US 11,972,624 B2
Delivery system
Peter Douglas Jackson, Alpharetta, GA (US); Robert Lee Martin, Jr., Kenosha, WI (US); Daniel James Thyer, Charlotte, NC (US); and Justin Michael Brown, Coppell, TX (US)
Assigned to Rehrig Pacific Company, Los Angeles, CA (US)
Filed by Rehrig Pacific Company, Los Angeles, CA (US)
Filed on Jul. 13, 2023, as Appl. No. 18/221,575.
Application 18/221,575 is a continuation of application No. 17/978,662, filed on Nov. 1, 2022, granted, now 11,783,606.
Claims priority of provisional application 63/274,303, filed on Nov. 1, 2021.
Claims priority of provisional application 63/274,310, filed on Nov. 1, 2021.
Prior Publication US 2023/0360419 A1, Nov. 9, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 30/148 (2022.01); B65B 11/04 (2006.01)
CPC G06V 30/153 (2022.01) [B65B 11/045 (2013.01)] 24 Claims
OG exemplary drawing
 
8. A computing system for identifying SKUs in a stack of a plurality of packages comprising:
at least one processor; and
at least one non-transitory computer-readable media storing:
at least one machine learning model that has been trained with a plurality of images of packages; and
instructions that, when executed by the at least one processor, cause the computing system to:
a) receive a plurality of stack images of the stack of the plurality of packages, wherein each of the plurality of stack images is of a different face of the stack of the plurality of packages;
b) segment the plurality of stack images into a plurality of package faces, including first package faces in a first layer of the stack of the plurality of packages;
c) using the at least one machine learning model, infer a brand based upon each of the first package faces;
d) based upon step b), detect a hidden package in the first layer of the stack of the plurality of packages, wherein the hidden package is not visible in any of the plurality of stack images; and
e) based upon steps a) to d), associate a hidden package brand with the hidden package.