Automatically shrink reconstruction volume by hrobarts · Pull Request #2221 · TomographicImaging/CIL

hrobarts · 2025-10-10T15:23:22Z

Changes

Code to create a reduced reconstruction volume from an AcquisitionData object
If using method='manual'

Creates an image geometry with limits specified with the limits argument. If a dimension is not included in the limits dictionary, the full size of the axis is used
If using method='otsu or threshold
Performs a binned reconstruction
Finds the maximum pixel intensity along each direction
Finds pixels above the threshold or a threshold found with an Otsu filter
Identified limits between sample and background where no pixels above the threshold are found
If preview=True
Plots the limits on the maximum intensity reconstruction
kwargs
buffer applies a buffer around the automatically calculated limits
mask_radius applies a mask to the reconstruction before finding limits
otsu_classes changes number of material classes to assume in the Otsu thresholding
min_component_size specified a minimum number of connected components in pixels to consider when choosing the limits - to avoid a small amount of noise changing the limits

vs = VolumeShrinker(data, recon_backend='astra')
ig_reduced = vs.run(limits={'horizontal_x':(10, 150), 'vertical':(5, 50)}, preview=True)

vs = VolumeShrinker(data, recon_backend='astra')
ig_reduced = vs.run(method='threshold', preview=True, threshold=0.9, buffer=10)

If debug logging is enabled, we plot a histogram with the threshold and reduced box

vs = VolumeShrinker(data, recon_backend='astra')
ig_reduced = vs.run(method='otsu', preview=True, otsu_classes=2, buffer=10, min_component_size=10)

Testing you performed

Demo TomographicImaging/CIL-Demos#283

I think it makes sense for this code to be in utilities and possibly be called by AcquisitionData.get_ImageGeometry() however when I did this I had trouble with circular imports. Maybe it should be moved to utilities under framework?

Related issues/links

May close #1998

Checklist

I have performed a self-review of my code
I have added docstrings in line with the guidance in the developer guide
I have updated the relevant documentation
I have implemented unit tests that cover any new or modified functionality
CHANGELOG.md has been updated with any functionality change
Request review from all relevant developers

hrobarts · 2025-12-02T13:50:21Z

For discussion:

Could this be a method on image_data

The user would have to reconstruct their data themselves before calling the get reduced image geometry method
Once they have reconstructed the full thing they might not be interested in reducing the volume
If they bin the data they will have to convert the binned reduced geometry to their unbinned reduced geometry

Could this be a method on acquisition_geometry

The method needs access to the data so would have to pass the data as an argument?
Similar to the ag.get_ImageGeometry() method we could have ag.get_reduced_ImageGeometry(data)

Could this be a method on acqusition_data

Makes the most sense, but confusing get_ImageGeometry is a method on acquisition_geometry and this would be a method on acqusition_data. Could allow both?

hrobarts · 2025-12-02T16:17:22Z

What should the name be

get_reduced_ImageGeometry()
get_optimal_ImageGeometry() - makes more sense if we also do rotation at some point
get_shrunk_ImageGeometry()
or
get_ImageGeometry() with argument method=auto (or rotate in future?) or enums like in labels

…o shrink_volume

lauramurgatroyd

Hi @hrobarts
The shrinker tool is looking very useful!
I had some questions about the structure, plus some docstring and unit test suggestions, and maybe some more checking of valid parameters. Also wasn't sure about the mask parameter in the 'manual' case

Wrappers/Python/cil/utilities/shrink_volume.py

lauramurgatroyd · 2026-02-26T10:26:21Z

Wrappers/Python/cil/utilities/shrink_volume.py

+                    'vertical':(bounds['vertical'][0], bounds['vertical'][1], 1)})(ig_unbinned)
+        return ig
+
+    def get_recon(self, mask_radius):


Would be great to have docstrings on all the methods

Wrappers/Python/cil/utilities/shrink_volume.py

lauramurgatroyd · 2026-02-26T10:35:32Z

Wrappers/Python/cil/utilities/shrink_volume.py

+    recon_backend : {'tigre', 'astra'}
+        The backend to use for the reconstruction


Should mention that it uses the cil plugin, not the recon class.
Is this because the recon class doesn't yet support astra FDK?

