IP Library Granted Patent US 12,536,534
Granted Patent B2
US 12,536,534 · App. 15/365,559 · Granted Jan 27, 2026

Identity verification systems and methods

Inventor: Frank Joseph Lamont Gauld (Edinburgh, GB)
Assignee: NCR Atleos Corporation
G06Q20/4014G06F16/951G06F21/32G06Q40/028G06Q20/042G06Q20/0425G06V40/33
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,536,534
App. No.
15/365,559
Granted
Jan 27, 2026
Kind
B2
Abstract

Various embodiments herein each include at least one of systems, methods, and software for identity verification. Some such embodiments identify a user based on analysis of their handwriting. Some such embodiments may be utilized to authenticate an individual, for emergency authentication or when other authentication solutions are not available, as one of a two or more step authentication process, authenticating checks, authenticating a signature or other handwriting on another document, and the like. One embodiment, in the form of a method, includes storing a set of handwriting characteristics of an individual identified during processing of a plurality of handwriting samples of the individual. The method may then determine whether an input handwriting sample is handwriting of the individual based on the stored set of handwriting characteristics of the individual and then output a result of the determining.

Claims (41)

1 . A method comprising:

analyzing a set of handwriting characteristics for each of a plurality of individuals for whom handwriting data from previously processed checks that have been received and stored in a database stored on a data storage device, each set of handwriting characteristics analyzed by a process that executes on a computer processor against a plurality of handwriting samples of each individual from archived documents stored in the database to identify characteristics of an individual's handwriting across a plurality of documents associated in stored data with an identity of an individual's identity;

wherein the process identifies and extracts unique characteristics of the individual's handwriting by analyzing archived documents stored in the database to build a unique handwriting fingerprint that enables authentication of an individual in emergency situations when other identification items or solutions are not available;

the process including:

identifying and extracting representations of characteristics of handwriting from stored images of handwriting samples, the characteristics including neatness, fluidity, rhythmic, artistic embellishment, slant or inclination of characters, character forms, movement of a writing utensil when forming characters or portions thereof, proportions of character portions compared to a whole character or other portions, height ratios of characters with relation to one or more other characters, loop shape and form, pressure as is evident from line width, writing utensil lifting when forming certain characters, exit and entry strokes for characters, retracing, character and word spacing, case of letters, and other characteristics;

storing, in the database on the data storage device, a set of extracted handwriting characteristics of the individual identified during processing of the plurality of handwriting samples of each individual, wherein a stored set of handwriting characteristics of the individual is stored in the database storing sets of handwriting characteristics of a set of individuals generated from previously processed check images, each set of handwriting characteristics stored as a data representation of a plurality of handwriting characteristics including data representing of:

character slant;

character forms;

writing instrument direction of movement when forming a character or portion thereof;

proportions between character portions;

height ratios between characters; and

character loop forms;

wherein storing further includes maintaining at least two or more separate data representations for a given individual who utilizes two or more distinct handwriting styles and maintaining the at least two or more separate data representations as handwriting fingerprints for corresponding individuals in the database;

receiving, by the computer processor, an input handwriting sample captured by an imaging device as an image of a check through a web service call;

computing, by the computer processor, whether the input handwriting sample captured by the imaging device as the image of the check is handwriting of the individual based on an individual's stored handwriting fingerprints, wherein computing further includes determining a degree of uniqueness for the input handwriting sample, wherein computing further includes generating a value for the degree of uniqueness for the input handwriting sample in view of the individual's stored handwriting fingerprints based on the characteristics in a set corresponding to the individual in the individual's stored handwriting fingerprints and based on corresponding characteristics extracted from the input handwriting sample, wherein the degree of uniqueness is generated based on a degree of matching of certain characteristics of a handwriting fingerprint with a handwriting input, a number of matches, and a frequency of matches;

outputting, from the computer processor, a result of the computing as the degree of uniqueness via the web service call for determination by a terminal whether the check can be used as payment or deposited at the terminal and providing the degree of uniqueness as an authentication of the individual when the individual fails to have other identification items needed for authentication of the individual associated with the check; and

providing the method as a web service to the terminal.

2 . The method of claim 1 , wherein the plurality of handwriting samples are stored check images.

3 . The method of claim 1 , wherein the input handwriting sample is received by a server performing at least a portion of the method from a source over a network and the result of the determining is output in a data transmission from the server to the source over the network.

4 . The method of claim 3 , wherein the source is the terminal.

5 . The method of claim 4 , wherein the terminal is an Automated Teller Machine (ATM).

