US 11,995,545 B2
Training a neural network model for recognizing handwritten signatures based on different cursive fonts and transformations
Reza Farivar, Champaign, IL (US); Fardin Abdi Taghi Abad, Champaign, IL (US); Anh Truong, Champaign, IL (US); Mark Watson, Urbana, IL (US); Austin Walters, Savoy, IL (US); Jeremy Goodsitt, Champaign, IL (US); and Vincent Pham, Champaign, IL (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Nov. 19, 2021, as Appl. No. 17/455,692.
Application 17/455,692 is a continuation of application No. 16/520,899, filed on Jul. 24, 2019, granted, now 11,195,172.
Prior Publication US 2022/0076254 A1, Mar. 10, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/08 (2023.01); G06F 18/24 (2023.01); G06Q 20/38 (2012.01); G06V 10/82 (2022.01); G06V 30/19 (2022.01); G06V 30/226 (2022.01); G06V 40/30 (2022.01); G06N 3/04 (2023.01)
CPC G06N 3/08 (2013.01) [G06F 18/24 (2023.01); G06Q 20/3825 (2013.01); G06V 10/82 (2022.01); G06V 30/19147 (2022.01); G06V 30/19173 (2022.01); G06V 30/226 (2022.01); G06V 40/33 (2022.01); G06N 3/04 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, by a device, information indicating names of individuals;
applying, by the device, different cursive fonts to each of the names to generate images of different cursive names;
applying, by the device, transformations to the images of the different cursive names to generate a set of images,
wherein a transformation, of the transformations, transforms a straight line portion of contiguous characters of a name, of the different cursive names, to follow a curve rather than a straight line;
training, by the device, a neural network model, with the set of images, to generate a trained neural network model;
receiving, by the device, an image of a signature,
wherein the signature is associated with a name of the names;
permitting or denying access, by the device using the trained neural network model, to another device based on the name associated with the signature; and
performing a transaction based on permitting or denying access to the other device.