_cnn_linear.py>` is used to train the CNN introduced in Tutorial 3

* :ref:`Tutorial 5 <sphx_glr_gallery_tuto_05_acquisition_split_measurements.py>` introduces a new type of measurement operator ('split') that simulates positive and negative measurements

* :ref:`Tutorial 6 <sphx_glr_gallery_tuto_06_dcnet_split_measurements.py>` uses a Denoised Completion Network with a trainable image denoiser to improve the results obtained in Tutorial 5

* :ref:`Tutorial 7 <sphx_glr_gallery_tuto_07_drunet_split_measurements.py>` shows how to perform image reconstruction using a pretrained plug-and-play denoising network.

* :ref:`Tutorial 8 <sphx_glr_gallery_tuto_08_lpgd_split_measurements.py>` shows how to perform image reconstruction using a learnt proximal gradient descent (AVAILABLE SOON).

* :ref:`Tutorial 9 <sphx_glr_gallery_tuto_09_dynamic.py>` explains motion simulation from an image, dynamic measurements and reconstruction.

* Explore :ref:`Bonus Tutorial <sphx_glr_gallery_tuto_bonus_advanced_methods_colab.py>` if you want to go deeper into Spyrit's capabilities


.. image:: ../fig/full.png
   :width: 800
   :align: center
   :alt: Principle of the image processing pipeline

|
|


.. _this page: https://github.com/openspyrit/spyrit/tree/3895b5e61fb6d522cff5e8b32a36da89b807b081/tutorial/images/test



.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial shows how to simulate measurements using the spyrit.core submodule. The simulation is based on three modules:">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_01_acquisition_operators_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_01_acquisition_operators.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">01. Acquisition operators</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial shows how to simulate measurements and perform image reconstruction. The measurement operator is chosen as a Hadamard matrix with positive coefficients. Note that this matrix can be replaced by any desired matrix.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_02_pseudoinverse_linear_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_02_pseudoinverse_linear.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">02. Pseudoinverse solution from linear measurements</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial shows how to simulate measurements and perform image reconstruction using PinvNet (pseudoinverse linear network) with CNN denoising as a last layer. This tutorial is a continuation of the Pseudoinverse solution tutorial &lt;tuto_pseudoinverse_linear&gt; but uses a CNN denoiser instead of the identity operator in order to remove artefacts.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_03_pseudoinverse_cnn_linear_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_03_pseudoinverse_cnn_linear.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">03. Pseudoinverse solution + CNN denoising</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial shows how to train PinvNet with a CNN denoiser for reconstruction of linear measurements (results shown in the previous tutorial &lt;tuto_pseudoinverse_cnn_linear&gt;). As an example, we use a small CNN, which can be replaced by any other network, for example Unet. Training is performed on the STL-10 dataset.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_04_train_pseudoinverse_cnn_linear_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_04_train_pseudoinverse_cnn_linear.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">04. Train pseudoinverse solution + CNN denoising</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial is an extension of the Tutorials tuto_acquisition_operators and tuto_pseudoinverse_linear where:">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_05_acquisition_split_measurements_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_05_acquisition_split_measurements.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">05. Acquisition operators (advanced) - Split measurements and subsampling</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="06. Denoised Completion Network (DCNet)">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_06_dcnet_split_measurements_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_06_dcnet_split_measurements.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">06. Denoised Completion Network (DCNet)</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial shows how to perform image reconstruction using a DCNet (data completion network) that includes a DRUNet denoiser. DRUNet is a pretrained plug-and-play denoising network that has been pretrained for a wide range of noise levels. DRUNet admits the noise level as an input. Contratry to the DCNet described in Tutorial 6 &lt;tuto_dcnet_split_measurements&gt;, it requires no training.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_07_drunet_split_measurements_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_07_drunet_split_measurements.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">07. DCNet with plug-and-play DRUNet denoising</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial shows how to perform image reconstruction with unrolled Learned Proximal Gradient Descent (LPGD) for split measurements.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_08_lpgd_split_measurements_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_08_lpgd_split_measurements.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">08. Learned proximal gradient descent (LPGD) for split measurements</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This tutorial explains how to reconstruct a motion-compensated image from a dynamic scene.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_09_dynamic_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_09_dynamic.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">09. Acquisition and reconstruction of dynamic scenes</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="We refer to spyrit-examples/tutorial for a list of tutorials that can be run directly in colab and present more advanced cases than the main spyrit tutorials, such as comparison of methods for split measurements, or comparison of different denoising networks.">

.. only:: html

  .. image:: /gallery/images/thumb/sphx_glr_tuto_bonus_advanced_methods_colab_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_tuto_bonus_advanced_methods_colab.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Bonus. Advanced methods - Colab</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /gallery/tuto_01_acquisition_operators
   /gallery/tuto_02_pseudoinverse_linear
   /gallery/tuto_03_pseudoinverse_cnn_linear
   /gallery/tuto_04_train_pseudoinverse_cnn_linear
   /gallery/tuto_05_acquisition_split_measurements
   /gallery/tuto_06_dcnet_split_measurements
   /gallery/tuto_07_drunet_split_measurements
   /gallery/tuto_08_lpgd_split_measurements
   /gallery/tuto_09_dynamic
   /gallery/tuto_bonus_advanced_methods_colab



.. only:: html

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

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