# US 2015/0147002 A1 · System and method for registering an image sequence

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0147002 A1
- Document ID: US20150147002A1
- Status: Published application
- Kind: A1
- Application number: 14/405,437
- Filed: 2013-06-28
- Published: 2015-05-28
- Earliest claimed: 2012-06-28
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: Guillaume Leopold Theodorus Frederik Hautvast
- Web page: https://patentgrep.com/patent/US20150147002A1

## CPC

- G06T 5/70: Denoising; Smoothing
- G06T 7/30: Determination of transform parameters for the alignment of images, i.e. image registration
- G06T 7/38: Registration of image sequences
- G06T 2207/10016: Video; Image sequence
- G06T 2207/20182: Noise reduction or smoothing in the temporal domain; Spatio-temporal filtering
- G06T 2207/30004: Biomedical image processing
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

System (100) for registering an image sequence, comprising:—an input (120) for obtaining the image sequence (200), the image sequence comprising a plurality of images (201 -205) arranged sequentially within an image range;—a transformation processor (140) for establishing transformations between pairs of consecutive images in the image sequence (200) to obtain a plurality of transformations (211-214);—an alignment processor (160) for, based on the plurality of transformations (211-214), aligning the image sequence (200) to a reference image (203) to obtain a registered image sequence (230);—wherein the alignment processor (160) is arranged for establishing the reference image (203) from the plurality of images (201-205) based on a transformation metric, the transformation metric quantifying a degree of transformation required for aligning the image sequence (200) to the reference image (203).

## Claims

### Claim 1

A system for registering an image sequence, comprising:
an input for obtaining the image sequence, the image sequence comprising a plurality of images arranged sequentially within an image range;
a transformation processor for establishing transformations between adjacent pairs of consecutive images in the image sequence to obtain a plurality of transformations as transformation matrices;
an alignment processor for, based on the plurality of transformations, aligning the image sequence to a reference image to obtain a registered image sequence;
wherein the alignment processor arranged for establishing the reference image from the plurality of images based on a transformation metric, the transformation metric quantifying a degree of transformation required for aligning the image sequence the reference image, wherein the alignment processor is arranged for quantifying the degree of transformation in accordance with the transformation metric based on multiplication of the transformation matrices of the adjacent pairs of consecutive images.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the alignment processor is arranged for establishing the reference image by selecting the reference image from the plurality of images based on a minimization of the transformation metric.

### Claim 3 (depends on 2)

The system according to claim 2, wherein the alignment processor is arranged for establishing the reference image by i) for at least a subset of the plurality of images, calculating a transformation value based on the transformation metric, and ii) selecting one of said images having a lowest transformation value as the reference image.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the alignment processor is arranged for establishing an interpolated image as the reference image, the interpolated image having a position in the image range between a pair of consecutive images in die image sequence.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the alignment processor is arranged for establishing a region of interest in the image sequence, and wherein the transformation metric quantifies a degree of transformation required for aligning the region of interest in the image sequence to the region of interest in the reference image.

### Claim 6 (depends on 5)

The system according to claim 5, further comprising a region of interest detector for detecting the region of interest in the image sequencer.

### Claim at position 7

7-8. (canceled)

### Claim 9 (depends on 1)

The system according to claim 1, wherein the alignment processor is arranged for quantifying the degree of transformation by i) for each of the plurality of images, establishing a transformation matrix product being indicative of the transformation from a respective one of the plurality of images to the reference image, and ii) establishing a transformation value based on a sum of the transformation matrix products.

### Claim 10 (depends on 1)

The system according to claim 1, wherein the transformation matrix is a 4×4 transformation matrix.

### Claim 11 (depends on 1)

The system according to claim 1, wherein the alignment processor is arranged for generate a confidence value based on the transformation metric, the confidence value being indicative of a quality of aligning the image sequence to the reference image.

### Claim 12 (depends on 1)

The system according to claim 1, wherein the alignment processor arranged for establishing the reference image from the plurality of images further based on an image similarity metric, the image similarity metric quantifying a degree of correspondence obtained when aligning the image sequence to the reference image.

### Claim 13 (depends on 1)

The system according to claim 1, wherein the alignment processor is arranged for establishing the reference image from the plurality of images further based on a position of the reference image within the image range.

### Claim 14 (depends on 1)

A workstation or imaging apparatus comprising the system according to claim 1.

### Claim 15 (depends on 15)

A method for registering an image sequence, comprising:
obtaining the image sequence, the image sequence comprising a plurality of images arranged sequentially within an image range;
establishing transformations between adjacent pairs of consecutive images in the image sequence to obtain a plurality of transformations as transformation matrices;
based on the plurality of transformations, aligning the image sequence to a reference image to obtain a registered image sequence;
wherein the method further comprises:
establishing the reference image from the plurality of images based on a transformation metric, the transformation metric quantifying a degree of transformation required for aligning the image sequence to the reference image; and
quantifying the degree of transformation in accordance with the transformation metric based on multiplication of the transformation matrices of the adjacent pairs of consecutive images. cm 16. A computer program product comprising instructions for causing a processor system to perform the method according to claim 15.
