Robust subspace recovery via dual sparsity pursuit

Patent metadata

Field	Value
Publication number	US-9418318-B2
Application number	US-201414468725-A
Country	US
Kind code	B2
Filing date	Aug 26, 2014
Priority date	Aug 30, 2013
Publication date	Aug 16, 2016
Grant date	Aug 16, 2016

Abstract

A computer-implemented method of detecting a foreground data in an image sequence using a dual sparse model framework includes creating an image matrix based on a continuous image sequence and initializing three matrices: a background matrix, a foreground matrix, and a coefficient matrix. Next, a subspace recovery process is performed over multiple iterations. This process includes updating the background matrix based on the image matrix and the foreground matrix; minimizing an L−1 norm of the coefficient matrix using a first linearized soft-thresholding process; and minimizing an L−1 norm of the foreground matrix using a second linearized soft-thresholding process. Then, background images and foreground images are generated based on the background and foreground matrices, respectively.

First claim

We claim:

1. A computer-implemented method of detecting a foreground data in an image sequence using a dual sparse model framework, the method comprising:
creating an image matrix based on a continuous image sequence;
initializing a background matrix, a foreground matrix, and a coefficient matrix;
performing a subspace recovery process over a plurality of iterations until convergence of one or more of the background matrix, the foreground matrix, and the coefficient matrix over the plurality of iterations, the subspace recovery process comprising:
updating the background matrix based on the image matrix and the foreground matrix,
updating the coefficient matrix by minimizing an L−1 norm of the coefficient matrix using a first linearized soft-thresholding process, and
updating the foreground matrix by minimizing an L−1 norm of the foreground matrix using a second linearized soft-thresholding process;

generating one or more background images based on the background matrix; and
generating one or more foreground images based on the foreground matrix.

2. The method of claim 1 , wherein the subspace recovery process utilizes one or more tuning parameters.

3. The method of claim 2 , wherein tuning parameters are applied by the first linearized soft-thresholding process and the second linearized soft-thresholding process.

4. The method of claim 3 , wherein the subspace recovery process further comprises:
updating the tuning parameters during each of the plurality of iterations.

5. The method of claim 4 , wherein the tuning parameters comprise one or more Lagrange multiplier values.

6. The method of claim 1 , further comprising:
generating a foreground image sequence based on the one or more foreground images; and
generating a background image sequence based on the one or more background images.

7. The method of claim 6 , wherein
the background image sequence depicts a lymphatic system or a blood vessel; and
the foreground image sequence depicts a passage of fluid through the lymphatic system or the blood vessel to an organ or tissue.

8. The method of claim 7 , further comprising:
using the foreground image sequence to generate a measurement of intensity variation.

9. A computer-implemented method of performing intensity variation segmentation using a dual sparse model framework, the method comprising:
receiving a myocardial perfusion image sequence comprising a plurality of images depicting fluid passing through a cardiac structure over time;
creating an image matrix based on a continuous image sequence;
initializing a background matrix, a foreground matrix, and a coefficient matrix;
performing a subspace recovery process over a plurality of iterations until convergence of one or more of the background matrix, the foreground matrix, and the coefficient matrix over the plurality of iterations, the subspace recovery process comprising:
updating the background matrix based on the image matrix and the foreground matrix,
updating the coefficient matrix by minimizing an L−1 norm of the coefficient matrix using a first linearized soft-thresholding process, and
updating the foreground matrix by minimizing an L−1 norm of the foreground matrix using a second linearized soft-thresholding process; and

using the foreground matrix to generate a measurement of intensity variation across the myocardial perfusion image sequence.

10. The method of claim 9 , further comprising:
generating a sequence of images depicting motion of the cardiac structure over time based on the background matrix.

11. The method of claim 9 , wherein the subspace recovery process utilizes one or more tuning parameters.

12. The method of claim 11 , wherein tuning parameters are applied by the first linearized soft-thresholding process and the second linearized soft-thresholding process.

13. The method of claim 12 , wherein the tuning parameters comprise one or more Lagrange multiplier values.

14. The method of claim 11 , wherein the subspace recovery process further comprises:
updating the one or more tuning parameters during each of the plurality of iterations.

15. An article of manufacture for detecting a foreground data in an image sequence using a dual sparse model framework, the article of manufacture comprising a non-transitory, tangible computer-readable medium holding computer-executable instructions for performing a method comprising:
creating an image matrix based on a continuous image sequence;
initializing a background matrix, a foreground matrix, and a coefficient matrix;
performing a subspace recovery process over a plurality of iterations until convergence of one or more of the background matrix, the foreground matrix, and the coefficient matrix over the plurality of iterations, the subspace recovery process comprising:
update the background matrix based on the image matrix and the foreground matrix,
update the coefficient matrix by minimizing an L−1 norm of the coefficient matrix using a first linearized soft-thresholding process, and
update the foreground matrix by minimizing an L−1 norm of the foreground matrix using a second linearized soft-thresholding process;

generating one or more background images based on the background matrix; and
generating one or more foreground images based on the foreground matrix.

16. The article of manufacture of claim 15 , wherein the method further comprises:
generating a foreground image sequence based on the one or more foreground images; and
generating a background image sequence based on the one or more background images.

17. The article of manufacture of claim 16 , wherein
the background image sequence depicts a lymphatic system or a blood vessel; and
the foreground image sequence depicts a passage of fluid through the lymphatic system or the blood vessel to an organ or tissue.

18. The article of manufacture of claim 17 , wherein the method further comprises:
using the foreground image sequence to generate a measurement of intensity variation.

Assignees

• Siemens Ag
• Univ North Carolina State

Inventors

• Nadar Mariappan S
• Bian Xiao
• Wang Qiu
• Cetingul Hasan Ertan
• Krim Hamid
• Plaetevoet Lucas

Classifications

• G06T2207/30104

  Vascular flow; Blood flow; Perfusion · CPC title

• G06K9/645 Primary

  Physics · mapped topic

• G06T7/0081

  Physics · mapped topic

• G06T2207/20144

  Physics · mapped topic

• G06T2207/30048

  Heart; Cardiac · CPC title