Methods and systems for temperature measurement with machine learning algorithm

Patent metadata

Field	Value
Publication number	US-10823618-B2
Application number	US-201815879878-A
Country	US
Kind code	B2
Filing date	Jan 25, 2018
Priority date	Jan 25, 2018
Publication date	Nov 3, 2020
Grant date	Nov 3, 2020

Abstract

A method includes capturing an image, and performing region detection on the captured image. The region detection includes identifying an object represented in the captured image. The method further includes detecting emissivity of the identified object and determining the temperature of the object based on the detected emissivity of the object.

First claim

What is claimed is:

1. A method, comprising:
capturing an image;
performing region detection on the captured image, said region detection including identifying an object represented in the captured image;
detecting emissivity of the identified object; and
determining a temperature of the object based on the detected emissivity of the object;
wherein said detecting step includes looking up emissivity data from an emissivity data store based on a result of identifying said object.

2. The method of claim 1 , wherein the identifying of the object includes using a machine learning algorithm.

3. The method of claim 2 , wherein the machine learning algorithm comprises R-CNN (region proposal with a convolutional neural network).

4. The method of claim 1 , wherein:
the identified object is a first object;
the region detection includes identifying a second object represented in the captured image, the second object different from the first object;
the method further comprising:
detecting emissivity of the second object; and
determining a temperature of the second object based on the detected emissivity of the second object;
wherein the determined temperature of the second object is different from the determined temperature of the first object.

5. The method of claim 4 , wherein the region detection includes identifying a third object represented in the captured image, the third object different from the first and second objects;
the method further comprising:
detecting emissivity of the third object; and
determining a temperature of the third object based on the detected emissivity of the third object;
wherein the determined temperature of the third object is different from the determined temperature of the second object and is also different from the determined temperature of the first object.

6. The method of claim 5 , further comprising:
using results of the temperature of the first, second and third objects to synthesize a temperature map of the captured image; the first, second and third objects represented in the temperature map.

7. The method of claim 1 , wherein the captured image is at least partially an infra-red image.

8. The method of claim 7 , wherein the captured image is partially a visible light image.

9. The method of claim 1 , wherein the captured image is a composite image formed from respective outputs of two digital cameras.

10. An optical temperature-detection apparatus, comprising:
a wavelength splitter;
a digital camera oriented to capture an image of an object through the wavelength splitter;
a signal processing unit in communication with the digital camera to receive image data from the digital camera; and
an emissivity data storage unit in communication with the signal processing unit to supply emissivity data to the signal processing unit.

11. The apparatus of claim 10 , wherein the signal processing unit is configured to perform region detection with respect to the received image data, said region detection including identifying the object.

12. The apparatus of claim 11 , wherein:
the object is a first object; and
said region detection includes identifying a second object represented by the image data.

13. The apparatus of claim 12 , wherein the signal processing unit receives first emissivity data and second emissivity data from the emissivity data storage unit; the first emissivity data indicative of an emissivity of the first object; the second emissivity data indicative of an emissivity of the second object.

14. The apparatus of claim 13 , wherein the signal processing unit is configured to determine a temperature of the first object based in part on the first emissivity data and to determine a temperature of the second object based in part on the second emissivity data.

15. The apparatus of claim 10 , further comprising:
a timing unit coupled to the digital camera for controlling a timing at which the digital camera captures image data with respect to the object.

16. An optical temperature-detection apparatus, comprising:
a digital camera;
a processor coupled to the digital camera to receive image data from the digital camera; and
a memory in communication with the processor, the memory storing program instructions, the processor operative with the program instructions to perform functions as follows:
performing region detection on an image represented by the image data, said region detection including identifying an object represented in the image;
detecting emissivity of the identified object; and

determining a temperature of the object based on the detected emissivity of the object;
wherein said detecting function includes looking up emissivity data from an emissivity data store based on a result of identifying said object.

17. The apparatus of claim 16 , wherein the identifying of the object includes using a machine learning algorithm.

18. The apparatus of claim 17 wherein said using the machine learning algorithm includes training the machine learning algorithm.

19. The apparatus of claim 17 wherein said using the machine learning algorithm includes operating the machine learning algorithm in an inference mode.

20. The apparatus of claim 17 , wherein the machine learning algorithm comprises R-CNN (region proposal with a convolutional neural network).

21. The apparatus of claim 16 , wherein:
the identified object is a first object;
the region detection includes identifying a second object represented in the image data;
the processor further operative with the program instructions to perform functions as follows:
detecting emissivity of the second object; and
determining a temperature of the second object based on the detected emissivity of the second object.

22. The apparatus of claim 21 , wherein the region detection includes identifying a third object represented in the image data;
the processor further operative with the program instructions to perform functions as follows:
detecting emissivity of the third object; and
determining a temperature of the third object based on the detected emissivity of the third object.

Assignees

• Gen Electric

Inventors

• Wang Guanghua
• Bewlay Bernard

Classifications

• G01J5/605 Primary

  using visual determination · CPC title

• G01J5/0003 Primary

  for sensing the radiant heat transfer of samples, e.g. emittance meter · CPC title

• G06V20/00

  Scenes; Scene-specific elements (control of digital cameras H04N23/60) · CPC title

• G06V10/143

  Sensing or illuminating at different wavelengths · CPC title

• G06V20/10

  Terrestrial scenes (scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59) · CPC title