Device and method for denoising a vector-valued image

Patent metadata

Field	Value
Publication number	US-10839488-B2
Application number	US-201716092066-A
Country	US
Kind code	B2
Filing date	Apr 24, 2017
Priority date	May 3, 2016
Publication date	Nov 17, 2020
Grant date	Nov 17, 2020

Abstract

The present invention relates to a device (100) for denoising a vector-valued image, the device (100) comprising: a generator (10), which is configured to generate an initial loss function (L_I) comprising at least one initial covariance matrix (ICM) defining a model of correlated noise for each pixel of the vector-valued image; a processor (20), which is configured to provide a final loss function (L_F) comprising a set of at least one final covariance matrix (FCM) based on the initial loss function by modifying at least one submatrix and/or at least one matrix element of the initial covariance matrix (ICM); and a noise-suppressor (30), which is configured to denoise the vector-valued image using the final loss function (L_F) comprising the set of the at least one final covariance matrix (FCM).

First claim

The invention claimed is:

1. A device for denoising a vector-valued image in a medical imaging system, the device comprising:
a generator configured to generate an initial loss function comprising an initial covariance matrix defining a model of correlated noise for each pixel of the vector-valued image;
a processor configured to divide the initial covariance matrix into two or more matrices based on two or more different spatial frequency bands of the vector-valued image in order to determine a final loss function; and
a noise-suppressor configured to denoise the vector-valued image using the final loss function.

2. The device according to claim 1 , wherein the two or more different spatial frequency bands of the vector-valued image, are defined by at least one high spatial frequency band and by at least one low spatial frequency band.

3. The device according to claim 2 ,
wherein the processor is configured to provide a tuning between cross-talk removal and correlated noise removal of frequency noise.

4. The device according to claim 1 ,
wherein the generator is configured to generate the initial loss function by adding a regularization term to a matrix product of the initial covariance matrix and the vector-valued image.

5. The device according to claim 4 ,
wherein the generator is configured to generate the initial loss function by adding the regularization term comprising a regularization strength parameter.

6. The device according to claim 1 ,
wherein the initial covariance matrix is constant for all pixel positions across the vector-valued image.

7. The device according to claim 1 ,
wherein the processor is configured to perform:
a frequency dependent covariance tuning in a material projection domain of a maximum-likelihood CT reconstruction of the vector-valued image; and/or
projection denoising with a Gaussian noise model of the vector-valued image.

8. The device according to claim 1 ,
wherein the processor is configured to reduce absolute values of off-diagonal elements of the initial covariance matrix at edges of material inhomogeneities of the vector-valued image.

9. The device according to claim 8 ,
wherein the processor is configured to extract the edges of the material inhomogeneities from the vector-valued image with a reduced noise level.

10. The device according to claim 9 ,
wherein the processor is configured to extract the edges of the material inhomogeneities by applying at least one of a Sobel operator, a Prewitt operator, a Marr-Hildreth operator, a Laplacian operator, and a differential edge detection to the vector-valued image.

11. A medical imaging system comprising a device according to claim 1 .

12. A computer-implemented method for denoising a vector-valued image in a medical imaging system, the method comprising:
generating an initial loss function comprising an initial covariance matrix defining a model of correlated noise for each pixel of the vector-valued image;
dividing the initial covariance matrix into two or more matrices based on two or more different spatial frequency bands of the vector-valued image in order to determine a final loss function; and
denoising the vector-valued image using the final loss function.

13. The method according to claim 12 , further comprising reducing absolute values of off-diagonal elements of the initial covariance matrix at edges of material inhomogeneities of the vector-valued image.

Assignees

• Koninklijke Philips Nv

Inventors

• Bergner Frank
• Brendel Bernhard Johannes
• Koehler Thomas

Classifications

• G06T12/30

  Image post-processing, e.g. metal artefact correction · CPC title

• G06T2207/10072

  Tomographic images · CPC title

• G06T2211/424

  Iterative · CPC title

• G06F17/16

  Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)} · CPC title

• G06T2211/408

  Dual energy · CPC title