US 11,727,568 B2
Rapid illness screening of a population using computer vision and multispectral data
Justin Starr, Baden, PA (US); Ashish Basuray, Petaluma, CA (US); Roman Kazantsev, Fremont, CA (US); and Randolf Reiss, Dripping Springs, TX (US)
Assigned to JAR Scientific, LLC, West Mifflin, PA (US)
Filed by JAR Scientific LLC, West Mifflin, PA (US)
Filed on Mar. 23, 2021, as Appl. No. 17/209,999.
Claims priority of provisional application 62/993,856, filed on Mar. 24, 2020.
Prior Publication US 2021/0304406 A1, Sep. 30, 2021
Int. Cl. G06T 1/00 (2006.01); G06T 7/00 (2017.01); G06F 3/01 (2006.01); G01K 1/02 (2021.01)
CPC G06T 7/0014 (2013.01) [G06F 3/017 (2013.01); G01K 1/026 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30196 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A method, comprising:
obtaining, using a camera system comprising two or more cameras deployed in two or more locations of an environment, imagery of one or more individuals in an environment;
analyzing, using a processor, the imagery using a trained model to identify at least one of the one or more individuals and recognize gestures of the at least one individual that are indicative of a respiratory sickness behavior above a calculated threshold;
the respiratory sickness behavior comprising one or more of coughing, touching a predefined area of the body, and gestural movements to indicate a restricted airflow;
tracking, using the camera system, the at least one of the one or more individuals displaying the respiratory sickness behavior across the two or more locations;
presenting, using a display device, information about the at least one individual being tracked and displaying the respiratory sickness behavior above the calculated threshold.