Electronic device for automatic registration of images

• US 6,539,127 B1
• Filed: 07/27/1999
• Issued: 03/25/2003
• Est. Priority Date: 07/28/1998
• Status: Expired due to Term

• Current Assignee: Inria Institut National de Recherche EN Informatique et EN Automatique
• Original Assignee: Inria Institut National de Recherche EN Informatique et EN Automatique
• Inventors: Roche, Alexis, Ayache, Nicholas, Malandain, Gregoire, Pennec, Xavier
• Primary Examiner(s): Mehta, Bhavesh
• Assistant Examiner(s): PATEL, KANJIBHAI B
• Application Number:
  US09/361,313
• Time in Patent Office: 1,337 Days
• Field of Search: 382/154, 382/162, 382/128-134, 382/274, 382/151, 382/181, 382/254, 382/276, 382/278, 382/284, 382/289, 382/291, 382/294, 382/295, 382/296, 382/300, 348/42, 702/85, 702/150-154
• US Class Current: 382/294
• CPC Class Codes: G06T 3/4061 by injecting details from different spectral ranges

Abstract

An electronic data processing device receives first and second data sets representing first and second comparable digital images. It comprises registration means comprising a first module which calculates a main function representative of the correlation ratios between the data of the first and second sets, and a second module which determines a registration transformation between one of the images and the other from the main function.

Claims

1. An electronic data processing device, comprising:

a receiving means for receiving first and second data sets representing first and second comparable digital images, each datum of said sets comprising a position component and an intensity component;

a registration means for determining, from these two data sets, a registration transformation between one of the images and the other; and

an output means for outputting the registration transformation;

wherein the registration means includes;

(a) a first module able to calculate a main function representative of the correlation ratios between the data of the first set and the data of the second set; and

(b) a second module able to determine the registration transformation from said main function.

2. A device according to claim 1, wherein the calculation of the main function is performed in a space having a chosen norm.

3. A device according to claim 2, wherein the space is a Hilbert space.

4. A device according to claim 1, further comprising:

a means for allowing the designation of said first image as “

reference”

image, and of said second image as “

floating”

image, so that the registration transformation is determined for registering the floating image with regard to the reference image.

5. A device according to claim 1, wherein the second module is arranged so as to choose a maximum value of the main function according to a chosen criterion and to determine the registration transformation from the maximum value.

6. A device according to claim 1, wherein the registration transformation is determined within a class of arbitrary n-dimensional geometric transformations, n being chosen at least from the values 2 and 3 according to the dimension of the images.

7. A device according to claim 6, wherein the n-dimensional geometric transformations are determined within a class of affine n-dimensional transformations referred to as “

rigid”

.

8. A device according to claim 1, wherein the registration means is arranged so as to estimate the main function from the data of the first and second data sets satisfying a chosen criterion.

9. A device according to claim 1, wherein the comparable digital images are medical images.

10. A device according to claim 1, wherein the first and the second image originate from different acquisition techniques.

11. An electronic data processing device, comprising:

a receiving means for receiving first and second data sets representing first and second comparable digital images, each datum of said sets comprising a position component and an intensity component;

a registration means for determining, from these two data sets, a registration transformation between one of the images and the other; and

an output means for outputting the registration transformation;

wherein the registration means includes;

(a) a first module able to calculate a main function representative of the correlation ratios between the data of the first set and the data of the second set;

(b) a second module able to determine the registration transformation from said main function; and

(c) a third module able to estimate a joint probability density function from the first and second data sets;

and wherein the first module is arranged so as to calculate the main function from the joint probability density function.

12. A device according to claim 11, wherein the calculation of the joint probability density function pertains to the intensity components of the data of the data sets.

13. A device according to claim 11, further comprising;

a fourth module arranged so as to transform the first data set into a third data set comprising data whose position components correspond to each of the position components of the data of the second data set.

14. A device according to claim 13, wherein the transformation of the first data set comprises the application of an initialization registration transformation and an interpolation.

15. A device according to claim 14, wherein the interpolation is of a trilinear type.

16. A device according to claim 14, wherein the registration means further includes a third module able to estimate a joint probability density function from the first and second data sets;

and the first module is arranged so as to calculate the main function from the joint probability density function;

wherein the second module is arranged so as to determine whether the registration transformation satisfies a chosen criterion; and

wherein in the event of non-satisfaction of the criterion the registration transformation is addressed to the third module in such a way as to replace the transformation previously used for obtaining the third data set with a view to a new registration transformation determination.

17. A device according to claim 11, wherein the calculation of the main function is performed in a space having a chosen norm.

18. A device according to claim 17, wherein the space is a Hilbert space.

19. A device according to claim 11, further comprising:

a means for allowing the designation of said first image as “

reference”

image, and of said second image as “

floating”

image, so that the registration transformation is determined for registering the floating image with regard to the reference image.

20. A device according to claim 11, wherein the registration transformation is determined within a class of arbitrary n-dimensional geometric transformations, n being chosen at least from the values 2 and 3 according to the dimension of the images.

21. A device according to claim 20, wherein the n-dimensional geometric transformations are determined within a class of affine n-dimensional transformations referred to as “

rigid”

.

22. A device according to claim 11, wherein the comparable digital images are medical images.

23. A device according to claim 11, wherein the first and the second image originate from different acquisition techniques.

24. A method of processing two data sets respectively representing first and second comparable digital images, each datum of said sets comprising a position component and an intensity component, said method comprising:

receiving the two data sets;

determining a registration transformation between one of the first and second images and the other, from the two data sets; and

outputting the registration transformation;

wherein the determining step includes the following substeps;

a) estimating a joint probability density function from the data of the first set and of the second set;

b) calculating from the joint probability density function a main function representative of the correlation ratios between the data of the first set and of the second set, and

c) determining the registration transformation from the calculated main function.