US 11,809,999 B2
Object recognition scanning systems and methods for implementing artificial based item determination
Edward Barkan, Miller Place, NY (US); Mark Drzymala, Saint James, NY (US); and Darran Michael Handshaw, Sound Beach, NY (US)
Assigned to Zebra Technologies Corporation, Lincolnshire, IL (US)
Filed by ZEBRA TECHNOLOGIES CORPORATION, Lincolnshire, IL (US)
Filed on Feb. 24, 2020, as Appl. No. 16/799,317.
Prior Publication US 2021/0264215 A1, Aug. 26, 2021
Int. Cl. G06K 9/62 (2022.01); G06F 16/583 (2019.01); G06N 3/08 (2023.01); G06N 7/00 (2023.01); G06K 9/03 (2006.01); G06N 3/088 (2023.01); G06F 18/21 (2023.01); G06N 7/01 (2023.01); G06V 30/12 (2022.01); G06V 30/19 (2022.01); G06V 30/10 (2022.01)
CPC G06N 3/088 (2013.01) [G06F 16/5854 (2019.01); G06F 18/217 (2023.01); G06N 3/08 (2013.01); G06N 7/01 (2023.01); G06V 30/12 (2022.01); G06V 30/19173 (2022.01); G06V 30/10 (2022.01)] 20 Claims
OG exemplary drawing
 
1. An object recognition scanning system comprising:
an imager having a field of view (FOV) extending over a scanning area, the imager configured to image one or more items within the FOV;
one or more processors configured to receive image data of an imaged item imaged by the imager during a scanning session; and
an object recognition model stored in a memory communicatively coupled to the one or more processors,
the memory storing instructions that, when executed by the one or more processors, cause the one or more processors to:
determine, by the object recognition model taking the image data as input, a product identification probability for the imaged item, the product identification probability being indicative of a level of confidence that the imaged item is one of multiple items stored in the object recognition model;
responsive to the product identification probability meeting or exceeding a first product identification threshold value and being less than a second product identification threshold value, determine a product type based on identifying at least a portion of a barcode within the image data and transmit the product type to a host when the product type matches the one of multiple items stored in the object recognition model; and
responsive to the product identification probability meeting or exceeding the second product identification threshold value, determine the product type based on the one of multiple items stored in the object recognition model and transmitting the product type to the host without regard for the at least a portion of a barcode.