US 12,437,567 B2
Computer-implemented method for optical character recognition
Daniel Albertini, Vienna (AT); David Dengg, Vienna (AT); Martin Cerman, Vienna (AT); Peter Sperl, Vienna (AT); and Hannes Duchkowitsch, Vienna (AT)
Assigned to ANYLINE GmbH, Vienna (AT)
Appl. No. 17/754,198
Filed by ANYLINE GmbH, Vienna (AT)
PCT Filed Sep. 30, 2020, PCT No. PCT/EP2020/077362
§ 371(c)(1), (2) Date Mar. 25, 2022,
PCT Pub. No. WO2021/064018, PCT Pub. Date Apr. 8, 2021.
Claims priority of application No. 19200582 (EP), filed on Sep. 30, 2019.
Prior Publication US 2022/0327849 A1, Oct. 13, 2022
Int. Cl. G06K 9/00 (2022.01); G06V 10/82 (2022.01); G06V 30/10 (2022.01); G06V 30/19 (2022.01)
CPC G06V 30/10 (2022.01) [G06V 10/82 (2022.01); G06V 30/1916 (2022.01)] 11 Claims
OG exemplary drawing
 
1. A computer-implemented method for optical character recognition, the method comprising:
receiving image data of an input image having one or more characters;
applying two or more different data processing operations to the received image data,
wherein at least one of the two or more different data processing operations is configured to return the one or more characters of the input image,
wherein each data processing operation is selected from a predefined set of available data processing functions according to a processing configuration,
wherein the available data processing functions are precompiled functions; and
before applying the two or more different data processing operations, loading the processing configuration and a network configuration from a remote service at runtime,
wherein the processing configuration determines a sequence of the two or more different data processing operations, and
wherein at least one of the two or more different data processing operations includes propagating data through one or more artificial neural networks configured by the loaded network configuration.