US 12,307,770 B2
Method for detecting and responding to conditions within a space
Kelby Green, Mountain View, CA (US); Kanav Dhir, Mountain View, CA (US); Dan Ryan, Mountain View, CA (US); Damien Leostic, Mountain View, CA (US); Marine Dunoguier, Mountain View, CA (US); and Habib Bukhari, Mountain View, CA (US)
Assigned to VergeSense, Inc., Mountain View, CA (US)
Filed by VergeSense, Inc., Mountain View, CA (US)
Filed on Apr. 13, 2023, as Appl. No. 18/134,490.
Claims priority of provisional application 63/330,642, filed on Apr. 13, 2022.
Prior Publication US 2023/0334863 A1, Oct. 19, 2023
Int. Cl. G06V 20/40 (2022.01); G06T 7/70 (2017.01); G06V 10/20 (2022.01); G06V 10/40 (2022.01); G06V 20/52 (2022.01); G06V 20/70 (2022.01); G08B 23/00 (2006.01)
CPC G06V 20/44 (2022.01) [G06T 7/70 (2017.01); G06V 10/255 (2022.01); G06V 10/40 (2022.01); G06V 20/52 (2022.01); G06V 20/70 (2022.01); G08B 23/00 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for detecting conditions within a space comprising:
during a first period, at a first sensor block in a population of sensor blocks arranged in the space, capturing a first image via an optical sensor arranged in the first sensor block;
extracting a first set of features from the first image;
based on the first set of features:
detecting a first set of objects in a first field of view of the first sensor block during the first time period; and
deriving a location and an object type of each object in the first set of objects; and
compiling locations and object types of the first set of objects into a first object list; and
during a second time period, at a remote computer system:
accessing a corpus of object lists, the corpus of object lists comprising the first object list and generated based on images captured by the population of sensor blocks;
compiling locations of a second set of objects represented in the corpus of object lists into a map of the space based on known locations of the population of sensor blocks;
accessing a nominal condition of the space, the nominal condition defining:
a set of inclusion objects within a threshold distance of an anchor object type; and
a set of exclusion objects within the threshold distance of the anchor object type and distinct from the set of inclusion objects;
detecting the anchor object type in the map; and
in response to detecting a subset of objects, in the second set of objects, within the threshold distance of the anchor object type in the map and deviating from the set of inclusion objects and the set of exclusion objects:
identifying the subset of objects, proximal a first location in the map, as anomalous; and
generating a notification to investigate the first location for the subset of objects.