Machine-learning models for image processing

Patent metadata

Field	Value
Publication number	US-12482286-B2
Application number	US-202519217154-A
Country	US
Kind code	B2
Filing date	May 23, 2025
Priority date	Apr 8, 2024
Publication date	Nov 25, 2025
Grant date	Nov 25, 2025

Abstract

Presented herein are systems and methods for the employment of machine learning models for image processing. A mobile application for client-side image processing and validation, which interacts with and leverages native image processing software of the client device, where the image processing software and the mobile application include any number of machine-learning models for identifying a document and attributes of the document for recognition and validation. This mobile application uses the image processing software from a client operating system to control the camera. The image processing software generates various types of information about a video frame and the document, and the mobile application invokes APIs or software libraries of the image processing software to access the information and validate the frame and document.

First claim

What is claimed is:

1 . A method for client-side processing and validation of document imagery, the method comprising:
obtaining, by a computing device associated with an end-user, video data comprising a plurality of frames from a camera of the computing device; at a predetermined interval:
selecting, by the computing device, a frame from the video data for processing at an imaging software program locally executed on the computing device;
obtaining, by the computing device, from the imaging software program a set of boundaries corresponding to one or more edges of a document in the selected frame according to pixel intensity of a set of pixels containing the document in the selected frame; and
obtaining, by the computing device, from the imaging software program one or more luminance values for the frame, including a document brightness value for the set of pixels containing the document;
determining, by the computing device, a background brightness value for a background set of pixels outside the set of boundaries of the document, and a contrast brightness value using the document brightness value and the background brightness value; and in response to determining that the document brightness value satisfies a first luminance threshold and the contrast brightness value satisfies a second luminance threshold, validating, by the computing device, the document of the frame selected at the predetermined interval.

2 . The method of claim 1 , wherein the computing device failed to validate a prior frame of the video data selected at a prior instance of the predetermined interval, and wherein the computing device selects the frame of the video data in response to failing to validate the prior frame.

3 . The method of claim 2 , wherein the computing device interrupts operation of the camera for capturing the video data in response to validating the document of the selected frame.

4 . The method of claim 1 , further comprising:
encrypting, by the computing device, the validated selected frame; and transmitting, by the computing device, the encrypted validated selected frame and mobile client device metadata to a backend server.

5 . The method of claim 1 , further comprising:
determining, by the computing device, an aspect ratio for the document using a vertical value and a horizontal value of the one or more dimensions of the document, wherein the computing device validates the document of the frame based further upon the aspect ratio and an aspect ratio threshold.

6 . The method of claim 1 , further comprising determining, by the computing device, an angle value of a corner of the document at an intersection of a pair of the boundaries corresponding to a vertical value and a horizontal value of the one or more dimensions, wherein the computing device validates the one or more dimensions of the document based upon the angle value of the corner satisfying an angle tolerance threshold.

7 . The method of claim 1 , wherein the computing device determines an area ratio between an area of portion of the image having the document and a total area of the selected frame, and wherein the computing device validates the document based further upon the area ratio of the document satisfying a document area threshold.

8 . The method of claim 1 , wherein the computing device determines the document brightness value based on a pixel intensity of the set of pixels containing the document.

9 . The method of claim 1 , wherein the computing device determines the background brightness value based on a pixel intensity of the background set of pixels outside the document.

10 . The method of claim 1 , wherein computing device validates the document of the frame based on comparing document brightness value against the first luminance threshold having an upper limit and a lower limit.

11 . A system comprising:
a computing device associated with an end-user comprising at least one processor, configured to:
obtain video data comprising a plurality of frames from a camera of the computing device;
at a predetermined interval:
select a frame from the video data for processing at an imaging software program locally executed on the computing device;
obtain from the imaging software program a set of boundaries corresponding to one or more edges of a document in the selected frame according to pixel intensity of a set of pixels containing the document in the selected frame; and
obtain from the imaging software program one or more luminance values for the frame, including a document brightness value for the set of pixels containing the document;

determine a background brightness value for a background set of pixels outside the set of boundaries of the document, and a contrast brightness value using the document brightness value and the background brightness value; and
in response to determining that the document brightness value satisfies a first luminance threshold and the contrast brightness value satisfies a second luminance threshold, validate the document of the frame selected at the predetermined interval.

12 . The system of claim 11 , wherein the computing device failed to validate a prior frame of the video data selected at a prior instance of the predetermined interval, and wherein the computing device selects the frame of the video data in response to failing to validate the prior frame.

13 . The system of claim 12 , wherein the computing device is further configured to interrupt operation of the camera that captures the video data in response to validating the document of the selected frame.

14 . The system of claim 11 , wherein the computing device is further configured to:
encrypting, by the computing device, the validated selected frame; and transmitting, by the computing device, the encrypted validated selected frame and mobile client device metadata to a backend server.

15 . The system of claim 11 , wherein the computing device is further configured to:
determining, by the computing device, an aspect ratio for the document using a vertical value and a horizontal value of the one or more dimensions of the document, wherein the computing device validates the document of the frame based further upon the aspect ratio and an aspect ratio threshold.

16 . The system of claim 11 , wherein the computing device is further configured to determine an angle value of a corner of the document at an intersection of a pair of the boundaries corresponding to a vertical value and a horizontal value of the one or more dimensions, and
wherein the computing device validates the one or more dimensions of the document based upon the angle value of the corner satisfying an angle tolerance threshold.

17 . The system of claim 11 , wherein the computing device is further configured to determine an area ratio between an area of portion of the image having the document and a total area of the selected frame, and wherein the computing device validates the document based further upon the area ratio of the document satisfying a document area threshold.

18 . The system of claim 11 , wherein the computing device is further configured to determine the document brightness value based on a pixel intensity of the set of pixels containing the document.

19 . The system of claim 11 , wherein the computing device is further configured to determine the background brightness value based on a pixel intensity of the background set of pixels outside the document.

20 . The system of claim 11 , wherein the computing device

Assignees

• Citibank Na

Inventors

• Sureka Ashutosh K
• Adimatyam Venkata Sesha Kiran Kumar
• Silver Miriam
• Funken Daniel
• Pham Toan
• Kapadia Vicky
• Palanivel Sakthivel
• Bhakoo Anurag

Classifications

• G06F21/602 Primary

  Providing cryptographic facilities or services · CPC title

• G06V30/42

  based on the type of document · CPC title

• G06V30/41

  Analysis of document content (recognition of printed characters based on code marks G06V30/224) · CPC title

• G06V30/19093

  Proximity measures, i.e. similarity or distance measures · CPC title

• G06V10/993

  Evaluation of the quality of the acquired pattern · CPC title