6 . A system comprising:

a processor, a memory device, a network interface device, and instructions stored on the memory device that are executable by the processor to perform data processing activities, the data processing activities executed by the processor comprising:

analyzing a set of handwriting characteristics for each of a plurality of individuals for whom handwriting data from previously processed checks that have been received in digital image form and respective representations of which are stored in a database stored on the memory device, each set of handwriting characteristics analyzed by a process that executes against a plurality of handwriting samples of each individual from archived documents stored in the database to identify characteristics of an individual's handwriting across a plurality of documents associated in stored data with an identity of an the individual's identity;

wherein the process identifies and extracts unique characteristics of the individual's handwriting by analyzing archived documents stored in the database to build a unique handwriting fingerprint that enables authentication of an individual in emergency situations when other identification items or solutions are not available;

the process including:

identifying and extracting representations of characteristics of handwriting from stored handwriting samples, the characteristics including neatness, fluidity, rhythmic, artistic embellishment, slant or inclination of characters, character forms, movement of a writing utensil when forming characters or portions thereof, proportions of character portions compared to a whole character or other portions, height ratios of characters with relation to one or more other characters, loop shape and form, pressure as is evident from line width, writing utensil lifting when forming certain characters, exit and entry strokes for characters, retracing, character and word spacing, case of letters, and other characteristics;

storing, in the database on the memory device, a set of extracted handwriting characteristics of the individual identified during processing of a plurality of handwriting samples of each individual, wherein the database stores sets of handwriting characteristics for a set of individuals generated from previously processed check images, each set of handwriting characteristics stored as a data representation of a plurality of handwriting characteristics including data representation of:

character slant;

character forms;

writing instrument direction of movement when forming a character or portion thereof;

proportions between character portions;

height ratios between characters; and

character loop forms;

wherein storing further includes maintaining at least two or more separate data representations for a given individual who utilizes two or more distinct handwriting styles and maintaining the at least two or more separate data representations as handwriting fingerprints for corresponding individuals in the database;

receiving, by the processor, an input handwriting sample captured by an imaging device as an image of a check through a web service call;

computing, by the processor, whether the input handwriting sample captured as the image of the check received via the network interface device is handwriting of the individual based on an individual's stored handwriting fingerprints wherein computing further includes determining a degree of uniqueness for the input handwriting sample, wherein computing further includes generating a value for the degree of uniqueness for the input handwriting sample in view of the individual's stored handwriting fingerprints based on the characteristics in a set corresponding to the individual in the individual's stored handwriting fingerprints and based on corresponding characteristics extracted from the input handwriting sample, wherein the degree of uniqueness is generated based on a degree of matching of certain characteristics of a handwriting fingerprint with a handwriting input, a number of matches, and a frequency of matches;

outputting, from the processor, a result of the computing via the network interface device via the web service call for determination by a terminal whether the check can be used as payment or deposited at the terminal and providing the degree of uniqueness as an authentication of the individual when the individual fails to have other identification items needed for authentication of the individual associated with the check; and

providing the data processing activities as a web service to the terminal.

7 . The system of claim 6 , wherein:

the plurality of handwriting samples are stored check images.

