Plug im!

Open access software for signal and image processing
(http://www.plugim.fr/)

IFPEN provides its internal signal and image processing platform for those who want the most advanced features combined with a simple and understandable user interface.
The developed plug-ins, directly linked to publications, as well as other digital operators, are detailed below. All of them are available under plug im!, as well as many other plug-ins for signals, images and/or volumes processing.

Note for readers

Below, some plug-ins available under plug im! are detailed; the plug-in is briefly defined with a help for its use. Plug-ins can be modified/updated for various reasons. The latest modifications will be indicated in News.

Do not hesitate to contact me for any question, remark or suggestion (CONTACT).

About Plug im! use

'Open' button:

'...' button:

Some definitions

Binary image:
An input representing a binary microstructure; either a binary 2D or 3D image, the white part (255) representing the microstructure to analyze, or a triphasic 2D or 3D image with specific intensities: 0 for the surrounding void, meaningless, 255 for the solid phase, and the rest is considered as the porous phase. The triphasic image allows to define a mask, distinguishing the surrounding void from the material, i.e. the sample of interest.
The triphasic image can be the output of the plug-in "Extraction of porosity". This option is originally added to consider real images acquired with a specific device, electron tomographic microscope. Nevertheless, real images obtained from any device can be considered, opening wide application perspectives.
In practice: the input, if it is a binary image, has to be biphasic or triphasic and has to be of type 'unsigned char', 'integer', 'float' or 'double'. 'RGB' images are not allowed. The data type can be seen at the top left of the Plug im! window.
A maximal volume is also imposed, which is equal to 1,000,000,000 voxels.

Skeletonization:
In most of the following operators, this is an optional step. We recommand the computation of a homotopic skeleton of the interconnected microstructure to analysis. Theoretically, it is the smallest subset of the microstructure keeping the connectivity degree; the topology is unchanged.
The skeleton of a microstructure can be considered as the  input binary image of plug-ins, as in the literature, for topological measures for instance, and some of our articles too.

The skeleton allows a consistent reduction in computation time.
Specific to by-iterative-erosions plug-ins: if the skeleton is considered ('Use skeleton'), the computation time of the morphological erosions is reduced too, as the valued skeleton is considered; the distance map value is assigned to each skeleton point. Moreover, in this case, the skeleton has to be load using the 'Open' button, and the microstructure using the '...' button.

In preparation

A-protocol:

...

Available

H-tortuosity-by-iterative-erosions (2D/3D)

Associated article: Chaniot J., Moreaud M., Sorbier L., Jeulin D., Becker J.-M. and Fournel T. (2020). "Heterogeneity assessment based on average variations of morphological tortuosity for complex porous structures characterization", Image Analysis & Stereology 39(1):111-128

Specifications:

  • H-tortuosity as seen by a spherical particle of given radius which increases step by step
  • linked to constrictivity, characterizing the bottleneck effect and/or the hindrance
Computation:

The steps defining the H-tortuosity are applied to the microstructure, which is eroded step by step. The iterative erosions using a unit sphere as structuring element, allow to consider a large interval of radii, representing the size of the hypothetic percolating particle.

Inputs:

  • binary image
  • optional: microstructure's skeleton

Outputs:

  • H-tortuosity values as a function of the radius (*.txt file)
  • map of the mean relative tortuosities (*.fda file)

Parameters:

  • sampling choice (1D-sampling or stratified-sampling)
  • maximal distance Dmax
  • number of random points N
  • boolean 'Use skeleton', when the microstructure's skeleton is considered as binary image. Additional parameter (if checked):
  1. "...": selection of the microstructure

H-tortuosity (2D/3D)

Associated article: Chaniot J., Moreaud M., Sorbier L., Jeulin D., Becker J.-M. and Fournel T. (2020). "Heterogeneity assessment based on average variations of morphological tortuosity for complex porous structures characterization", Image Analysis & Stereology 39(1):111-128

