# US 8,542,889 B2 · Systems, methods, and computer-readable media for capturing a signature for use in a document

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

- Number: US 8,542,889 B2
- Document ID: US8542889B2
- Status: Granted
- Kind: B2
- Application number: 12/907,835
- Filed: 2010-10-19
- Published: 2012-04-19
- Granted: 2013-09-24
- Earliest claimed: 2010-10-19
- Assignees: Apple Inc.
- Applicants: Matthew Sarnoff
- Inventors: Matthew Sarnoff
- Web page: https://patentgrep.com/patent/US8542889B2

## CPC

- H04N 1/195: the array comprising a two-dimensional [2D] array
- G06V 30/1456: {based on user interactions}
- G06V 40/30: Writer recognition; Reading and verifying signatures
- G06V 40/33: {based only on signature image, e.g. static signature recognition}
- H04N 1/19594: {using a television camera or a still video camera}
- G06V 30/10: Character recognition
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Systems, methods, and computer-readable media are provided for capturing a signature, and placing a representation of the captured signature in an appropriate field of a document. A camera or other appropriate sensor can capture an image of a signature provided by a user on a piece of paper. The signature can be digitized to create a representation that a device may use in a displayed document. To determine where to place the representation, a horizontal line of a document can be identified by selectively rendering portions of the document adjacent to an input position, and identifying one or more boundaries for a detected horizontal line. The representation can be scaled to fit in a detected field of the document.

## Claims

### Claim 1

A method for creating a digital representation of a signature, the method comprising:
capturing an image of a signature, wherein the signature is aligned with a reference line associated with the image;
identifying a central row of pixels in the captured image corresponding to the reference line;
identifying a closest row of pixels above the central row of pixels that comprises a consecutive sequence of white pixels of at least a threshold length, and for which a threshold number of additional rows comprise a consecutive sequence of white pixels;
identifying a top boundary within the image for the signature, wherein the top boundary comprises a row having a particular amount of light portions, wherein the top boundary comprises the identified closest row of pixels;
identifying a bottom boundary within the image for the signature, wherein the bottom boundary comprises a row having a particular amount of light portions, and wherein the reference line is positioned between the top boundary and bottom boundary; and
applying a vectorization process to portions of the captured image that are between the top boundary and the bottom boundary to create a representation of the image.

### Claim 2 (depends on 1)

The method of claim 1, comprising:
directing a user to position the signature provided on an object opposite a camera such that the signature is aligned with the reference line that is provided by a display window.

### Claim 3 (depends on 2)

The method of claim 2, comprising:
displaying a preview of the image of the signature in a display window, wherein the display window comprises the reference line.

### Claim 4 (depends on 3)

The method of claim 3, comprising:
directing the user to move the object relative to the camera so that the signature is aligned with the reference line in the display window.

### Claim 5 (depends on 1)

The method of claim 1, comprising:
converting individual pixels of the captured image to black and white pixels using a thresholding function.

### Claim 6 (depends on 1)

The method of claim 1, wherein:
the consecutive sequence of white pixels comprises at least thirty white pixels.

### Claim 7

An electronic device, comprising:
a camera operative to capture an image of a signature provided on a substrate;
display circuitry operative to provide a portion of the captured image in a display window; and
control circuitry configured to:
process the captured image to create a black and white image of the signature;
identify a position of a reference line provided on the display window relative to the captured image;
identify a reference row of pixels corresponding to the captured reference line;
successively select each row of pixels above the identified reference row of pixels;
identify a first row of the successively selected rows of pixels that comprises a sequence of pixels as an upper boundary for the signature above the reference line;
identify a lower boundary for the signature below the reference line; and
generate a vector representation of the black portions of the image between the upper boundary and the lower boundary, wherein the vector representation corresponds to the signature.

### Claim 8 (depends on 7)

The electronic device of claim 7, wherein the control circuitry is configured to:
convert a captured color image to a grayscale image; and
apply an adaptive thresholding function to the grayscale image to create the black and white image of the signature.

### Claim 9 (depends on 7)

The electronic device of claim 7, wherein the sequence of pixels comprises a consecutive series of white pixels.

### Claim 10 (depends on 7)

The electronic device of claim 7, wherein the control circuitry is operative to:
generate the vector representation of the signature based on the portion of the image provided in the display window.

### Claim 11 (depends on 7)

The electronic device of claim 7, wherein:
the vector representation reproduces changes in line thickness and velocity of at least one stroke of the signature.

### Claim 12

A method for digitizing a signature, the method comprising:
capturing a color image of a signature while the signature is positioned relative to the displayed reference line;
processing the captured image to generate a black and white image corresponding to the captured image;
identifying a reference line for aligning the signature;
identifying a central row of pixels in the captured image corresponding to the reference line;
identifying a top boundary as a closest row of pixels above the central row of pixels that comprises a consecutive sequence of white pixels of at least a threshold length, and for which a threshold number of additional rows comprise a consecutive sequence of white pixels; and
applying a raster-to-vector process to the portion of the black and white image in regions adjacent to the reference line to generate a representation of the signature.

### Claim 13 (depends on 12)

The method of claim 12, comprising:
identifying a lower boundary below the central row.

### Claim 14 (depends on 12)

The method of claim 12, comprising:
encrypting the representation to block unauthorized use of the representation.

### Claim 15 (depends on 14)

The method of claim 14, comprising:
associating the encrypted representation with a user.

### Claim 16 (depends on 12)

The method of claim 12, wherein:
associating the representation with an application, wherein only the application can make use of the representation.

### Claim 17

A non-transitory, computer-readable medium comprising computer program logic recorded thereon wherein the computer program logic is configured to:
capture an image of a signature, wherein the signature is aligned with a reference line associated with the image;
identify a central row of pixels in the captured image corresponding to the reference line;
identify a top boundary for the signature above the central row of pixels, wherein the top boundary comprises a first row of the image having a first consecutive amount of light portions;
identify a bottom boundary for the signature below the central row of pixels, wherein the bottom boundary comprises a second row of the image having a second consecutive amount of light portions, and wherein the reference line is positioned between the top boundary and the bottom boundary, wherein the first and second amounts of light portions are greater than a light portion threshold; and
apply a vectorization process to portions of the captured image that are between the top boundary and the bottom boundary to create a representation of the signature.