IP Library › Granted Patent US 12,613,520
Granted Patent B2
US 12,613,520 · App. 18/210,105 · Granted Apr 28, 2026

Live drone aegis and autonomous drone response

Inventors: Timothy Sulzer (Jenkintown, PA); Marcus Day (Jenkintown, PA); Samuel Mohebban (Washington, DC); Quinn Graehling (Columbus, OH); Kieran Carroll (Penn Valley, PA)
Assignee: ZeroEyes, Inc.
G05D1/0094G05D1/0022G05D1/0088G06V20/58
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 12,613,520
App. No.
18/210,105
Granted
Apr 28, 2026
Kind
B2
Abstract

A machine learning (“ML”) model may be used to detect a presence of an object in one or more frames received from a camera sensor. The ML model may insert bounding boxes around the object and annotate the bounding boxes with one or more attributes of the object. The one or more frames and the annotated bounding boxes may be stored in a database configured to be searchable by at least one attribute of the one or more attributes. It may be determined whether the object is true positive (“TP”) event or a false positive (“FP”) event. The ML model may be re-trained using one or more of the database and the determination. If the object is a TP event, an alert may be transmitted to one or more devices with a location of the object that is based off of location information received from the camera sensor.

Claims (39)

1 . An object detection and response method, the method comprising:

receiving a video stream from a camera sensor of one or more of a stationary camera and a first mobile autonomous vehicle (“MAV”);

selecting a set of frames from the video stream;

detecting, by a machine learning model, a presence of an object in one or more frames of the set of frames;

inserting, by the machine learning model, bounding boxes in an area of the object in each of the one or more frames;

annotating, by the machine learning model, the bounding boxes with one or more attributes of the object;

storing the one or more frames and the annotated bounding boxes in a database, the database configured to be searchable by at least one attribute of the one or more attributes;

determining whether the object is true positive (“TP”) event or a false positive (“FP”) event;

re-training the machine learning model using one or more of the database and the determination; and

if it is determined that the object is a TP event, directing a second mobile autonomous vehicle (“MAV”) to a location of the object, wherein the location of the object is determined by location information associated with the camera sensor.

2 . The method of claim 1 , wherein the re-training comprises varying one or more parameters of the machine learning model.

3 . The method of claim 2 , wherein the one or more parameters are associated with one or more characteristics of the camera sensor.

4 . The method of claim 1 , wherein the location information comprises one or more of global positioning system (“GPS”) coordinates and longitude, latitude, and altitude.

5 . The method of claim 4 , wherein the machine learning model is executed by a server system in communication with the camera sensor and the one or more devices via a communications interface.

6 . The method of claim 4 , wherein the machine learning model is executed by one or more of the first MAV and the second MAV in communication with the one or more devices via a communications interface.

7 . The method of claim 1 , wherein the second MAV is configured to automatically launch and travel to the location of the object and one or more of track the object with camera sensors and intervene with one or more threat neutralization and/or distraction countermeasures.

8 . The method of claim 7 , wherein the intervening with one or more threat neutralization and/or distraction countermeasures is controlled manually by a user.

9 . The method of claim 1 , wherein the determining whether the object is a TP or a FP is done manually by a user.

10 . The method of claim 1 , wherein the determining whether the object is a TP or a FP is done automatically by the machine learning model.

11 . A system for object detection and response, the system comprising:

a processor operatively coupled to a memory configured to store computer-readable instructions that, when executed by the processor, cause the processor to:

receive a video stream from a camera sensor of one or more of a stationary camera and a first mobile autonomous vehicle (“MAV”);

select a set of frames from the video stream;

detect, by a machine learning model, a presence of an object in one or more frames of the set of frames;

insert, by the machine learning model, bounding boxes in an area of the object in each of the one or more frames;

annotate, by the machine learning model, the bounding boxes with one or more attributes of the object;

store the one or more frames and the annotated bounding boxes in a database, the database configured to be searchable by at least one attribute of the one or more attributes;

determine whether the object is true positive (“TP”) event or a false positive (“FP”) event;

re-train the machine learning model using one or more of the database and the determination; and

if it is determined that the object is a TP event, directing a second mobile autonomous vehicle (“MAV”) to a location of the object, wherein the location of the object is determined by location information associated with the camera sensor.

12 . The system of claim 11 , wherein the re-training comprises varying one or more parameters of the machine learning model.

13 . The system of claim 12 , wherein the one or more parameters are associated with one or more characteristics of the camera sensor.

14 . The system of claim 11 , wherein the location information comprises one or more of global positioning system (“GPS”) coordinates and longitude, latitude, and altitude.

15 . The system of claim 14 , wherein the machine learning model is executed by a server system in communication with the camera sensor and the one or more devices via a communications interface.

16 . The system of claim 14 , wherein the machine learning model is executed by one or more of the first MAV and the second MAV in communication with the one or more devices via a communications interface.

17 . The system of claim 11 , wherein the second MAV is configured to automatically launch and travel to the location of the object and one or more of track the object with camera sensors and intervene with one or more threat neutralization and/or distraction countermeasures.

18 . The system of claim 17 , wherein the intervening with one or more threat neutralization and/or distraction countermeasures is controlled manually by a user.

19 . The system of claim 11 , wherein the determining whether the object is a TP or a FP is done manually by a user.

20 . The system of claim 11 , wherein the determining whether the object is a TP or a FP is done automatically by the machine learning model.

Assignments (4)
RELEASE OF SECURITY INTEREST Recorded May 7, 2025
From: OCEAN II PLO LLC
To: ZEROEYES, INC.
Reel/Frame 071053/0720 →
SECURITY INTEREST Recorded May 6, 2025
From: ZEROEYES, INC.; ZEROEYES GOVERNMENT SOLUTIONS LLC
To: HERCULES CAPITAL, INC.
Reel/Frame 071037/0255 →
SECURITY INTEREST Recorded Sep 27, 2024
From: ZEROEYES, INC.
To: OCEAN II PLO LLC
Reel/Frame 068724/0436 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 11, 2023
From: SULZER, TIMOTHY; DAY, MARCUS; MOHEBBAN, SAMUEL; GRAEHLING, QUINN; CARROLL, KIERAN
To: ZEROEYES, INC.
Reel/Frame 065824/0417 →
Continuity (3)
Provisional Application 63352281 · Jun 15, 2022
Provisional Application 63352289 · Jun 15, 2022
Related Publication 20230409033A1 · Dec 21, 2023
References Cited (11)
US 11308335B2 · Sulzer et al. · 2022 [cited by applicant]
US 12087059B1 · Gordon · 2024 [cited by examiner]
US 20160019427A1 · Martin · 2016 [cited by examiner]
US 20180233009A1 · Spector · 2018 [cited by examiner]
US 20180341262A1 · Yeshurun · 2018 [cited by examiner]
US 20200364468A1 · Sulzer · 2020 [cited by examiner]
US 20210117724A1 · Taheri · 2021 [cited by examiner]
US 20230252791A1 · Webb · 2023 [cited by examiner]
US 20230409054A1 · Bradley · 2023 [cited by examiner]
International Search Report and Written Opinion dated Sep. 21, 2023, of counterpart International Application No. PCT/US2023/25359. [cited by applicant]
International Preliminary Report on Patentability of the International Searching Authority and Written Opinion mailed Dec. 26, 2024, of counterpart International Application No. PCT/US2023/02359. [cited by applicant]