IP Library › Granted Patent US 12,602,517
Granted Patent B2
US 12,602,517 · App. 17/504,275 · Granted Apr 14, 2026

Systems and methods for automatic redaction of sensitive information from video streams

Inventors: Marco Pistoia (Amawalk, NY); William Moriarty (West Chester, PA); Hargun Kalsi (Monmouth Junction, NJ); Matthew Murphy (Wigan, GB); Shaohan Hu (Yorktown Heights, NY)
Assignee: JPMORGAN CHASE BANK, N.A.
G06F21/64G06K19/06037G06K19/0723H04N7/183
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 12,602,517
App. No.
17/504,275
Granted
Apr 14, 2026
Kind
B2
Abstract

Systems and methods for automatic redaction of sensitive information from video streams are disclosed. According to one embodiment, a method for automatic redaction of sensitive information from video streams may include: (1) receiving, by an image processing computer program executed by an electronic device, a video stream of an area; (2) identifying, by the image processing computer program, an object capable of having sensitive information thereon in the video stream of the area; and (3) redacting or obscuring, by the image processing computer program, the object in the video stream.

Claims (14)

1 . A method for automatic redaction of sensitive information from video streams, comprising:

receiving, by an image processing computer program executed by an electronic device, a video stream of an office, a conference room, or a workspace;

accessing, by the image processing computer program, a database comprising known locations of known fixed objects within the office, the conference room, or the workspace that are capable of having sensitive information thereon, wherein the known fixed objects consist of a whiteboard, a blackboard, a display, a desk, or a table;

identifying, by the image processing computer program and using a deep neural network model, an object capable of having sensitive information thereon in the video stream of the office, the conference room, or the workspace, wherein the sensitive information comprises text or image data;

matching, by the image processing computer program, a location of the object to one of the known locations of the known fixed objects within the office, the conference room, or the workspace; and

redacting or obscuring, by the image processing computer program, the object in the video stream in response to the location of the object matching one of the known locations.

2 . The method of claim 1 , further comprising:

identifying, by the image processing computer program, content on the object;

wherein the image processing computer program redacts or obscures the content on the object in the video stream.

3 . The method of claim 2 , further comprising:

determining, by the image processing computer program, that the content is sensitive information by recognizing character data in the content using optical character recognition and matching, by the image processing computer program, the character data to sensitive character data in a sensitive information database;

wherein the image processing computer program redacts or obscures the sensitive information on the object in the video stream.

4 . The method of claim 1 , wherein the sensitive information comprises business confidential information and/or personal identifiable information.

5 . The method of claim 1 , wherein the video stream is received from a fixed image capture device.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 24, 2025
From: PISTOIA, MARCO; MORIARTY, WILLIAM; KALSI, HARGUN; MURPHY, MATTHEW; HU, SHAOHAN
To: JPMORGAN CHASE BANK, N.A.
Reel/Frame 070934/0426 →
Continuity (1)
Related Publication 20230120993A1 · Apr 20, 2023
References Cited (8)
US 9854206B1 · Ren · 2017 [cited by examiner]
US 10885632B1 · Stahl · 2021 [cited by examiner]
US 11546141B1 · Guest · 2023 [cited by examiner]
US 20190311298A1 · Kopp · 2019 [cited by examiner]
US 20210073412A1 · Kvochko · 2021 [cited by examiner]
US 20220207181A1 · Ng · 2022 [cited by examiner]
US 20220414264A1 · Ananthanarayanan · 2022 [cited by examiner]
CN 111859451B · 2024 [cited by examiner]