Non-contact temperature measurement in thermal imaging systems and methods
Systems and methods include an image capture component configured to capture infrared images of a scene, and a logic device configured to identify a target in the images, acquire temperature data associated with the target based on the images, evaluate the temperature data and determine a corresponding temperature classification, and process the identified target in accordance with the temperature classification. The logic device identifies a person and tracks the person across a subset of the images, identify a measurement location for the target in a subset of the images based on target feature points identified by a neural network, and measure a temperature of the location using corresponding values from one or more captured thermal images. The logic device is further configured calculate a core body temperature of the target using the temperature data to determine whether the target has a fever and calibrate using one or more black bodies.
1. A method comprising:
identifying a target in a plurality of infrared images;
acquiring temperature data associated with the target based, at least in part, on the infrared images, wherein the acquiring comprises sampling temperature data from a measurement location over time and comparing a target temperature trajectory to one or more trajectory models;
evaluating the temperature data to determine a corresponding temperature classification, wherein the evaluating comprises applying a corresponding trajectory model to the sampled temperature data and applying the model to project a core body temperature;
processing the identified target in accordance with the temperature classification; and
detecting a trajectory alarm condition for the sampled temperature data following a corresponding trajectory model.
2. The method of claim 1 , wherein the identifying comprises analyzing the infrared images using a convolutional neural network (CNN), the method further comprising:
training the CNN to identify facial feature points comprising a chin, a mouth, a nose, a left eye, and/or a right eye;
training the CNN to identify a distance between the target and an image capture device; and
training the CNN to identify one or more objects in the image comprising a mask, eye glasses, and/or a hat.
3. The method of claim 1 , wherein:
the identifying comprises identifying one or more facial feature points for the target;
the acquiring comprises identifying a region of interest based on the one or more facial feature points and measuring a temperature of the region of interest using corresponding values from one or more of the infrared images; and
the evaluating comprises measuring a head pose orientation based on the one or more facial feature points to determine whether the target is in a measurement position.
4. The method of claim 1 , wherein:
the temperature data is associated with the measurement location on the target comprising a cheek, a tear duct, and/or a forehead;
the evaluating comprises correlating the measurement location with the core body temperature of the target; and
the method further comprises analyzing a visual image of the target to identify the measurement location.
5. The method of claim 1 , further comprising:
disposing a passive black body in view of a thermal camera configured to capture the infrared images;
disposing a temperature sensor adjacent to the passive black body;
sensing a temperature associated with the black body;
wirelessly transmitting the sensed temperature to the thermal camera; and
calibrating the thermal camera using the sensed temperature and at least one of the infrared images comprising the passive black body.
6. The method of claim 1 , further comprising:
measuring a temperature of a surface with a known temperature, wherein the surface is a black body, an object with a known temperature, and/or a device that measures ambient temperature; and
adjusting the temperature data using the measured temperature to account for measurement error.
7. The method of claim 1 , further comprising:
providing a hollow body in view of a thermal camera that captures the infrared images, the hollow body comprising a high emissivity surface at an end facing the thermal camera; and
processing the image of the high emissivity surface as a black body.
8. The method of claim 1 , wherein:
the acquiring comprises:
capturing a plurality of visible light images with a visible light camera,
tracking the target in the visible light images,
identifying the measurement location on the target, and capturing the infrared images with a thermal camera; and
the visible light camera has a wider field of view than the thermal camera.
9. The method of claim 8 , further comprising:
instructing the thermal camera to zoom towards the measurement location; and
wherein the measurement location comprises more pixels in the infrared images than in the visible light images.
10. The method of claim 1 , wherein the identifying comprises:
capturing a plurality of visible light images of a scene;
detecting a desired measurement location in the visible light images;
acquiring the infrared images of the desired measurement location;
analyzing the infrared images to determine a plurality of landmarks within the measurement location;
selecting one of the landmarks based on a likelihood of accurate temperature measurement;
establishing a region of interest associated with the selected landmark;
measuring a maximum temperature within the region of interest; and
updating reference temperature data for the scene using the measured temperature.
