Microscopy Image Browser Graphcut segmentation

This window give access to semi-automated image segmentation using the maxflow/mincut graphcut method.

The Graph cut segmentation is based on Max-flow/min-cut algorithm written by Yuri Boykov and Vladimir Kolmogorov and implemented for MATLAB by Michael Rubinstein. The max-flow/min-cut algorithm is applied not to individual pixels but to groups of pixels (superpixels (2D), or supervoxels(3D)) that may be generated either using the SLIC algorithm written by Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurelien Lucchi, Pascal Fua, and Sabine S?sstrunk or by the Waterhed algorithm. The objects that have intensity contrast are best described with the SLIC superpixels, while the objects that have distinct boundaries with the Watershed superpixels. Utilization of superpixels requires some time to calculate them but pays off during the following segmentation.

Back to Index --> User Guide --> Menu --> Tools Menu

collapse all on page

General example
Mode panel
Subarea panel
Calculation of superpixels/supervoxels
Image segmentation settings
Image segmentation example
References

collapse all

General example

A demonstration of the Graphcut segmentation is available in the following video:

https://youtu.be/dMeoIZPaDS4

How to use

How to use:

Use two labels to mark areas that belong to background and the objects of interest
Start the Graphcut segmentation tool: Menu->Tools->Semi-automatic segmentation Graphcut
Set one of the modes: 2D/3D
Define type of superpixels/supervoxels: SLIC, or Watershed
Generate superpixels/supervoxels (Press the Superpixels/Graph button)
Check the size of the generated superpixels and modify the size if needed
Press the Segment button to start segmentation

Note! some functions have to be compiled, please check the System Requirements page for details.

Mode panel

The Mode panel offers possibility to select a desired working mode for the segmentation.

◉ 2D, current slice only, performs segmentation on the slice that is currently shown in the Image View panel
◉ 2D, slice-by-slice, performs 2D segmentation for each slice of the dataset individually
◉ 3D, volume, performs 3D segmentation for complete or selected portion (see Selected Area section below) of the dataset
◉ 3D, volume, grid, a special mode of 3D graphcut, where the dataset is chopped into several subvolumes (defined by Chop edit boxes, see below) and the dataset which is centered at the Image View panel is gets segmented (for convenience, turn on the marker of the center point, toolbar->center marker button ). Chopping of large volume into several small subvolumes (e.g.400x400x400 pixels allows effective interactive segmentation of this large volume To segment all subvolumes press the Segment All button

Subarea panel

The Subarea panel allows selection of the sub-area of the dataset for processing. If dataset is too big it can be processed in parts or binned using this panel.

X:... defines the width of the dataset to process. Please use two numbers separated by a colon sign (:)
Y:... defines the height of the dataset to process
Z:... defines the z-slices of the dataset to process
from Selection button populates the X:, Y:, Z: fields using coordinates of a bounding box that describes the Selection layer
Current View button limits the X: and Y: parameters to the image that is currently displayed in the Image View panel
Reset resets the Subarea fields to the dimensions of the dataset
Bin x times... defines a binning factor for the data before segmentation. It allows to perform faster but with less details.
Attention! The auto update mode ([✓] auto update) is not available for the binned datasets!

collapse all

Calculation of superpixels/supervoxels

Before the segmentation, the pixels of the opened dataset should be clustered using the SLIC or Watershed algorithms. The picture below shows comparison between two types of superpixels. The upper panels show the SLIC superpixels that were good to segment a dark lipid droplet that has a good intensity contrast. The Watershed superpixels gave better segmentation of objects that were surrounded with boundaries.

