# US 2021/0174478 A1 · Keystone corrections with quadrilateral objects

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

- Number: US 2021/0174478 A1
- Document ID: US20210174478A1
- Status: Published application
- Kind: A1
- Application number: 16/077,695
- Filed: 2017-06-16
- Published: 2021-06-10
- Earliest claimed: 2017-06-16
- Assignees: Hewlett-Packard Development Company, L.P.
- Applicants: Hewlett-Packard Development Company, L.P.
- Inventors: Xiangfei QIAN
- Web page: https://patentgrep.com/patent/US20210174478A1

## CPC

- G06T 5/80: Geometric correction
- G06T 7/80: Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- H04N 1/387: Composing, repositioning or otherwise {geometrically} modifying originals
- G06T 2207/10024: Color image
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

An example method includes receiving by a processor of an image capturing device, an image of a quadrilateral object captured by a camera of the image capturing device. A homography matrix is calculated by the processor based on the image of the quadrilateral object and a known dimension of the quadrilateral object. The processor performs a keystone correction based on the homography matrix to convert a subsequent image captured by the camera into a corrected image.

## Claims

### Claim 1

A method, comprising:
receiving, by a processor of an image capturing device, an image of a quadrilateral object captured by a camera of the image capturing device;
calculating, by the processor, a homography matrix based on the image of the quadrilateral object and a known dimension of the quadrilateral object; and
performing, by the processor, a keystone correction based on the homography matrix to convert a subsequent image captured by the camera into a corrected image.

### Claim 2 (depends on 1)

The method of claim 1, wherein the calculating the homography matrix comprises:
calculating, by the processor, an initial homography matrix based on the known dimension of the quadrilateral object and the image;
generating, by the processor, a rectangular image of the quadrilateral object using the initial homography matrix;
calculating, by the processor, a respective offset of each one of four corners of the rectangular image of the quadrilateral object and the quadrilateral object in the image; and
generating, by the processor, the homography matrix based on the respective offset of the each one of the four corners.

### Claim 3 (depends on 1)

The method of claim 1, wherein the known dimension comprises a length and a width of the quadrilateral object.

### Claim 4 (depends on 1)

The method of claim 1, wherein the known dimension is received via a user interface from a user.

### Claim 5 (depends on 1)

The method of claim 1, wherein the known dimension is detected based on a three dimensional camera of the image capturing device.

### Claim 6 (depends on 1)

The method of claim 1, wherein the receiving and the calculating are repeated when the image capturing device is moved.

### Claim 7

A non-transitory computer readable storage medium encoded with instructions executable by a processor of image capturing device, the non-transitory computer-readable storage medium comprising:
instructions to capture an image of a quadrilateral object via the image capturing device;
instructions to calculate a homography matrix based on the image of the quadrilateral object and a known dimension of the quadrilateral object;
instructions to capture a subsequent image; and
instructions to perform a keystone correction on the subsequent image based on the homography matrix to convert a subsequent image captured by the image capturing device into a corrected image.

### Claim 8 (depends on 7)

The non-transitory computer readable storage medium of claim 7, wherein the instructions to calculate the homography matrix comprises:
instructions to calculate an initial homography matrix based on the known dimension of the quadrilateral object and the image;
instructions to generate a rectangular image of the quadrilateral object using the initial homography matrix;
instructions to calculate a respective offset of each one of four corners of the rectangular image of the quadrilateral object and the quadrilateral object in the image; and
instructions to generate the homography matrix based on the respective offset of the each one of the four corners.

### Claim 9 (depends on 7)

The non-transitory computer readable storage medium of claim 7, wherein the known dimension comprises a length and a width of the quadrilateral object.

### Claim 10 (depends on 7)

The non-transitory computer readable storage medium of claim 7, wherein the known dimension is received via a user interface from a user.

### Claim 11 (depends on 7)

The non-transitory computer readable storage medium of claim 7, wherein the known dimension is detected based on a three dimensional camera of the image capturing device.

### Claim 12 (depends on 7)

The non-transitory computer readable storage medium of claim 7, wherein the receiving and the calculating are repeated when the image capturing device is moved.

### Claim 13

An apparatus, comprising:
a red, green, and blue (RGB) camera to capture an image of a quadrilateral object;
a three dimensional (3D) camera to detect dimensions of the quadrilateral object; and
a processor in communication with the RGB camera and the 3D camera to calculate a homography matrix based on the image of the quadrilateral object and the dimensions of the quadrilateral object, wherein processor performs a keystone correction based on the homography matrix to convert a subsequent image captured by the RGB camera into a corrected image.

### Claim 14 (depends on 13)

The apparatus of claim 13, wherein to calculate the homography matrix, the processor is to:
calculate an initial homography matrix based on the known dimension of the quadrilateral object and the image;
generate a rectangular image of the quadrilateral object using the initial homography matrix;
calculate a respective offset of each one of four corners of the rectangular image of the quadrilateral object and the quadrilateral object in the image; and
generate the homography matrix based on the respective offset of the each one of the four corners.

### Claim 15 (depends on 1)

The apparatus of claim 1, wherein the processor is to detect movement of the apparatus and to calculate an updated homography matrix based on an updated image of the quadrilateral object and the dimensions of the quadrilateral object.
