# US 2026/0087595 A1 · Systems and methods for identifying and correcting blurred components within images

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

- Number: US 2026/0087595 A1
- Document ID: US20260087595A1
- Status: Published application
- Kind: A1
- Application number: 19/405,071
- Filed: 2025-12-01
- Published: 2026-03-26
- Earliest claimed: 2022-08-19
- Assignees: Capital One Services, LLC
- Applicants: Capital One Services, LLC
- Inventors: Joseph James Ford, III; Thomas Oliver Cantrell
- Web page: https://patentgrep.com/patent/US20260087595A1

## CPC

- G06T 5/50: using two or more images, e.g. averaging or subtraction
- G06V 30/19173: {Classification techniques}
- G06T 5/60: using machine learning, e.g. neural networks
- G06T 5/73: Deblurring; Sharpening
- G06V 10/70: using pattern recognition or machine learning (optical pattern recognition or electronic computations therefor G06V10/88)
- G06V 10/87: {using selection of the recognition techniques, e.g. of a classifier in a multiple classifier system}
- G06V 30/133: {Evaluation of quality of the acquired characters}
- G06V 30/164: Noise filtering
- G06V 30/1801: {Detecting partial patterns, e.g. edges or contours, or configurations, e.g. loops, corners, strokes or intersections (extracting features by contour coding G06V30/182)}
- G06V 30/187: {Frequency domain transformation; Autocorrelation}
- G06V 30/42: based on the type of document
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Methods and systems are described herein for identifying the location and nature of any blur within one or more images received as a user communication and generating an appropriate correction. The system utilizes a first machine learning model, which is trained to identify blurred components of inputted images and determine whether the blurred components are located in portions of the inputted images comprising textual information. The system may apply a corrective action selected by the first machine learning model, which may comprise stitching blurred images together to a sharp product image and/or some other method appropriate for rectifying images received.

## Claims

### Claim 1

A method, comprising:
generating a feature input representing a first image depicting textual information and non-textual information;
inputting the feature input into a machine learning model to obtain an output indicating that (i) the first image has a blurred component and (ii) an error type associated with the blurred component; and
executing, based on the error type, a corrective action to the first image to generate a second image comprising the textual information and the non-textual information without the blurred component.

### Claim 2 (depends on 1)

The method of claim 1, wherein the machine learning model comprises a first machine learning model, executing the corrective action comprises:
inputting the first image and the error type to a second machine learning model to generate the non-textual information without the blurred component; and
generating, using the second machine learning model, the second image.

### Claim 3 (depends on 1)

The method of claim 1, wherein executing the corrective action comprises:
extracting a portion of a third image corresponding to a location of the blurred component in the first image, wherein the portion of the third image is unblurred; and
combining the portion of the third image with the first image to generate the second image.

### Claim 4 (depends on 1)

The method of claim 1, wherein executing the corrective action comprises:
generating a request for an additional image comprising the textual information and the non-textual information; and
providing the request to a user device.

### Claim 5 (depends on 1)

The method of claim 1, wherein the feature input comprises a first feature input and the output comprises a first output, the method further comprises:
generating a second feature input representing a third image depicting the textual information and the non-textual information; and
inputting the second feature input into the machine learning model to obtain a second output indicating that the third image lacks any blurred components.

### Claim 6 (depends on 1)

The method of claim 1, wherein the feature input comprises a first feature input, the output comprises a first output, and the error type comprises a first error type, the method further comprises:
generating a second feature input representing a third image depicting the textual information and the non-textual information; and
inputting the second feature input into the machine learning model to obtain a second output indicating that (i) the third image has an additional blurred component and (ii) a second error type associated with the additional blurred component.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
extracting, from the first image, a greyscale ratio; and
comparing the greyscale ratio against a greyscale ratio threshold to identify the error type among a plurality of error types.

### Claim 8 (depends on 1)

The method of claim 1, wherein generating the feature input comprises:
detecting a first portion of the first image depicting the textual information and a second portion of the first image depicting the non-textual information; and
filtering the second portion of the first image.

### Claim 9 (depends on 1)

The method of claim 1, wherein generating the feature input comprises:
identifying an image type of the first image;
retrieving an image template associated with the image type indicating locations of relevant textual information within images of the image type; and
extracting, based on the image template, a portion of the first image comprising the textual information used to identify blurred components.

### Claim 10 (depends on 1)

The method of claim 1, wherein executing the corrective action comprises:
generating a first corrective action option for a first corrective action and a second corrective action option for a second corrective action; and
detecting a selection of the first corrective action option or the second corrective action option to cause execution of the first corrective action or the second corrective action, respectively.

### Claim 11 (depends on 1)

The method of claim 1, wherein generating the feature input comprises:
receiving, via a user device, a communication comprising the first image; and
generating, using the machine learning model, the feature input based on the communication.

### Claim 12 (depends on 1)

The method of claim 1, further comprising:
capturing, using a camera of a user device, the first image, wherein the error type associated with the blurred component is based on at least one of the camera of the user device or the capturing of the first image.

### Claim 13

A system, comprising:
memory storing computer program instructions; and
one or more processors that execute the computer program instructions to cause the one or more processors to:
generate a feature input representing a first image depicting textual information and non-textual information;
input the feature input into a machine learning model to obtain an output indicating that (i) the first image has a blurred component and (ii) an error type associated with the blurred component; and
execute, based on the error type, a corrective action to the first image to generate a second image comprising the textual information and the non-textual information without the blurred component.

### Claim 14 (depends on 13)

The system of claim 13, wherein the machine learning model comprises a first machine learning model, the corrective action being executed comprises the one or more processors being configured to:
input the first image and the error type to a second machine learning model to generate the non-textual information without the blurred component; and
generate, using the second machine learning model, the second image.

### Claim 15 (depends on 13)

The system of claim 13, wherein the corrective action being executed comprises the one or more processors being configured to:
extract a portion of a third image corresponding to a location of the blurred component in the first image, wherein the portion of the third image is unblurred; and
combine the portion of the third image with the first image to generate the second image.

### Claim 16 (depends on 13)

The system of claim 13, wherein the corrective action being executed comprises the one or more processors being configured to:
generate a request for an additional image comprising the textual information and the non-textual information; and
provide the request to a user device.

### Claim 17 (depends on 13)

The system of claim 13, wherein the feature input comprises a first feature input and the output comprises a first output, the one or more processors are further configured to:
generate a second feature input representing a third image depicting the textual information and the non-textual information; and
input the second feature input into the machine learning model to obtain a second output indicating that the third image lacks any blurred components.

### Claim 18 (depends on 13)

The system of claim 13, wherein the feature input comprises a first feature input, the output comprises a first output, and the error type comprises a first error type, the one or more processors are further configured to:
generate a second feature input representing a third image depicting the textual information and the non-textual information; and
input the second feature input into the machine learning model to obtain a second output indicating that (i) the third image has an additional blurred component and (ii) a second error type associated with the additional blurred component.

### Claim 19 (depends on 13)

The system of claim 13, wherein the feature input being generated comprises the one or more processors being configured to:
receive, via a user device, a communication comprising the first image; and
generate, using the machine learning model, the feature input based on the communication.

### Claim 20

One or more non-transitory computer-readable media storing computer program instructions that, when executed by one or more processors, effectuate operations comprising:
obtaining a feature input representing a first image depicting textual information and non-textual information;
providing the feature input to a machine learning model to obtain an output indicating that (i) the first image has a blurred component and (ii) an error type associated with the blurred component; and
executing, based on the error type, a corrective action to the first image to generate a second image comprising the textual information and the non-textual information without the blurred component.
