US 12,307,813 B2
Systems and methods of detecting incorrect mask usage
Terence Neill, Lisburn (GB); and Alexis B. Ehrlich, Boca Raton, FL (US)
Assigned to Tyco Fire & Security GmbH, Neuhausen am Rheinfall (CH)
Filed by Tyco Fire & Security GmbH, Neuhausen am Rheinfall (CH)
Filed on Feb. 19, 2021, as Appl. No. 17/180,074.
Prior Publication US 2022/0269878 A1, Aug. 25, 2022
Int. Cl. G06V 40/16 (2022.01); G08B 21/18 (2006.01)
CPC G06V 40/171 (2022.01) [G08B 21/18 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A method of detecting incorrect mask usage, comprising:
monitoring video frames of one or more video feeds of an environment comprising an enforcement area and at least one non-enforcement area;
detecting, via image processing, one or more persons in the video frames;
determining, via image processing, that the one or more persons is inside of the enforcement area of the environment based on a location of the one or more persons within a first area of the video frames associated with the enforcement area and not a second area of the video frames associated with the at least one non-enforcement area;
determining, via image processing, based at least in part on the one or more persons being inside the enforcement area, whether the one or more persons are in violation of a mask wearing policy indicative of an occupancy rate threshold formed by modifying a maximum occupancy rate of the enforcement area by a person-to-person health preserving minimum spacing distance; and
generating an alert in response to determining that the one or more persons are in violation of the mask wearing policy,
wherein the one or more video feeds are captured by one or more cameras that are calibrated with a camera enrollment process that uses one or more reference objects of a known physical size and known distance from the one or more cameras by an image processing algorithm that provides ratios for converting pixel distances of video frames to physical distances in the environment.