Dual-band temperature detection systems and methods

Patent metadata

Field	Value
Publication number	US-12430940-B2
Application number	US-202217981337-A
Country	US
Kind code	B2
Filing date	Nov 4, 2022
Priority date	May 8, 2020
Publication date	Sep 30, 2025
Grant date	Sep 30, 2025

Abstract

Systems and methods include an infrared camera configured to capture an infrared image of a scene, a visible-light camera configured to capture a visible-light image of the scene, and a logic device configured to simultaneously capture a pair of images of the scene comprising the infrared image of the scene and the visible image of the scene, align the pair of images so that a pixel location in one of the pair of images has a corresponding pixel location in the other image, classify the visible image, annotate the infrared image based, at least in part, on the classification of the visible image, and add the annotated infrared image to a neural network training dataset for use in training a neural network for infrared image classification. A beamsplitter is arranged to reflect a first image of the scene towards the infrared camera and pass through a second image of the scene to the visible-light camera, and a first blackbody is attached thereto and positioned in a field of view of the infrared camera.

First claim

What is claimed is:

1. A method comprising:
simultaneously capturing a stream of image pairs representing frames of a scene comprising an infrared image of the scene and a visible light image of the scene, wherein the image pairs are captured from substantially a same perspective through separate optical paths of a beamsplitter;
aligning, by a logic device, the images in each of the image pairs so that a pixel location in one of the images of the image pair has a corresponding pixel location in the other image of the image pair;
analyzing, by the logic device, the visible light image to identify an object of interested moving through the scene, including identifying a measurement location on the object appropriate for temperature measurement, wherein the object of interest is a human, and analyzing the visible light image comprises:
using the visible light image to estimate a head size of the human; and
determining a distance to the human based on the estimated head size;

wherein using the visible light image to estimate a head size comprises:
estimating the human's age and/or gender; and
using the human's age and/or gender to estimate the head size; and

wherein the method further comprises calculating, by the logic device, a calibrated temperature of the identified measurement location from the infrared image based at least in part on the determined distance.

2. The method of claim 1 , wherein aligning the image in each of the image pairs so that the pixel location in one of the images has a corresponding pixel location in the other image of the image pair further comprises determining a transformation between the pair of images so that the pixel location in each image represents the same location in the scene; and applying the transformation to the image pair to align the images.

3. The method of claim 1 , wherein analyzing the visible light image further comprises
identifying the human's canthus in the visible light image; and
determining whether sufficient pixels representing the canthus are available in the infrared image to generate a temperature reading of the canthus;
wherein using the visible light image to estimate a head size comprises:
estimating the human's age and gender; and
using the human's age and gender to estimate the head size.

4. The method of claim 1 , wherein simultaneously capturing the stream of image pairs representing frames of the scene further comprises capturing the image pairs of the scene through the beamsplitter, which is arranged to reflect a first image of the scene towards a first camera and pass through a second image of the scene to a second camera.

5. The method of claim 4 , wherein the method is performed by a dual-band camera, the method comprising positioning a first blackbody in a field of view of the first camera, such that the infrared image includes an image of the first blackbody.

6. The method of claim 5 , further comprising positioning the first blackbody at a location proximate to an expected location of the object of interest to approximate an atmospheric transmission of the object of interest in the scene, the first blackbody having an emissivity approximating an expected emissivity of the measurement location on the object of interest; and
calibrating a responsivity of the first camera based on an image of the first blackbody in the image pair.

7. A method comprising:
simultaneously capturing a stream of image pairs representing frames of a scene comprising an infrared image of the scene and a visible light image of the scene, wherein the image pairs are captured from substantially a same perspective through separate optical paths of a beamsplitter;
aligning, by a logic device, the images in each of the image pairs so that a pixel location in one of the images of the image pair has a corresponding pixel location in the other image of the image pair;
analyzing, by the logic device, the visible light image to identify an object of interested moving through the scene, including identifying a measurement location on the object appropriate for temperature measurement, wherein the object of interest is a human, and analyzing the visible light image comprises:
using the visible light image to estimate a head size of the human; and
determining a distance to the human based on the estimated head size;

wherein the method further comprises calculating, by the logic device, a calibrated temperature of the identified measurement location from the infrared image based at least in part on the determined distance;
wherein simultaneously capturing the stream of image pairs representing frames of the scene further comprises capturing the image pairs of the scene through the beamsplitter, which is arranged to reflect a first image of the scene towards a first camera and pass through a second image of the scene to a second camera;
wherein the method is performed by a dual-band camera, the method comprising positioning a first blackbody in a field of view of the first camera, such that the infrared image includes an image of the first blackbody; and
wherein analyzing the infrared image further comprises analyzing the infrared image using a trained neural network based at least in part on the analyzed visible image including calculating a temperature based on a reference offset and a temperature measurement of the first blackbody.

8. The method of claim 5 , wherein the first blackbody is a thermistor.

9. The method of claim 5 , further comprising positioning the first blackbody at a location proximate to the first camera such that the first blackbody would appear out of focus to the first camera; and
positioning a collimating lens between the first blackbody and the first camera to focus its rays for imaging by the first camera; and
wherein the collimating lens is not positioned between the identified measurement location on the object and the first camera.

10. A system comprising:
an infrared camera configured to capture an infrared image of a scene;
a visible-light camera configured to capture a visible-light image of the scene; and
a logic device configured to:
simultaneously capture a stream of image pairs representing frames of the scene comprising the infrared image of the scene and the visible image of the scene, wherein the image pairs are captured from substantially the same perspective through separate optical paths of a beamsplitter;
align the images in each of the image pairs so that a pixel location in one of the images of the image pair has a corresponding pixel location in the other image of the image pair;
analyze the visible light image to identify an object of interest moving through the scene, including identifying a measurement location on the object appropriate for temperature measurement, wherein the object of interest is a human, and the logic device is configured to analyze the visible light image by performing operations comprising:
using the visible light image to estimate a head size of the human; and
determining a distance to the human based on the estimated head size;

wherein using the visible light image to estimate a head size comprises:
estimating the human's age and/or gender; and
using the human's age and/or gender to estimate the head size; and

wherein the logic device is further configured to calculate a calibrated temperature of the identified measurement location from the infrared image based, at least in part, on the analyzed visible light image.

11. The system of claim 10 , wherein the logic device is further configured to
determine a transformation between the images in the pair of images so that the pixel location in each image represents the same location in the scene;

Assignees

• Flir Systems Ab

Inventors

• Richards Austin A
• Tremblay Marcel
• Frank Jeffrey D
• Prieto-Moreno Andres

Classifications

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/10048

  Infrared image · CPC title

• G06T2207/10024

  Color image · CPC title

• G01J2005/202

  Arrays · CPC title