IP Library Granted Patent US 12705327
Granted Patent B1
US 12705327 · App. 18/893,107 · Granted Aug 11, 2026

System and method for user-specific captchas

Inventors: Gregory David Hansen (Fuquay Varina, NC); Darrin Keith Wylie (San Antonio, TX); Lance David Brown (San Antonio, TX); Brittney Chiu Childers (San Antonio, TX); Liana Nicole Hamel (San Antonio, TX); Yolandra Jovan Hendrix (Dallas, TX); Karen Barnett Niemeyer (Helotes, TX); Evelyn Teresa Rimmer (San Antonio, TX)
Assignee: United Services Automobile Association (USAA)
G06F21/36G06F21/316G06F2221/2103G06F2221/2133
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 12705327
App. No.
18/893,107
Granted
Aug 11, 2026
Kind
B1
Abstract

The embodiments provide a system and method for improved CAPTCHA challenges that utilize user-specific information. In some embodiments, personalized information about assets currently or previously owned assets, including properties and/or vehicles, are collected. The system then builds a dataset (a “user-specific CAPTCHA dataset”) that is comprised of images including the user-owned assets. The user-specific CAPTCHA dataset can then be used to create personalized, or user-specific, CAPTCHA challenges that include images from the data set. For systems that implement CAPTCHA challenges for multiple different users, each user-specific dataset may be associated to a particular user identifier (such as a username or email address).

Claims (33)

1 . A computer implemented method for gathering images for use in a Completely Automate Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) challenge, the computer implemented method comprising the steps of:

receiving information about an asset owned by a user;

submitting a query for an image representative of the asset owned by the user, the query including the information about the asset, and receiving a returned image, wherein submitting the query includes sending the query to an online search engine;

applying a machine learning algorithm to the returned image to detect an object matching the information about the asset;

upon failing to detect the object within the returned image using the machine learning algorithm, causing the returned image to be presented, via a user interface, to a human reviewer for confirmation that the returned image includes an item matching the information about the asset; and

storing the returned image as an asset image, wherein the returned image is configured to be used in a CAPTCHA challenge for the user at a later time.

2 . The computer implemented method according to claim 1 , wherein the submitting the query includes querying a database of images.

3 . The computer implemented method according to claim 1 , wherein the returned image includes metadata, and wherein the metadata includes a tag associated with the information about the asset.

4 . The computer implemented method according to claim 1 , wherein the machine learning algorithm further comprises a neural network.

5 . The computer implemented method according to claim 1 , wherein the asset is a vehicle.

6 . The computer implemented method according to claim 1 , wherein the asset is a real estate property.

7 . A system for gathering images for use in a Completely Automated Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) challenge, the system comprising:

a processor and memory, the memory storing instructions executable by the processor to:

receive information about an asset owned by a user;

submit a query for an image representative of the asset owned by the user, the query including the information about the asset, and receiving a returned image, wherein submitting the query includes sending the query to an online search engine;

apply a machine learning algorithm to the returned image to detect an object matching the information about the asset;

upon failing to detect the object within the returned image using the machine learning algorithm, causing the returned image to be presented, via a user interface, to a human reviewer for confirmation that the returned image includes an item matching the information about the asset; and

store the returned image as an asset image in the memory, wherein the returned image is configured to be used in a CAPTCHA challenge for the user at a later time.

8 . The system according to claim 7 , wherein submitting the query includes querying a database of images.

9 . The system according to claim 7 , wherein the returned image includes metadata, and wherein the metadata includes a tag associated with the information about the asset.

10 . The system according to claim 7 , wherein the machine learning algorithm includes a neural network.

11 . The system according to claim 7 , wherein the asset is a vehicle.

12 . The system according to claim 7 , wherein the asset is a real estate property.

13 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations for gathering images for use in a Completely Automated Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) challenge, the operations comprising:

receiving information about an asset owned by a user;

submitting a query for an image representative of the asset owned by the user, the query including the information about the asset, and receiving a returned image, wherein the submitting the query includes sending the query to an online search engine;

applying a machine learning algorithm to the returned image to detect an object matching the information about the asset;

upon failing to detect the object within the returned image using the machine learning algorithm, causing the returned image to be presented, via a user interface, to a human reviewer for confirmation that the returned image includes an item matching the information about the asset; and

storing the returned image as an asset image, wherein the returned image is configured to be used in a CAPTCHA challenge for the user at a later time.

14 . The non-transitory computer-readable medium according to claim 13 , wherein the submitting the query includes querying a database of images.

15 . The non-transitory computer-readable medium according to claim 13 , wherein the returned image includes metadata, and wherein the metadata includes a tag associated with the information about the asset.

16 . The non-transitory computer-readable medium according to claim 13 , wherein the machine learning algorithm further comprises a neural network.

17 . The non-transitory computer-readable medium according to claim 13 , wherein the asset is selected from a group consisting of a vehicle and a real estate property.