Image example of clusters

Superpixels ▼, define type of superpixels to use
Size of superpixels..., defines approximate size of superpixels (SLIC only)
Compactness..., a number between 1 and 99 that defines how square the superpixels should be; for example, 99 - results in quite square superpixels (SLIC only)
Reduce number of superpixels..., a factor that is allowing to increase size of superpixels, the larger numbers results in larger superpixels (Watershed only)
Color channel ▼, index of the color channel to be used for calculation of superpixels
Type of signal ▼, use black-on-white for electron microscopy and light-on black for light microscopy
The Chop...editboxes, allow to chop the dataset into smaller subvolumes for the ◉ 3D volume grid mode or for calculation of the SLIC supervoxels
The [✓] autosave checkbox - when enabled, the resulting graphcut structure with the generated supervoxels is automatically saved to disk for future use
The [✓] parfor checkbox - when enabled, the Watershed clustering for the 3D volume grid mode is calculated using the parallel processing, which improves calculation performance in several times
[✓] use PixelIdsList - when enabled, the generation of final models is based on detected indices of supervoxels. This mode may improve performance in some situations, but requires more memory % Recalculate Graph - allows to recalculate the graph using a new coefficient (Coef). In general, the larger coefficients give stronger growth from the seeds. Sometimes, however, the large coefficients results in segmentation of areas that are distant from the seeds, which is considered as an artefact of the method.
Superpixels/Graph press this button to initiate generation of superpixels and their final organization into a graph
Import press to import superpixels and the generated graph from a disk or MATLAB
Export press to export superpixels and the generated graph to a file, MATLAB, a new model or as Lines3D graph object (not recommended for many superpixels, see also here https://youtu.be/xrsTVqD7kOQ)
Preview superpixels the generated superpixels may be previewed by pressing this button

Image segmentation settings

Both the Watershed and Graphcut workflows use provided labels that mark areas belonging to the Object and Background to perform the fine segmentation. Comparing to the Graphcut workflow, the Watershed workflow is a bit less interactive; it requires more time for the each execution and separates only objects that have distinct boundaries, for example membrane enclosed organelles.

On the other hand, the Graphcut workflow spends more time on the image preprocessing (calculation of the superpixels and generation of a graph) but each following interaction is fast. Using this workflow it is possible to separate objects that have both boundaries and intensity contrast. In general the Graphcut workflow is recommended for most of the cases.

Below, description of the Image segmentation settings:

Background ▼ defines a material of the model that labels the background areas
Object ▼ defines a material of the model that labels the object to be segmented
Update lists refreshes the lists of materials
[✓] Auto update - enables auto update of the segmentation results each time when material is modified. It is mostly useful for relatively small datasets (~400x400x400 pixels).
Important: please do not use the ⇧ Shift+A key shortcut, but only A shortcut. Also, when this mode is used it is recommended to recalculate the final segmentation by pressing the Segment button. Also the auto update mode is not available if the Bin mode is used.

Image segmentation example

	Load a sample dataset: Menu->File->Import image from->URL, enter the address: http://mib.helsinki.fi/tutorials/WatershedDemo/watershed_demo1.tif Press the + button in the Segmentation panel to add material to the model and name is as 'Background' (use the right mouse button to call a popup menu) Use the brush tool to label an area that belongs to cytoplasm

Press the A button to add selected area to the first material (Background) of the model Press the + button again to add another material and name it as 'Seeds' Draw labels inside mitochondria.

Press the A button to add selected area to the second material (Seeds) of the model Start the Graphcut segmentation tool: Menu->Tools->Semi-automatic segmentation->Graphcut. Select the Watershed ▼ type of superpixels Make sure that the proper materials are selected for both Background and Object in the Image segmentation settings Press the Segment button to segment mitochondria Add more seeds to the background and object materials to improve segmentaion Press the Segment button again of use the [✓] Auto update mode for instant update of the segmentation results The segmented mitochondria are placed to the Mask layer Optionally smooth mitochondria: Menu->Mask->Smooth Mask

References

Graph Cut:

Max-flow/min-cut algorithm written by Yuri Boykov and Vladimir Kolmogorov (Please note that this algorithm is licensed only for research purposes).
MATLAB wrapper for maxflow is written by Michael Rubinstein.
SLIC superpixels and supervoxels by Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurelien Lucchi, Pascal Fua, and Sabine S?sstrunk.
Region Adjacency Graph (RAG) and its modification for watershed was written by David Legland, INRA, France, 2013-2015 and was used in calculation of adjusent superpixels