US 11,727,697 B2
Object detection and image classification based optical character recognition
Dennis Schultz, St. Louis, MO (US); Daniel Thomas Harrison, Newmarket (CA); Christopher Anthony Kemp, Toronto (CA); and Michael A. Salem, Dallas, TX (US)
Assigned to Salesforce, Inc., San Francisco, CA (US)
Filed by salesforce.com, inc., San Francisco, CA (US)
Filed on Jun. 30, 2021, as Appl. No. 17/364,343.
Application 17/364,343 is a continuation of application No. 16/773,813, filed on Jan. 27, 2020, granted, now 11,100,319.
Prior Publication US 2021/0326589 A1, Oct. 21, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/62 (2022.01); G06V 30/413 (2022.01); G06V 30/19 (2022.01); G06V 30/10 (2022.01)
CPC G06V 20/63 (2022.01) [G06V 30/19173 (2022.01); G06V 30/413 (2022.01); G06V 30/10 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer implemented method for optical character recognition, comprising:
receiving an image displaying one or more labels;
detecting the one or more labels by providing the image as input to one or more object detection models, wherein an object detection model is configured to identify a label of a label type in an input image;
for a detected label in the image:
determining a text of the detected label using optical character recognition;
accessing expected phrase types for phrases of a label type of the detected label;
analyzing the text based on an expected phrase type to identify discrepancies in the text;
responsive to identifying a discrepancy in a text of a particular label, sending a recommendation for improving the image.