Method for compressed sensing and processing of image data

Patent metadata

Field	Value
Publication number	US-12430720-B2
Application number	US-202318155289-A
Country	US
Kind code	B2
Filing date	Jan 17, 2023
Priority date	Jan 18, 2022
Publication date	Sep 30, 2025
Grant date	Sep 30, 2025

Abstract

A method can be used for sensing and processing image data for an object to be imaged. The object is scanned incompletely by virtue of regions (eB) of the object being sensed, where the sensed image regions (eB) alternate with non-sensed image regions (neB) of the object. Image data (rBD) of the non-sensed image regions (neB) are reconstructed on the basis of the sensed image data (eBD) of the sensed image regions (eB). A noise signal (N) of the sensed image data (eBD) of the sensed regions (eB) is ascertained and transferred to the reconstructed image data (rBD) of the non-sensed regions (neB), so that a user obtains a homogeneous visual impression in relation to the noise arising in the overall image data of the object visualized in a resultant overall image (rGB Inv ).

First claim

The invention claimed is:

1. A method for sensing and processing image data, the method comprising:
incompletely scanning an object to be imaged, wherein incomplete scanning results in sensed image regions (eB) of the object alternating with non-sensed image regions (neB) of the object;
reconstructing image data of the non-sensed image regions (neB) on the basis of the sensed image data (eBD) of the sensed image regions (eB) to form reconstructed image data (rBD);
ascertaining a noise signal of the sensed image data (eBD) of the sensed regions (eB); and
transferring the noise signal of the sensed image data (eBD) to the reconstructed image data (rBD) of the non-sensed regions (neB), to generate a substantially uniform noise distribution across overall image data of the object in a resultant overall image (rGB Inv ).

2. The method according to claim 1 , further comprising:
applying a selected noise signal to the reconstructed image data (rBD) as a noise signal of the non-sensed regions (neB),
comparing the effect of the selected noise signal on the reconstructed image data (rBD) of the non-sensed regions (neB) to the sensed image data (eB) of the sensed image regions (eB), and
verifying an observance of specified similarity criteria, and
if the specified similarity criteria is not satisfied, making a decision regarding application of a modified noise signal to the reconstructed image data (rBD) of the non-sensed regions (neB), on the basis of a result of the comparison.

3. The method according to claim 1 , wherein
the reconstructed image data (rBD) of the non-sensed image regions (neB) are generated by a simulation model impressing the noise signal on the reconstructed image data (rBD).

4. The method according to claim 1 , wherein
the resultant overall image (rGB) is predicted by a simulation model proceeding from the reconstructed image data (rBD) of the non-sensed image regions (neB), with parameters of a microscope (M) used for sensing of the sensed image data (eBD) of the sensed image regions (eB) being taken into account within a scope of the prediction.

5. The method according to claim 1 , wherein
an expected noise signal of the sensed image data (eBD) of the sensed image regions (eB) is estimated before the object is incompletely scanned, and
said expected noise signal is impressed following the incomplete scanning, on the reconstructed image data (rBD) of the non-sensed image regions (neB).

6. The method according to claim 1 , wherein the incomplete scanning is by an electron microscope.

7. The method according to claim 1 , wherein the object is incompletely scanned with at least one first spectral region and one second spectral region, and
image data of non-sensed image regions (neB) of at least one spectral region are reconstructed in the form of reconstructed image data (rBD).

8. The method according to claim 7 , wherein wavelength ranges of the at least one first spectral region and the second spectral region do not overlap one another.

9. The method according to claim 1 , wherein the incomplete scanning is implemented only at specific points in time of a time period, and
non-sensed image regions (neB) of non-sensed points in time are reconstructed in the form of reconstructed image data (rBD).

10. The method according to claim 1 , wherein the incomplete scanning is implemented in a form of a z-stack, along an optical axis of an objective lens (OB) used for sensing of image data.

Assignees

• Zeiss Carl Microscopy Gmbh

Inventors

• Amthor Manuel
• Haase Daniel
• Engelmann Ralf

Classifications

• H01J37/222

  Image processing arrangements associated with the tube · CPC title

• G06T2207/10061

  from scanning electron microscope · CPC title

• H01J2237/221

  Image processing · CPC title

• H01J37/28

  with scanning beams {(H01J37/268, H01J37/292, H01J37/2955 take precedence)} · CPC title

• H01J2237/226

  Image reconstruction · CPC title