:orphan:

Hyperspectral Segmentation
==========================

This gallery demonstrates the lithological segmentation workflow using EnMap hyperspectral data and Bayesian inference.

Gallery Overview
----------------

These examples cover the complete pipeline from raw hyperspectral cubes to georeferenced lithological maps with associated uncertainty. The workflow integrates remote sensing feature extraction with probabilistic classification.

Key Concepts
------------

* **Feature Extraction**: Dimensionality reduction (MNF) and extraction of geologically relevant spectral features (e.g., absorption depths).
* **Bayesian Segmentation**: Using spatial priors and spectral likelihoods to classify surface units.
* **Geological Integration**: Comparing surface segmentation results with 3D geological model predictions.
* **Uncertainty Quantification**: Estimating classification confidence and entropy across the scene.

Prerequisites
-------------

* Basic understanding of hyperspectral remote sensing.
* Familiarity with Bayesian classification concepts.
* Installed packages: ``rasterio``, ``scikit-image``, ``scipy``, in addition to the core Mineye dependencies.

Workflow Stages
---------------

1. **Preprocessing & Feature Extraction**: Preparing the hyperspectral data for classification.
2. **Segmentation**: Running the Bayesian engine to produce lithological labels.
3. **Validation**: Evaluating results against known geological structures and extracting points for further modeling.

.. note::
   The examples in this gallery require EnMap L2A products. Paths to data should be configured in ``mineye.config.paths`` or provided as environment variables as shown in the scripts.



.. raw:: html

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

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example demonstrates the workflow for lithological segmentation using EnMap hyperspectral data. It covers feature extraction, data preparation, and the Bayesian segmentation process.">

.. only:: html

  .. image:: /examples_segmentation/images/thumb/sphx_glr_01_enmap_lith_segmentation_thumb.png
    :alt:

  :ref:`sphx_glr_examples_segmentation_01_enmap_lith_segmentation.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">EnMap Lithological Segmentation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example demonstrates how to extract lithological information from EnMap segmentation results and sample it for use in 3D geological modeling.">

.. only:: html

  .. image:: /examples_segmentation/images/thumb/sphx_glr_02_enmap_data_extraction_thumb.png
    :alt:

  :ref:`sphx_glr_examples_segmentation_02_enmap_data_extraction.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">EnMap Data Extraction and Point Sampling</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example demonstrates how to compare a 3D geological model&#x27;s predictions with lithological labels extracted from EnMap hyperspectral data.">

.. only:: html

  .. image:: /examples_segmentation/images/thumb/sphx_glr_03_enmap_gempy_comparison_thumb.png
    :alt:

  :ref:`sphx_glr_examples_segmentation_03_enmap_gempy_comparison.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">EnMap Likelihood and Model Comparison</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /examples_segmentation/01_enmap_lith_segmentation
   /examples_segmentation/02_enmap_data_extraction
   /examples_segmentation/03_enmap_gempy_comparison



.. only:: html

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

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