Specifications:
  • scalable topological descriptor; informations of different dimensions at distinct scales
  • random points sampling, handling disconnections and applicable on complex multi-scale microstructures, especially when entries and exits are delicate to impose

  • H-tortuosity values, named H-scalars, assessing the average variations of the peripheral morphological tortuosity
Computation:
  1. Optional step: skeletonization
  2. Sampling: definition of the set of N random points; 1D-sampling method (a), stratified-sampling method (b) (Fig.(a-b)). An additional sub-step is imposed in the case of biphasic images, avoiding boundary issues; a sampling area is defined to remove potentially problematic points (turquoise disk Fig.(c)).
  3. Distance transforms: At this step, the map of mean relative tortuosities and the maps associated to each points, if 'Save Data' checked, are computed.
  4. H-coefficients: set of mean morphological tortuosities, each one connected to a specific location, i.e. a random point, and a scale, i.e. a Euclidean distance (shades of purple Fig.(d)).
    A H-coefficient can be seen as either a viewpoint from this location at a specific distance, or the mean accessibility of this point for any other point located at this distance.
  5. H-scalars: set of viewpoints according to a distance d (from 1 to Dmax). Contrarily to the M-tortuosity, the H-tortuosity focuses on local features by quantifying them thanks to the local variations of the morphological tortuosity.
    A H-scalar, connected to a specific distance, is the mean tortuosity of a pair of random points with this distance as only constraint.

Input:

  • binary image

Outputs:

  • H-tortuosity values (*.txt file)
  • map of the mean relative tortuosities (*.fda file)
  • optional: maps associated to each starting point (*.fda file)

Parameters:

  • sampling choice (1D-sampling or stratified-sampling)
  • maximal distance Dmax (Fig.(d))
  • number of random points N
  • boolean 'Save Data', to save tortuosity maps associated to each starting point (*.fda file). Additional parameter (if checked):
  1. "...": selection of the save directory.


Deterministic M-tortuosity (2D/3D)

Associated article: Batista A.T.F., Baaziz W., Taleb A.-L., Chaniot J., Ersen O., Moreaud M., Legens C., Aguilar-Tapia A., Proux O., Hazemann J.-L., Diehl F., Chizallet C., Gay A.-S. and Raybaud P. (2020)."Atomic scale insight into the formation, size and location of platinum nanoparticles supported on γ-alumina", ACS Catalysis 10(7):4193–4204

Specifications:

  • M-tortuosity with an imposed set of starting points
  • image of objects with volume bigger than 1 voxel can be used to define the imposed set of starting points

Computation:
The steps are identical to the M-tortuosity, except for some pre-processing steps linked to the definition of the imposed set of starting points.
As objects can be used, real particles can be considered to define the starting locations. Nevertheless, they have to be reduced to a unique voxel, their center of mass. Due to some approximations, some starting points may not belong to the microstructure. The orthogonal projection can fix this issue, computing the closest feature point to each starting point outside to the microstructure.

'Center of Mass' and 'Orth. Projection' can be used combined; first center of mass computation then orthogonal projection, or separately, center of mass or orthogonal projection.

'Maps' and 'Paths' can be used combined; the geodesic paths are displayed over the morphological tortuosities maps, or separately; the morphological tortuosities maps or the geodesic paths.

Inputs:

    • binary image
    • imposed set of starting points

    Outputs:

    • deterministic M-tortuosity value (*.txt file)
    • set of M-coefficients (*.txt file)
    • map of the mean relative tortuosities (*.fda file)
    • optional: "Euclidean distance/geodesic distance/ morphological tortuosity" for each points pair (*.txt file)
    • optional: maps associated to each starting point (*.fda file)

    Parameters:

