IP Library Granted Patent US 12,387,137
Granted Patent B2
US 12,387,137 · App. 17/677,321 · Granted Aug 12, 2025

Processing images of wind farms and plant farms captured by a drone

Inventors: Rachana Vidhi (Palm Beach Gardens, FL); Prasanna Shrivastava (Palm Beach Gardens, FL)
Assignee: INVENTUS HOLDINGS, LLC
G06N20/00F03D17/00G06T7/11G06N3/02G06T2207/10032G06T2207/20081
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 12,387,137
App. No.
17/677,321
Granted
Aug 12, 2025
Kind
B2
Abstract

An artificial intelligence (AI) ensemble engine analyzes a set of images that include a wind turbine and a region of a plant farm to identify potential problem areas in the wind turbine and the plant farm. The AI ensemble engine also analyzes the potential problem areas of the wind turbine and the plant farm to determine whether maintenance is warranted for the potential problem areas to provide a list of problem areas. A notification manager determines whether each problem area in the list of problem areas is related to a wind turbine or agriculture and generates a notification based on the determination.

Claims (39)

1. A non-transitory machine readable medium having machine executable instructions causing a processor to execute operations, the machine executable instructions comprising:

an image processor configured to parse a set of images to generate a subset of images by discarding images without a portion of a wind turbine and a region of a plant farm;

an artificial intelligence (AI) ensemble engine having a number of modules trained to distinguish features of the portion of the wind turbine and the region of the plant farm, the AI ensemble engine:

analyzes the features of the subset of images that include the portion of the wind turbine and the region of a plant farm to identify potential problem areas in the wind turbine and the plant farm; and

analyzes the potential problem areas of the wind turbine and the plant farm to determine whether maintenance is warranted for the potential problem areas to provide a list of problem areas; and

a notification manager that determines whether each problem area in the list of problem areas is related to a wind turbine or agriculture of the plant farm and generates a notification to a turbine operator or a plant farm operator based on the determination.

2. The medium of claim 1 , wherein the notification manager provides the notification to a ticket generator interface in response to determining that a given problem area in the list of problem areas is related to a wind turbine, wherein the notification includes a location of the wind turbine extracted from a given image of the subset of images.

3. The medium of claim 2 , wherein the ticket generator interface generates a service ticket with an order to dispatch a maintenance crew in response to the notification.

4. The medium of claim 2 , wherein the given problem area is a crack in the wind turbine.

5. The medium of claim 2 , wherein the given problem area is a hot spot on the wind turbine.

6. The medium of claim 1 , wherein the notification manager provides a short message service (SMS) message and/or an email to the plant farm operator in response to determining that a given problem area in the list of problem areas is related to the plant farm, wherein the notification includes an image of the given problem area extracted from the subset of images.

7. The medium of claim 6 , wherein the given problem area is a dead spot on the plant farm.

8. The medium of claim 6 , wherein the given problem area is a collapsed wind turbine.

9. The medium of claim 1 , wherein the subset of images are captured by a drone surveying the wind turbine and the plant farm.

10. A system for analyzing images comprising:

a drone that captures a set of images, the set of images including images of a wind turbine on a wind farm and images of a plant farm proximal to the wind farm;

an image processor configured to parse a set of images to discard images without a portion of the wind turbine and a region of a plant farm to generate a subset of images;

an artificial intelligence (AI) ensemble engine operating on one or more computing devices having a memory storing executable instructions executed on a processor that cause the AI to:

distinguish features of the portion of the wind turbine and the region of the plant farm;

analyze the features of the subset of images to identify potential problem areas with the wind turbine and the plant farm; and

analyze the potential problem areas of the wind turbine and the plant farm to determine whether maintenance is warranted for the potential problem areas to generate a list of problem areas; and

a notification manager operating on the one or more computing devices that determines whether each problem area in the list of problem areas is related to a wind turbine or agriculture of the plant farm and generates a notification to a turbine operator or a plant farm operator based on the determination.

11. The system of claim 10 , wherein the notification manager provides the notification to a ticket generator interface in response to determining that a given problem area in the list of problem areas is related to a wind turbine, wherein the notification includes a location of the wind turbine extracted from a given image of the subset of images.

12. The system of claim 11 , wherein the ticket generator interface generates a service ticket with an order to dispatch a maintenance crew in response to the notification.

