# US 2021/0342755 A1 · Verification of proof of work using computer vision and/or machine learning

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

- Number: US 2021/0342755 A1
- Document ID: US20210342755A1
- Status: Published application
- Kind: A1
- Application number: 17/305,252
- Filed: 2021-07-02
- Published: 2021-11-04
- Earliest claimed: 2020-01-30
- Assignees: Capital One Services, LLC
- Applicants: Capital One Services, LLC
- Inventors: Michael MOSSOBA; Abdelkader M'Hamed BENKREIRA; Latika GULATI
- Web page: https://patentgrep.com/patent/US20210342755A1

## CPC

- G06Q 10/0631: Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06N 20/00: Machine learning
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A device receives image data associated with an environment of an individual and identifies verifiable content by processing the image data. The device processes the verifiable content to determine a likelihood that the individual is an employee or agent of an organization and to determine a set of likelihoods of the environment being a work environment. The device determines, based on the likelihood and the set of likelihoods, a number of verifiable hours worked by the individual during a given time period. The device identifies wages data that specifies a number of hours worked that has been credited to the individual for the given time period, determines that an amount of wages paid to the individual is an incorrect amount of wages, and provides an alert to another device that indicates that the amount of wages paid to the individual is the incorrect amount of wages.

## Claims

### Claim 1

A method, comprising:
receiving, by a device, image data associated with an environment of an individual;
identifying, by the device, verifiable content based on using one or more content recognition techniques to process the image data;
determining, by the device and based on a machine-learning driven analysis of the verifiable content, a set of threshold likelihoods that the individual is present at a work environment during particular hours of a given time period;
determining, by the device and based on the set of threshold likelihoods, a number of verifiable hours worked by the individual during the given time period;
identifying, by the device, wages data associated with a number of hours worked by the individual for the given time period;
determining, by the device and based on the number of verifiable hours worked by the individual and the number of hours worked by the individual for the given time period, that an amount of wages paid to the individual is an incorrect amount of wages; and
providing, by the device, an alert to another device that indicates that the amount of wages paid to the individual is the incorrect amount of wages.

### Claim 2 (depends on 1)

The method of claim 1, wherein identifying the verifiable content is based on training a data model with the image data to output the set of threshold likelihoods,
wherein the set of threshold likelihoods indicate whether one or more pixels, of the image data, include the verifiable content.

### Claim 3 (depends on 1)

The method of claim 1, wherein determining the set of threshold likelihoods comprises:
processing the verifiable content to verify that the individual is associated with an organization associated with the work environment.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
receiving audio data associated with the individual; and
verifying the number of verifiable hours worked by the individual using the audio data.

### Claim 5 (depends on 1)

The method of claim 1, wherein the device includes smart glasses.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
training a data model to receive the image data and to output a set of scores that indicate whether one or more pixels, of the image data, represent sensitive information; and
filtering the image data to obscure the sensitive information by modifying the one or more pixels.

### Claim 7 (depends on 1)

The method of claim 1, wherein the verifiable content includes at least one of:
data identifying a user device,
data identifying particular content displayed via the user device,
data identifying a piece of equipment, or
data identifying a product.

### Claim 8

A device, comprising:
one or more memories; and
one or more processors, coupled to the one or more memories, configured to:
receive image data associated with an environment of an individual;
identify verifiable content based on using one or more content recognition techniques to process the image data;
determine, based on a machine-learning driven analysis of the verifiable content, a set of threshold likelihoods that the individual is present at a work environment during particular hours of a given time period;
determine, based on the set of threshold likelihoods, a number of verifiable hours worked by the individual during the given time period;
identify wages data associated with a number of hours worked by the individual for the given time period;
determine, based on the number of verifiable hours worked by the individual and the number of hours worked by the individual for the given time period, that an amount of wages paid to the individual is an incorrect amount of wages; and
provide an alert to another device that indicates that the amount of wages paid to the individual is the incorrect amount of wages.

### Claim 9 (depends on 8)

The device of claim 8, wherein identifying the verifiable content is based on training a data model with the image data to output the set of threshold likelihoods,
wherein the set of threshold likelihoods indicate whether one or more pixels, of the image data, include the verifiable content.

### Claim 10 (depends on 8)

The device of claim 8, wherein the one or more processors, when determining the set of threshold likelihoods, are configured to:
process the verifiable content to verify that the individual is associated with an organization associated with the work environment.

### Claim 11 (depends on 8)

The device of claim 8, wherein the one or more processors are further configured to:
receive audio data associated with the individual; and
verify the number of verifiable hours worked by the individual using the audio data.

### Claim 12 (depends on 8)

The device of claim 8, wherein the device includes smart glasses.

### Claim 13 (depends on 8)

The device of claim 8, wherein the one or more processors are further configured to:
train a data model to receive the image data and to output a set of scores that indicate whether one or more pixels, of the image data, represent sensitive information; and
filter the image data to obscure the sensitive information by modifying the one or more pixels.

### Claim 14 (depends on 8)

The device of claim 8, wherein the verifiable content includes at least one of:
data identifying a user device,
data identifying particular content displayed via the user device,
data identifying a piece of equipment, or
data identifying a product.

### Claim 15

A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:
one or more instructions that, when executed by one or more processors of a device, cause the device to:
receive image data associated with an environment of an individual;
identify verifiable content based on using one or more content recognition techniques to process the image data;
determine, based on a machine-learning driven analysis of the verifiable content, a set of threshold likelihoods that the individual is present at a work environment during particular hours of a given time period;
determine, based on the set of threshold likelihoods, a number of verifiable hours worked by the individual during the given time period;
identify wages data associated with a number of hours worked by the individual for the given time period;
determine, based on the number of verifiable hours worked by the individual and the number of hours worked by the individual for the given time period, that an amount of wages paid to the individual is an incorrect amount of wages; and
provide an alert to another device that indicates that the amount of wages paid to the individual is the incorrect amount of wages.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein identifying the verifiable content is based on training a data model with the image data to output the set of threshold likelihoods,
wherein the set of threshold likelihoods indicate whether one or more pixels, of the image data, include the verifiable content.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the one or more instructions, that cause the device to determine the set of threshold likelihoods, cause the device to:
process the verifiable content to verify that the individual is associated with an organization associated with the work environment.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the one or more instructions further cause the device to:
receive audio data associated with the individual; and
verify the number of verifiable hours worked by the individual using the audio data.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the device includes smart glasses.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the one or more instructions further cause the device to:
train a data model to receive the image data and to output a set of scores that indicate whether one or more pixels, of the image data, represent sensitive information; and
filter the image data to obscure the sensitive information by modifying the one or more pixels.
