US 12,242,452 B2
Text-based machine learning extraction of table data from a read-only document
Hongyang Yu, Wentworth Point (AU); Hanieh Borhanazad, Sydney (AU); and Sandip Mandlecha, Pune (IN)
Assigned to Coupa Software Incorporated, San Mateo, CA (US)
Filed by Coupa Software Incorporated, San Mateo, CA (US)
Filed on Jan. 17, 2024, as Appl. No. 18/415,062.
Application 18/415,062 is a continuation of application No. 17/973,511, filed on Oct. 25, 2022, granted, now 11,914,567.
Application 17/973,511 is a continuation of application No. 17/074,957, filed on Oct. 20, 2020, granted, now 11,500,843, issued on Nov. 15, 2022.
Claims priority of application No. 202011037847 (IN), filed on Sep. 2, 2020.
Prior Publication US 2024/0160616 A1, May 16, 2024
Int. Cl. G06F 16/22 (2019.01); G06F 16/93 (2019.01); G06N 3/045 (2023.01); G06V 30/412 (2022.01); G06V 30/414 (2022.01)
CPC G06F 16/2282 (2019.01) [G06F 16/93 (2019.01); G06N 3/045 (2023.01); G06V 30/412 (2022.01); G06V 30/414 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
extracting text rectangle (TR)-level numerical data and grid data from a digital electronic document using one or more neural networks;
generating a feature map by projecting the TR-level numerical data onto a grid defined by the grid data and de-biasing one or more long text rectangle data items of the TR-level numerical data;
generating row-level numerical data by processing the feature map using a first-running convolutional filter;
generating a filtered feature map by filtering the feature map using a grid filter;
generating column-level numerical data by processing the filtered feature map using a second-running convolutional filter; and
converting the row-level numerical data and the column-level numerical data to a structured data format.