13. The system of claim 11 , wherein the given problem area is a crack in the wind turbine.

14. The system of claim 11 , wherein the given problem area is a hot spot on the wind turbine.

15. The system of claim 11 , wherein the notification manager provides a short message service (SMS) message and/or an email to the plant farm operator in response to determining that a given problem area in the list of problem areas is related to the plant farm, wherein the notification includes an image of the given problem area extracted from the subset of images.

16. The system of claim 15 , wherein the given problem area is a dead spot on the plant farm.

17. The system of claim 15 , wherein the given problem area is an erosion region on the plant farm.

18. A method for processing images comprising:

receiving, by an image processing system operating on one or more computing devices having a memory storing executable instructions to be executed by a processor, a set of images captured by a drone surveying a wind farm with wind turbines and/or a plant farm;

parsing, by the image processing system, a set of images to discard images without a portion of the wind turbine and a region of a plant farm to generate a subset of images;

identifying, by an artificial intelligence (AI) ensemble engine operating on the one or more computing devices, potential problem areas related to wind turbines or plant farms in the subset of images, wherein the AI ensemble engine is trained to recognize features related to wind turbines and plant farms;

analyzing, by a feature analyzer of the AI ensemble engine operating on the one or more computing devices, the potential problem areas to determine problem areas related to the wind turbines and/or the plant farm; and

determining, by a notification manager operating on the one or more computing devices, whether maintenance is needed for each problem area and generating a notification to a turbine operator or a plant farm operator based on the determination.

19. The method of claim 18 , further comprising generating, by the notification manager, a notification that includes an image of a problem area and a location of the problem area.

20. The medium of claim 1 , wherein features are low-level features, and the AI ensemble engine further:

combines low-level features to form high-level features and produce a distributed feature representation, and

identifying the high-level features as related to the portion of the wind turbine or the region of the plant farm based on the distributed feature representation.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 22, 2022
From: VIDHI, RACHANA; SHRIVASTAVA, PRASANNA
To: INVENTUS HOLDINGS, LLC
Reel/Frame 059065/0483 →
Continuity (1)
Related Publication 20230267364A1 · Aug 24, 2023
References Cited (21)
US 8545187B2 · Presz, Jr. · 2013 [cited by examiner]
US 9470579B2 · Ritter et al. · 2016 [cited by applicant]
US 10231441B2 · Horton · 2019 [cited by applicant]
US 20120136630A1 · Murphy · 2012 [cited by applicant]
US 20150278640A1 · Johnson · 2015 [cited by examiner]
US 20180003161A1 · Michini et al. · 2018 [cited by applicant]
US 20180230967A1 · Beatrice · 2018 [cited by examiner]
US 20180286141A1 · Van de Cotte · 2018 [cited by examiner]
US 20200049129A1 · Girardot · 2020 [cited by examiner]
US 20200225655A1 · Cella · 2020 [cited by examiner]
US 20200327264A1 · Wang · 2020 [cited by examiner]
US 20210110724A1 · Wake et al. · 2021 [cited by applicant]
US 20210210170A1 · Gardner · 2021 [cited by examiner]
US 20210246881A1 · Priore, III · 2021 [cited by examiner]
US 20220065228A1 · Ponnada · 2022 [cited by examiner]
EP 2527649A1 · 2012 [cited by applicant]
WO WO2020156629A1 · 2020 [cited by examiner]
“Shang Zhang, an Improved Algorithm for Wind Turbine Blade Defect Detection, Oct. 18, 2022, IEEE Access, Version 23, pp. 1-11” (Year: 2022). [cited by examiner]
Stetco, Adrian, et al., “Machine learning methods for wind turbine condition monitoring: A review”, Renewable Energy 133 (2019) 620-635. [cited by applicant]
Yuan, Yuan, et al., “Advanced agricultural disease image recognition technologies: A review”, Information Processing in Agriculture, pp. 1-12, available at www.sciencedirect.com. [cited by applicant]
Ciaburro, Giuseppe, et al., “Machine Learning-Based Tools for Wind Turbine Acoustic Monitoring”, Appl. Sci. 2021, 11, 6488, https:/doi.org/10.3390/app11146488, pp. 1-17. [cited by applicant]