    • '...': selection of the imposed set of starting points (*.txt OR an image *.tif or *.fda)
    • boolean 'Image', if an image is used for the starting points. Additional parameters (if checked):
    1. boolean 'Center of Mass': computation of the center of mass of each object
    2. boolean 'Orth. Projection': orthogonal projection of each point on the microstructure
    • boolean 'Save Dist/Tor' (optional output): save information for each points pair; distances and tortuosity (*.txt file)
    • boolean 'Save Data' (optional output): maps associated to each starting point (*.fda file). Additional parameters (if checked):
    1. 'Maps': morphological tortuosities map of each point
    2. 'Paths': geodesic paths associated to each point
    3. '...': selection of the save directory

    M-tortuosity-by-iterative-erosions (2D/3D)

    Associated article: Chaniot J., Moreaud M., Sorbier L., Fournel T. and Becker J.-M. (2019). "Tortuosimetric operator for complex porous media characterization", Image Analysis & Stereology 38(1) :25-41

    Specifications:

    • M-tortuosity as seen by a spherical particle of given radius which increases step by step
    • linked to constrictivity, characterizing the bottleneck effect and/or the hindrance
    Computation:

    The steps defining the M-tortuosity are applied to the microstructure, which is eroded step by step. The iterative erosions using a unit sphere as structuring element, allow to consider a large interval of radii, representing the size of the hypothetic percolating particle.

    Input:

    • binary image
    • optional: microstructure's skeleton

    Outputs:

    • M-tortuosity values as a function of the radius (*.txt file)
    • sets of M-coefficients as a function of the radius (*.txt file)
    • map of the mean relative tortuosities (*.fda file)

    Parameters:

    • sampling choice (1D-sampling or stratified-sampling)
    • number of random points N
    • boolean 'Use skeleton', when the microstructure's skeleton is considered as binary image. Additional parameter (if checked):
    1. "...": selection of the microstructure

    M-tortuosity (2D/3D)

    Associated article: Chaniot J., Moreaud M., Sorbier L., Fournel T. and Becker J.-M. (2019). "Tortuosimetric operator for complex porous media characterization", Image Analysis & Stereology 38(1) :25-41

    Specifications:
    • scalable topological descriptor; informations of different dimensions
    • random points sampling, handling disconnections and applicable on complex multi-scale microstructures, especially when entries and exits are delicate to impose

    • the M-tortuosity is translation, rotation and homothety invariant. Moreover, it is stable for periodic microstructures.
    Computation:
    1. Optional step: skeletonization
    2. Sampling: definition of the set of N random points; 1D-sampling method (a), stratified-sampling method (b).  Fig.(a-b) displays the difference between the two sampling methods.
    3. Distance transforms: At this step, the map of mean relative tortuosities and the maps associated to each points, if 'Save Data' checked, are computed.
    4. M-coefficients: set of mean morphological tortuosities, each one connected to a specific location, i.e. a random point.
      A M-coefficient can be seen as either a viewpoint of the microstructure from this location, or the mean accessibility of this point for any random point.
    5. M-scalar: The M-tortuosity value, named the M-scalar, is an assessment of the morphological tortuosity. It aggregates information of each M-coefficient into a scalar value.
      The M-scalar is a global viewpoint of the microstructure, or the mean accessibility of any random point for any other random point.

    Input:

    • binary image

    Outputs:

    • M-tortuosity value  (*.txt file)
    • set of M-coefficients (*.txt file)
    • map of the mean relative tortuosities (*.fda file)
    • optional: "Euclidean distance/geodesic distance/ morphological tortuosity" for each points pair (*.txt file)
    • optional: maps associated to each starting point (*.fda file)

    Parameters:

    • sampling choice (1D-sampling or stratified-sampling)
    • number of random points N
    • boolean 'Save Dist/Tor' (optional output): save information for each points pair; distances and tortuosity (*.txt file)
    • boolean 'Save Data' (optional output): maps associated to each starting point (*.fda file) . Additional parameters (if checked):
    1. 'Maps': morphological tortuosities map of each point
    2. 'Paths': geodesic paths associated to each point
    3. '...': selection of the save directory

     .