Yes I guess, this is just the same functionality as the COR corrector

Wrappers/Python/test/test_shrink_volume.py

lauramurgatroyd · 2026-02-26T11:27:39Z

Wrappers/Python/test/test_shrink_volume.py

+
+
+class TestVolumeShrinker(unittest.TestCase):
+


would be good to have a test with the mask parameter used

What kind of test are you thinking? Testing the impact of the mask on the reduce_reconstruction_volume function? At the moment I'm only testing this on a very small test recon and I'm not sure how to make a mask predictably change the behaviour. I don't think this is something I have time to do but happy to add to a to-do list!

Or do you mean test the kwargs are correctly initialised when we pass them to run?

yes I meant the impact on the reduce_reconstruction_volume

I guess if we had a circular mask that we know makes the region smaller than the automatic region on each dimension

lauramurgatroyd · 2026-02-26T11:32:15Z

Wrappers/Python/test/test_shrink_volume.py

+
+        bounds = vs.reduce_reconstruction_volume(self.test_recon, binning=1, method='threshold', kwargs={'threshold':0.1})
+        for dim in ['horizontal_x', 'horizontal_y', 'vertical']:
+            self.assertEqual(bounds[dim], (2,7))


probably good to have a test where they aren't the same in each dimension? If not too difficult to achieve

I'm added an asymmetrical volume

lauramurgatroyd · 2026-02-26T13:28:55Z

Wrappers/Python/cil/utilities/shrink_volume.py

+                fig, axes = plt.subplots(nrows=1, ncols=3, figsize=(8, 2.5))
+
+                axes[0].imshow(arr, cmap=plt.cm.gray)
+                axes[0].set_title('Original')


Might be useful to add to the titles the direction and that its the max value (like in the other plots)

Wrappers/Python/cil/utilities/shrink_volume.py

lauramurgatroyd · 2026-02-26T14:02:25Z

You will need to add the volume shrinker to this file in the documentation:
https://github.com/TomographicImaging/CIL/blob/master/docs/source/utilities.rst

Co-authored-by: Laura Murgatroyd <60604372+lauramurgatroyd@users.noreply.github.com> Signed-off-by: Hannah Robarts <77114597+hrobarts@users.noreply.github.com>

hrobarts

Hi @lauramurgatroyd thank you for your review! I think I've addressed all the comments now.

Wrappers/Python/cil/utilities/shrink_volume.py

hrobarts · 2026-02-27T09:01:23Z

Wrappers/Python/cil/utilities/shrink_volume.py

+            If using 'threshold' method, specify the intensity threshold
+            between sample and background. Default is None.


Yes it fails confusingly, I've added a check now!

hrobarts · 2026-02-27T09:07:47Z

Wrappers/Python/cil/utilities/shrink_volume.py

+        mask_radius: float, optional
+            Radius of circular mask to apply on the reconstructed volume, before
+            automatically cropping the reconstruction volume. Default is None.


Have updated to say it's used for the cropping and in preview

hrobarts · 2026-02-27T09:08:02Z

Wrappers/Python/cil/utilities/shrink_volume.py

+            if method.lower() == 'manual':
+                mask_radius = kwargs.pop('mask_radius', None)
+                recon, binning = self.get_recon(mask_radius=mask_radius)


I think it's useful for the preview as well. If you had data that created a region of interest sample with a ring in the recon then it might be useful to use a mask to check how your manual limits look on the volume. I've updated the docstring to say the mask is also used for preview.

hrobarts · 2026-02-27T11:58:55Z

Wrappers/Python/test/test_shrink_volume.py

+
+
+class TestVolumeShrinker(unittest.TestCase):
+


What kind of test are you thinking? Testing the impact of the mask on the reduce_reconstruction_volume function? At the moment I'm only testing this on a very small test recon and I'm not sure how to make a mask predictably change the behaviour. I don't think this is something I have time to do but happy to add to a to-do list!

hrobarts · 2026-02-27T11:59:41Z

Wrappers/Python/test/test_shrink_volume.py

+
+
+class TestVolumeShrinker(unittest.TestCase):
+


Or do you mean test the kwargs are correctly initialised when we pass them to run?

