US 12,079,683 B2
Methods and systems of harvesting data for training machine learning (ML) model
Paul R. Poloniewicz, Waxhaw, NC (US)
Assigned to Hand Held Products, Inc., Charlotte, NC (US)
Filed by Hand Held Products, Inc., Charlotte, NC (US)
Filed on Jun. 29, 2023, as Appl. No. 18/343,904.
Application 18/343,904 is a continuation of application No. 17/075,389, filed on Oct. 20, 2020, granted, now 11,734,531.
Prior Publication US 2023/0342575 A1, Oct. 26, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 7/14 (2006.01); G06F 18/214 (2023.01); G06K 7/10 (2006.01); G06N 20/00 (2019.01); G06T 3/60 (2024.01); G06T 7/00 (2017.01); G06T 7/11 (2017.01); G06T 7/70 (2017.01)
CPC G06K 7/1413 (2013.01) [G06F 18/214 (2023.01); G06K 7/10881 (2013.01); G06N 20/00 (2019.01); G06T 3/60 (2013.01); G06T 7/0002 (2013.01); G06T 7/11 (2017.01); G06T 7/70 (2017.01); G06K 2007/10524 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30168 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by a processor, indicia data from an indicia scanner, wherein the indicia data comprises at least decoded data obtained based on decoding an indicium in an image, an image tile comprising a portion of the indicium, and/or location of one or more corners of the portion of the indicium in the image;
generating, by the processor, an image of an ideal indicium based on at least the decoded data;
modifying, by the processor, the image of the ideal indicium to generate a modified image of the ideal indicium;
retrieving, by the processor, a clean image tile, from the modified image of the ideal indicium, wherein the clean image tile comprises a portion of the ideal indicium; and
generating, by the processor, training data, wherein the training data includes the portion of the indicium and the portion of the ideal indicium.