Assignments (8)
CORRECTIVE ASSIGNMENT TO CORRECT THE THE PROPERTIES SECTION BY INCLUDING IT WITH TEN PREVIOUSLY OMITTED PROPERTY NUMBERS PREVIOUSLY RECORDED ON REEL 65346 FRAME 367. ASSIGNOR(S) HEREBY CONFIRMS THE SECURITY INTEREST. Recorded Aug 13, 2025
From: NCR ATLEOS CORPORATION; CARDTRONICS USA, LLC
To: BANK OF AMERICA, N.A., AS ADMINISTRATIVE AGENT
Reel/Frame 072445/0072 →
CHANGE OF NAME Recorded May 20, 2024
From: NCR CORPORATION
To: NCR VOYIX CORPORATION
Reel/Frame 067464/0595 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 20, 2024
From: NCR VOYIX CORPORATION
To: NCR ATLEOS CORPORATION
Reel/Frame 067464/0882 →
CORRECTIVE ASSIGNMENT TO CORRECT THE DOCUMENT DATE AND REMOVE THE OATH/DECLARATION (37 CFR 1.63) PREVIOUSLY RECORDED AT REEL: 065331 FRAME: 0297. ASSIGNOR(S) HEREBY CONFIRMS THE SECURITY INTEREST. Recorded Oct 31, 2023
From: NCR ATLEOS CORPORATION
To: CITIBANK, N.A.
Reel/Frame 065627/0332 →
SECURITY INTEREST Recorded Oct 24, 2023
From: NCR ATLEOS CORPORATION
To: CITIBANK, N.A.
Reel/Frame 065331/0297 →
CORRECTIVE ASSIGNMENT TO CORRECT THE PROPERTY NUMBERS SECTION TO REMOVE PATENT APPLICATION: 15000000 PREVIOUSLY RECORDED AT REEL: 050874 FRAME: 0063. ASSIGNOR(S) HEREBY CONFIRMS THE SECURITY INTEREST. Recorded Apr 12, 2021
From: NCR CORPORATION
To: JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT
Reel/Frame 057047/0161 →
SECURITY INTEREST Recorded Oct 29, 2019
From: NCR CORPORATION
To: JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT
Reel/Frame 050874/0063 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 14, 2017
From: GAULD, FRANK JOSEPH LAMONT
To: NCR CORPORATION
Reel/Frame 041771/0359 →
Continuity (1)
Related Publication 20180150839A1 · May 31, 2018
References Cited (33)
US 4701960A · Scott · 1987 [cited by examiner]
US 5559895A · Lee · 1996 [cited by examiner]
US 6424728B1 · Ammar · 2002 [cited by examiner]
US 7168614B2 · Kotovich · 2007 [cited by examiner]
US 7545982B2 · Houle · 2009 [cited by examiner]
US 7636457B2 · Franke · 2009 [cited by examiner]
US 9520034B1 · Walker · 2016 [cited by examiner]
US 20010018739A1 · Anderson · 2001 [cited by examiner]
US 20020152170A1 · Dutta · 2002 [cited by examiner]
US 20030138135A1 · Chung · 2003 [cited by examiner]
US 20050035193A1 · Gustin · 2005 [cited by examiner]
US 20060041506A1 · Mason · 2006 [cited by examiner]
US 20060187698A1 · Schmidt · 2006 [cited by examiner]
US 20060202012A1 · Grano · 2006 [cited by examiner]
US 20070084911A1 · Crowell · 2007 [cited by examiner]
US 20080005579A1 · Gaines · 2008 [cited by examiner]
US 20090164372A1 · Dell · 2009 [cited by examiner]
US 20090173781A1 · Ramachandran · 2009 [cited by examiner]
US 20100161466A1 · Gilder · 2010 [cited by examiner]
US 20130028502A1 · Nepomniachtchi · 2013 [cited by examiner]
US 20130103582A1 · Singfield · 2013 [cited by examiner]
US 20130159183A1 · Lopez · 2013 [cited by examiner]
US 20140040141A1 · Gauvin · 2014 [cited by examiner]
US 20150254655A1 · Bondesen · 2015 [cited by examiner]
US 20170046560A1 · Tsur · 2017 [cited by examiner]
CN 103250132A · 2013 [cited by examiner]
Plamondon et. al, “Online and off-line handwriting recognition: a comprehensive survey”, https://ieeexplore.ieee.org/document/824821 , Jan. 2000 (Year: 2000). [cited by examiner]
Donato et. Al, “Automatic Signature Verification: The State of the Art”, https://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=4603093&punumber=5326 , published in IEEE Transactions on Systems, Man, and Cybernetics, Pa… [cited by examiner]
Desai Et. Al, “Forensic Examination Of Handwriting And Signatures”, International Journal of Innovative Research and Development, https://www.internationaljournalcorner.com/index.php/ijird_ojs/issue/view/8152 , 2013 (Ye… [cited by examiner]
Bertrand Et. Al “A Conditional Random Field Model for Font Forgery Detection”, https://ieeexplore.ieee.org/document/7333827 , 2015 (Year: 2015). [cited by examiner]
Chianese et. al, “Recovering dynamic information from static handwriting.”, Pattern Recognition, vol. 26, No. 3., 409, 418, accessed via https://www.researchgate.net/publication/220602224_Recovering_dynamic_information_… [cited by examiner]
Haffeman et. al, “Offline Handwritten Signature Verification—Literature Review”, https://arxiv.org/pdf/1507.07909v2 , 2015 (Year: 2015). [cited by examiner]
Forensic science simplified—“A Simplified Guide To Forensic Document Examination” https://web.archive.org/web/20131008222119/https://www.forensicsciencesimplified.org/docs/QuestionedDocuments.pdf , archived 2013 (Year: … [cited by examiner]