IP Library Granted Patent US 11,830,172
Granted Patent B1
US 11,830,172 · App. 17/691,786 · Granted Nov 28, 2023

Thermal anomaly detection and classification

Inventors: Clyde Bennet Wheeler (Santa Fe, NM); Hang Zhou (Santa Fe, NM)
Assignee: Descartes Labs, Inc.
G06T5/50G01J5/025H04N23/11G01J2005/0077G06T2207/10048G06T2207/30188
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 11,830,172
App. No.
17/691,786
Granted
Nov 28, 2023
Kind
B1
Abstract

A set of images is received. A thermal anomaly is detected in the received set of images. The detected thermal anomaly is classified. Examples of classifications of thermal anomalies include: stationary targets, gas flares, agricultural anomalies, and wildfires. An action is taken based at least in part on the classified detected anomaly.

Claims (27)

1. A system, comprising: a processor configured to: receive a set of images comprising multi-spectral band data, wherein a first image included in the received set of images is associated with location information; detect, at a first time, a thermal anomaly using the received set of images; determine that the detected thermal anomaly is a first ignition event having an associated detection time and associated location and, in response to the determination, generate an alert; and detect, at a subsequent time, a second ignition event, and selectively take an action regarding the detection of the second ignition event based at least in part on a determination of whether the second ignition event is associated with the first ignition event by updating boundaries of an existing fire in response to determining that the second ignition event is associated with the first ignition event; and a memory coupled to the processor and configured to provide the processor with instructions.

2. The system of claim 1 , wherein the processor is further configured to determine whether the detected thermal anomaly is associated with a stationary target.

3. The system of claim 1 , wherein the processor is further configured to determine whether the detected thermal anomaly is a gas flare.

4. The system of claim 1 , wherein the processor is further configured to determine whether the detected thermal anomaly is associated with agriculture.

5. The system of claim 1 , wherein the processor is further configured to determine whether the detected thermal anomaly is associated with a wildfire.

6. The system of claim 1 , wherein detecting the thermal anomaly includes applying a cloud mask.

7. The system of claim 6 , wherein the processor is further configured to generate the cloud mask.

8. The system of claim 1 , wherein the processor is further configured to generate a fire mask.

9. A method, comprising: receiving a set of images comprising multi-spectral band data, wherein a first image included in the received set of images is associated with location information; detecting, at a first time, a thermal anomaly using the received set of images; determining that the detected thermal anomaly is a first ignition event having an associated detection time and associated location and, in response to the determination, generating an alert; and detecting, at a subsequent time, a second ignition event, and selectively taking an action regarding the detection of the second ignition event based at least in part on a determination of whether the second ignition event is associated with the first ignition even by updating boundaries of an existing fire in response to determining that the second ignition event is associated with the first ignition event.

10. A computer program product embodied in a non-transitory computer readable storage medium and comprising computer instructions for: receiving a set of images comprising multi-spectral band data, wherein a first image included in the received set of images is associated with location information; detecting, at a first time, a thermal anomaly using the received set of images; determining that the detected thermal anomaly is a first ignition event having an associated detection time and associated location and, in response to the determination, generating an alert; and detecting, at a subsequent time, a second ignition event, and selectively taking an action regarding the detection of the second ignition event based at least in part on a determination of whether the second ignition event is associated with the first ignition event by updating boundaries of an existing fire in response to determining that the second ignition event is associated with the first ignition event.

11. The system of claim 8 , wherein the fire mask indicates, for each pixel in a tile, whether the pixel represents a fire or not.

12. The system of claim 8 , wherein the processor is further configured to draw one or more boundaries around fire pixels.

13. The system of claim 8 , wherein the processor is further configured to generate a set of fire segments.

14. The system of claim 13 , wherein the processor is further configured, for a given segment, to determine whether the segment is associated with an existing event.

15. The system of claim 1 , wherein selectively taking the action includes generating an alert in response to determining that the second ignition event is not associated with the first ignition event.

16. The method of claim 9 , further comprising determining whether the detected thermal anomaly is associated with a stationary target.

17. The method of claim 9 , further comprising determining whether the detected thermal anomaly is a gas flare.

18. The method of claim 9 , further comprising determining whether the detected thermal anomaly is associated with agriculture.

19. The method of claim 9 , further comprising determining whether the detected thermal anomaly is associated with a wildfire.

20. The method of claim 9 , wherein detecting the thermal anomaly includes applying a cloud mask.

21. The method of claim 20 , further comprising generating the cloud mask.

22. The method of claim 9 , further comprising generating a fire mask.

23. The method of claim 22 , wherein the fire mask indicates, for each pixel in a tile, whether the pixel represents a fire or not.

24. The method of claim 22 , further comprising drawing one or more boundaries around fire pixels.

25. The method of claim 22 , further comprising generating a set of fire segments.

26. The method of claim 25 , further comprising, for a given segment, determining whether the segment is associated with an existing event.

27. The method of claim 9 , wherein selectively taking the action includes generating an alert in response to determining that the second ignition event is not associated with the first ignition event.

Assignments (4)
SECURITY INTEREST Recorded Jun 10, 2025
From: GEOSYS-INTL, INC.; EARTHDAILY ANALYTICS USA, INC.; EARTHDAILY ANALYTICS CORP.; SKYFOREST INC.
To: TRINITY CAPITAL INC., AS COLLATERAL AGENT
Reel/Frame 071379/0919 →
RELEASE OF SECURITY INTEREST Recorded May 27, 2025
From: DESCARTES DEBT PARTNERS, LLC
To: DESCARTES LABS, INC.
Reel/Frame 071223/0433 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 11, 2025
From: DESCARTES LABS, INC.
To: EARTHDAILY ANALYTICS USA, INC.
Reel/Frame 070478/0433 →
SECURITY INTEREST Recorded Jul 27, 2022
From: DESCARTES LABS, INC.
To: DESCARTES DEBT PARTNERS, LLC
Reel/Frame 060647/0896 →