11. The method of claim 1 , further comprising: determining a plurality of temperature profiles associated with a plurality of paths for a target within a scene, wherein one or more of the paths extend across multiple scenes captured by a plurality of imaging devices; determining one of the paths associated with the target; and adjusting the temperature data in accordance with the temperature profile for the determined one of the paths.
12. The method of claim 1 , further comprising:
acquiring reference temperature samples by rendering a user interface having user prompts configured to guide the user to acquire accurate reference temperature samples;
rendering visual indicia associated with a desired target comprising an outline of a head and/or an alignment box, wherein the visual indicia adapts to a target distance and/or angle; and
rendering an alarm indication if the target is determined to have a fever.
13. A system configured to perform the method of claim 1 , the system comprising:
a thermal camera configured to capture the infrared images; and
a logic device configured to perform the identifying, the acquiring, the evaluating, and the processing.
14. The system of claim 13 , further comprising:
a passive black body configured to be disposed in view of the thermal camera;
a temperature sensor configured to be disposed adjacent to the passive black body and to sense a temperature associated with the black body;
a wireless interface configured to transmit the sensed temperature to the thermal camera; and
wherein the logic device is configured to calibrate the thermal camera using the sensed temperature and at least one of the infrared images comprising the passive black body.
15. A method comprising:
identifying a target in a plurality of infrared images;
acquiring temperature data associated with the target based, at least in part, on the infrared images;
evaluating the temperature data to determine a corresponding temperature classification;
processing the identified target in accordance with the temperature classification;
selecting a first plurality of measurement locations in one or more captured thermal images;
measuring local contrast of the first measurement locations and excluding measurement locations having a contrast higher than a high contrast threshold, resulting in a second plurality of measurement locations;
measuring a signal at each of the second measurement locations over time;
excluding measurement locations having variation over time over a variation threshold, resulting in a third plurality of measurement locations;
monitoring the signal at each of the third measurement locations; and
using the monitored signal to reduce error in signal measurement due to system drift.
16. A system configured to perform the method of claim 15 , the system comprising:
a thermal camera configured to capture the infrared images; and
a logic device configured to perform the identifying, the acquiring, the evaluating, and the processing.
17. A method comprising:
identifying a target in a plurality of infrared images;
acquiring temperature data associated with the target based, at least in part, on the infrared images;
evaluating the temperature data to determine a corresponding temperature classification;
processing the identified target in accordance with the temperature classification;
initializing a temperature distribution associated with a fever;
setting an isotherm alarm covering the temperature distribution;
capturing a thermal image and performing face detection in the thermal image in response to the isotherm alarm;
measuring a temperature of the face; and
updating the temperature distribution if the measured temperature is above a threshold.
18. The method of claim 17 , wherein:
the target is a person;
the identifying comprises identifying the person and tracking the person across a subset of the infrared images;
the acquiring comprises identifying a measurement location for the person in the subset of the infrared images and determining the temperature data associated with the location using corresponding values from one or more of the infrared images; and
the evaluating comprises calculating a core body temperature of the person using the temperature data.
19. The method of claim 17 , wherein:
the acquiring comprises analyzing thermal data acquired from a measurement location, environmental data associated with a camera location, crowd temperature data associated with a plurality of people associated with the camera location, and/or target temperature data over time using a convolutional neural network (CNN) trained to generate a core body temperature;
the evaluating comprises using a convolutional neural network (CNN) trained to classify the target as one of having a fever or not having a fever; and
the processing comprises generating an alarm condition if the temperature classification indicates the target has a fever.
20. A system configured to perform the method of claim 17 , the system comprising:
a thermal camera configured to capture the infrared images; and
a logic device configured to perform the identifying, the acquiring, the evaluating, and the processing.