# US 2013/0156281 A1 · Image registration method

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

- Number: US 2013/0156281 A1
- Document ID: US20130156281A1
- Status: Published application
- Kind: A1
- Application number: 13/814,760
- Filed: 2011-08-11
- Published: 2013-06-20
- Earliest claimed: 2010-08-31
- Assignees: Shanghai Microport Ep Medtech Co., Ltd.
- Inventors: Yiyong Sun; Jingjing Gong; Daozhi Liu
- Web page: https://patentgrep.com/patent/US20130156281A1

## CPC

- G06V 10/754: {involving a deformation of the sample pattern or of the reference pattern; Elastic matching}
- G06T 3/14: Transformations for image registration, e.g. adjusting or mapping for alignment of images
- G06T 7/33: using feature-based methods
- G06V 20/695: {Preprocessing, e.g. image segmentation}
- G06T 2207/10072: Tomographic images
- G06T 2207/10136: 3D ultrasound image
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

The present application relates to an image registration method, comprising: selecting a registration source image {p,} and a registration target image {qj}; applying a random perturbation to {pi} in accordance with a preset random perturbation control parameter σ so that it is deformed to obtain {pi′}, and obtaining a set of closest points, i.e., {qj′}, on {qj} corresponding to points on {pi′}; performing an iterative operation on {pi} and {qj′} in accordance with a preset initial coordinate transformation H0 to obtain a coordinate transformation {Hl}, 0<l<L, between {pi′} and {qj′}, and calculating an average distance {El} corresponding to the coordinate transformation {Hl} in accordance with the coordinate transformation {Hl}; judging magnitudes of the average distance {El} and a preset ideal average distance Ex, and terminating the registration when the average distance {El} is smaller than or equal to the preset ideal average distance.

## Claims

### Claim 1

An image registration method, characterized by comprising:
selecting a registration source image {p} and a registration target image {qj}, wherein 0<i<M, 0<j<N, M and N are positive integers, and M<<N;
applying a random perturbation to the image {pi} in accordance with a preset random perturbation control parameter σ so that it is deformed to obtain an image {pi′}, and obtaining closest points on the image {q,} corresponding to points on the image {pi′}, the obtained set of points being an image {qj′};
performing an iterative operation on the image {pi′} and the image {qj′} within a preset step length L in accordance with a preset initial coordinate transformation H0 to obtain a coordinate transformation {Hi}, 0<l<L, between the image {pi′} and the image {qj′},and calculating an average distance {El} corresponding to the coordinate transformation {Hl} in accordance with the coordinate transformation {Hl};
judging magnitudes of the average distance {El} and a preset ideal average distance Ex, and terminating the registration when the average distance {El} is smaller than or equal to the preset ideal average distance.

### Claim 2 (depends on 1)

The method according to claim 1, characterized in that the pi and qj are column vectors of 3*1.

### Claim 3 (depends on 2)

The method according to claim 2, characterized in that said obtaining closest points on the image {qj} corresponding to points on the image {pi′} is specifically:
searching for the closest points on the image {qj} corresponding to the points on the image {pi′} by means of KD binary tree.

### Claim 4 (depends on 3)

The method according to claim 3, characterized in that said judging magnitudes of the average distance {El} and the ideal average distance Ex is specifically:
judging in turn whether the average distance {El} is smaller than or equal to the ideal average distance E.

### Claim 5 (depends on 3)

The method according to claim 3, characterized in that said judging magnitudes of the average distance {El} and the ideal average distance Ex is specifically:
selecting the minimum average distance Emin in the average distance {El};
judging magnitudes of the minimum average distance Emin and the ideal average distance E.

### Claim 6 (depends on 1)

The method according to claim 1, characterized by, when the average distance {El} is larger than the preset ideal average distance, further comprising steps of:
judging in turn magnitudes of the average distance {El} and an initial average distance E0, wherein the initial average distance E0 is corresponding to the initial coordinate transformation H0;
when any value in the average distance {El} is smaller than or equal to the initial average distance E0, terminating the judgment between the average distance {El} and the initial average distance E0, replacing the initial coordinate transformation H0 with a coordinate transformation Hg corresponding to a current average distance Eg, decaying the random perturbation control parameter σ in accordance with a fixed decaying rate, recalculating an average distance {El′} in accordance with the coordinate transformation Hg and the decayed random perturbation control parameter σ, and judging whether the average distance {El′} is smaller than or equal to the preset ideal average distance Ex;
when all the values in the average distance {El} are larger than the initial average distance E0, after decaying the random perturbation control parameter σ in accordance with the fixed decaying rate, reapplying a random perturbation to the image {pi}, recalculating an average distance {E1″} by using the initial coordinate transformation H0, and judging whether the average distance {E1″} is smaller than or equal to the preset ideal average distance Ex.

### Claim 7 (depends on 1)

The method according to claim 1, characterized by, when the average distance {El} is larger than the preset ideal average distance, further comprising steps of:
selecting the minimum average distance Emin in the average distance {El};
judging magnitudes of the minimum average distance Emin and the initial average distance E0, wherein the initial average distance E0 is corresponding to the initial coordinate transformation H0;
when the minimum average distance Emin is smaller than or equal to the initial average distance E0, replacing the initial coordinate transformation H0 with a coordinate transformation Hmin corresponding to the minimum average distance Emin, decaying the random perturbation control parameter σ in accordance with a fixed decaying rate, recalculating an average distance {El′} in accordance with the coordinate transformation Hmin and the decayed random perturbation control parameter σ, and judging whether the average distance {El′} is smaller than or equal to the preset ideal average distance Ex;
when the minimum average distance Emin is larger than the initial average distance E0, after decaying the random perturbation control parameter σ in accordance with the fixed decaying rate, reapplying a random perturbation to the image {pi}, recalculating an average distance {E1″} by using the initial coordinate transformation H0, and judging whether the average distance {El′} is smaller than or equal to the preset ideal average distance Ex.

### Claim 8 (depends on 6)

The method according to claim 6, characterized by, after said recalculating an average distance {El′} or recalculating an average distance {El″}, further comprising:
judging whether a number of iteration times is larger than a preset number of iteration times K, and terminating the registration and reporting an error when the number of iteration times is larger than the preset number of iteration times K.

### Claim 9 (depends on 6)

The method according to claim 6, characterized in that the fixed decaying rate of the random perturbation control parameter σ is
1
2
.

### Claim 10 (depends on 1)

The method according to claim 1, characterized in that the step length L is in a range of 50˜200.