Wrappers/Python/test/test_shrink_volume.py

hrobarts · 2026-02-27T12:14:06Z

Wrappers/Python/cil/utilities/shrink_volume.py

+    recon_backend : {'tigre', 'astra'}
+        The backend to use for the reconstruction


Yes I guess, this is just the same functionality as the COR corrector

Wrappers/Python/cil/utilities/shrink_volume.py

lauramurgatroyd · 2026-02-27T14:02:11Z

Wrappers/Python/cil/utilities/shrink_volume.py

+        >>> cil_log_level = logging.getLogger()
+        >>> cil_log_level.setLevel(logging.DEBUG)
+        >>> vs = VolumeShrinker(data, recon_backend='astra')
+        >>> ig_reduced = vs.run(method='threshold' threshold=0.9)


missing a ','

lauramurgatroyd · 2026-02-27T14:05:27Z

Wrappers/Python/test/test_shrink_volume.py

+
+
+class TestVolumeShrinker(unittest.TestCase):
+


yes I meant the impact on the reduce_reconstruction_volume

lauramurgatroyd · 2026-02-27T14:16:28Z

Wrappers/Python/cil/utilities/shrink_volume.py

+        mask_radius: float, optional
+            Radius of circular mask to apply on the reconstructed volume, before
+            automatically cropping the reconstruction volume or displaying with 
+            preview. Default is None.


I think it should say here or under method that the mask_radius is not used except for the preview, for the manual case

lauramurgatroyd · 2026-02-27T14:17:24Z

Wrappers/Python/test/test_shrink_volume.py

+
+
+class TestVolumeShrinker(unittest.TestCase):
+


I guess if we had a circular mask that we know makes the region smaller than the automatic region on each dimension

hrobarts added 2 commits October 10, 2025 15:11

First draft shrink volume code

33584af

Make mask optional

10f42b7

hrobarts linked an issue Oct 29, 2025 that may be closed by this pull request

Utility to find a reduced reconstruction volume #2234

Closed

hrobarts removed a link to an issue Oct 30, 2025

Utility to find a reduced reconstruction volume #2234

Closed

hrobarts self-assigned this Nov 3, 2025

hrobarts and others added 6 commits December 9, 2025 09:54

Update logic if auto=False

6dcec8b

Merge branch 'master' into shrink_volume

cb6e148

Merge branch 'shrink_volume' of github.com:TomographicImaging/CIL int…

7c52da8

…o shrink_volume

Merge branch 'master' into shrink_volume

d366be1

Merge branch 'shrink_volume' of github.com:TomographicImaging/CIL int…

65fdfdc

…o shrink_volume

Rename functions

69822bc

gfardell requested a review from Neonbluestoplight February 10, 2026 13:56

hrobarts and others added 3 commits February 20, 2026 11:36

Merge branch 'master' into shrink_volume

41f926b

Add tests

e15f3db

Add colorbar in preview plots

e154b8b

hrobarts marked this pull request as ready for review February 25, 2026 09:04

hrobarts requested a review from lauramurgatroyd February 25, 2026 14:30

Update CHANGELOG

edc9b44

lauramurgatroyd requested changes Feb 26, 2026

View reviewed changes

hrobarts and others added 4 commits February 27, 2026 08:34

Merge branch 'master' into shrink_volume

ec69701

Apply suggestions from code review

83002f6

Co-authored-by: Laura Murgatroyd <60604372+lauramurgatroyd@users.noreply.github.com> Signed-off-by: Hannah Robarts <77114597+hrobarts@users.noreply.github.com>

Updates from review

6c32294

Add volumeshrinker to rst file

ea98fcb

hrobarts commented Feb 27, 2026

View reviewed changes

Fix docs

83df471

hrobarts requested a review from lauramurgatroyd February 27, 2026 13:51

lauramurgatroyd reviewed Feb 27, 2026

View reviewed changes

		recon_backend : {'tigre', 'astra'}
		The backend to use for the reconstruction

		If using 'threshold' method, specify the intensity threshold
		between sample and background. Default is None.

Conversation

hrobarts commented Oct 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes

Testing you performed

Related issues/links

Checklist

Uh oh!

hrobarts commented Dec 2, 2025

Uh oh!

hrobarts commented Dec 2, 2025

Uh oh!

lauramurgatroyd left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

lauramurgatroyd commented Feb 26, 2026

Uh oh!

hrobarts left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

hrobarts commented Oct 10, 2025 •

edited

Loading