Method of superimposing images

• US 20060098897A1
• Filed: 11/02/2005
• Published: 05/11/2006
• Est. Priority Date: 11/10/2004
• Status: Abandoned Application

• Current Assignee: AGFA Healthcare NV (Agfa-Gevaert N.V.)
• Original Assignee: Agfa-Gevaert N.V.
• Inventors: Dewaele, Piet
• Application Number:
  US11/264,945
• Publication Number:
  US 20060098897A1
• Time in Patent Office: Days
• US Class Current: 382/294
• CPC Class Codes: G06T 2207/30008 Bone
• CPC Class Codes: G06T 3/00 Geometric image transformations in the plane of the image
• CPC Class Codes: G06T 7/32 using correlation-based methods
• CPC Class Codes: G06T 7/337 involving reference images or patches
• CPC Class Codes: G06T 7/35 using statistical methods
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors

Abstract

A structure in a first image is mapped onto its corresponding structure in a second image by applying a geometric transformation to the first image so that both images are expressed in a common coordinate system. Starting from initial values the parameters of the geometric transformation are updated taking into account the result of an evaluation of a cost function. Finally the superimposed images may be blended.

Claims

1. A method of superimposing images comprising the steps of:

(1) obtaining a digital signal representation of at least two images A and B,

(2) applying at least the digital signal representation of image A to a display device coupled to a computer and displaying said image A, and

(3) applying a geometric transform to the digital signal representation of image B so as to map at least a geometric structure in image B onto a corresponding geometric structure in image A in a common coordinate system.

2. A method according to claim 1, wherein

(i) initial parameters of a geometric transform are computed which maps said geometric structure from image B onto a corresponding structure in image A,

(ii) applying said geometric transform with the computed parameters to the digital signal representation of image B so as to transform image B into a common coordinate system as image A,

(iii) evaluating a cost function to generate a cost value indicative of a degree of alignment of the transformed image B and the image A,

(iv) updating parameters of said geometric transform taking into account said cost value, and

(v) re-iterating steps (ii) to (iv) to obtain final parameters of said geometric transform when the generated cost value is optimal.

3. A method according to claim 1, wherein a blended image is generated, each pixel value of said blended image being a combination of a corresponding pixel value in image A and a corresponding pixel value in a transformed image resulting from applying said transform to image B.

4. A method according to claim 2, wherein a blended image is generated, each pixel value of said blended image being a combination of a corresponding pixel value in image A and a corresponding pixel value in a transformed image resulting from applying said transform to image B.

5. A method according to claim 1, wherein the transform with said final parameters is applied.

6. A method according to claim 3, wherein the contribution to said combination of pixel values of image A and of transformed image B is variable.

7. A method according to claim 4, wherein the contribution to said combination of pixel values of image A and of transformed image B is variable.

8. A computer program product adapted to carry out the steps of claim 1 when run on a computer.

9. A computer readable carrier medium comprising computer executable program code adapted to carry out the steps of claim 1.