# US 2015/0104111 A1 · Methods and systems of local signal equalization

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

- Number: US 2015/0104111 A1
- Document ID: US20150104111A1
- Status: Published application
- Kind: A1
- Application number: 14/278,371
- Filed: 2014-05-15
- Published: 2015-04-16
- Earliest claimed: 2013-10-11
- Assignees: Disney Enterprises, Inc.
- Applicants: Disney Enterprises, Inc.
- Inventors: Derek Bradley; Thabo Beeler
- Web page: https://patentgrep.com/patent/US20150104111A1

## CPC

- G06T 5/73: Deblurring; Sharpening
- G06T 5/10: using non-spatial domain filtering
- G06T 7/32: using correlation-based methods
- G06T 2207/20016: Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
- G06T 2207/20021: Dividing image into blocks, subimages or windows
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Methods and systems of equalizing spatial frequency signals for images are provided. Images distorted through defocus, motion blur, under-sampling or other reasons may be equalized. Equalized spatial frequency signals may be used to determine correspondence matching between images. Image patches from two or more image signals may be identified and analyzed to determine image patches that contain similar representations of the real scene, with sufficient local details for accurately computing cross-correlation.

## Claims

### Claim 1

A computer-implemented method, comprising:
identifying a first image patch and a second image patch, the first image patch corresponding to a first set of spatial frequency signals and the second image patch corresponding to a second set of spatial frequency signals;
determining a reference frequency; and
equalizing the first image patch and the second image patch based on the reference frequency.

### Claim 2 (depends on 1)

The method of claim 1, wherein the step of determining the reference frequency comprises identifying a first cutoff frequency of the first image patch, identifying a second cutoff frequency of the second image patch, determining the reference frequency as a lower value of the first cutoff frequency and the second cutoff frequency.

### Claim 3 (depends on 1)

The method of claim 1, further comprising determining correspondence matching between the first image patch and the second image patch.

### Claim 4 (depends on 2)

The method of claim 2, wherein the step of equalizing the first image patch and the second image patch comprises scaling the first set of spatial frequency signals.

### Claim 5 (depends on 2)

The method of claim 2, wherein the step of identifying the first cutoff frequency comprises determining a first set of highest frequencies in the first image patch and the step of identifying the second cutoff frequency comprises determining a second set of highest frequencies in the second image patch.

### Claim 6 (depends on 5)

The method of claim 5, wherein each frequency of the first set of highest frequencies and the second set of highest frequencies has an amplitude higher than a threshold.

### Claim 7 (depends on 6)

The method of claim 6, further comprising selecting the threshold according to a noise level.

### Claim 8 (depends on 2)

The method of claim 2, wherein the step of determining the reference frequency comprises:
generating a first local frequency map of the first image patch and a second local frequency map of the second image patch; and
determining a first scale-space representation of the first image patch and a second scale-space representation of the second image patch;
wherein the reference frequency is determined based on the first local frequency map, the second local frequency map, the first scale-space representation, and a second scale-space representation.

### Claim 9 (depends on 8)

The method of claim 8, wherein the first scale-space representation is a first set of Gaussian image pyramids, each of the first set of Gaussian image pyramids is centered on a pixel of the first image patch, the second scale-space representation is a second set of Gaussian image pyramids, and each of the second set of Gaussian image pyramids is centered on a pixel of the second image patch.

### Claim 10 (depends on 2)

The method of claim 2, wherein the first image patch and the second image patch are identified based on a correspondence matching window.

### Claim 11

A system, comprising:
a processor; and
non-transitory memory storing a set of instructions, the set of instructions configured to cause the processor to:
identify a first image patch and a second image patch, the first image patch corresponding to a first set of spatial frequency signals and the second image patch corresponding to a second set of spatial frequency signals;
determine a reference frequency; and
equalize the first image patch and the second image patch based on the reference frequency.

### Claim 12 (depends on 11)

The system of claim 11, wherein the step of determining the reference frequency comprises identifying a first cutoff frequency of the first image patch, identifying a second cutoff frequency of the second image patch, determining the reference frequency as a lower value of the first cutoff frequency and the second cutoff frequency.

### Claim 13 (depends on 11)

The system of claim 11, wherein the set of instructions are further configured to cause the processor to determine correspondence matching between the first image patch and the second image patch using the equalized spatial frequency signals.

### Claim 14 (depends on 12)

The system of claim 12, wherein the step of equalizing the first image patch and the second image patch comprises scaling the first set of spatial frequency signals using the reference frequency.

### Claim 15 (depends on 12)

The system of claim 12, wherein the step of identifying the first cutoff frequency comprises determining a first set of highest frequencies in the first image patch and the step of identifying the second cutoff frequency comprises determining a second set of highest frequencies in the second image patch.

### Claim 16 (depends on 15)

The system of claim 15, wherein each frequency of the first set of highest frequencies and the second set of highest frequencies has an amplitude higher than a threshold.

### Claim 17 (depends on 16)

The system of claim 16, wherein the set of instructions are configured to cause the processor to select the threshold according to a noise level.

### Claim 18 (depends on 12)

The system of claim 12, wherein the set of instructions are configured to cause the processor to:
generate a first local frequency map of the first image patch and a second local frequency map of the second image patch; and
determine a first scale-space representation of the first image patch and a second scale-space representation of the second image patch;
wherein the reference frequency is determined based on the first local frequency map, the second local frequency map, the first scale-space representation, and a second scale-space representation.

### Claim 19 (depends on 18)

The system of claim 18, wherein the first scale-space representation is a first set of Gaussian image pyramids, each of the first set of Gaussian image pyramids is centered on a pixel of the first image patch, the second scale-space representation is a second set of Gaussian image pyramids, and each of the second set of Gaussian image pyramids is centered on a pixel of the second image patch.

### Claim 20 (depends on 12)

The system of claim 12, the first image patch and the second image patch are identified based on a correspondence matching window.
