US 11,900,705 B2
Intelligent engineering data digitization
Neeru Narang, San Jose, CA (US); Guanglei Xiong, Pleasanton, CA (US); Ullas Balan Nambiar, Bangalore (IN); Emmanuel Munguia Tapia, San Jose, CA (US); Ditty Mathew, Vaikom (IN); Omar Razi, Bangalore (IN); David Alfonso Guerra, San Antonio, TX (US); and Thyagarajan Delli, Bangalore (IN)
Assigned to ACCENTURE GLOBAL SOLUTIONS LIMITED, Dublin (IE)
Filed by ACCENTURE GLOBAL SOLUTIONS LIMITED, Dublin (IE)
Filed on Apr. 2, 2021, as Appl. No. 17/221,564.
Prior Publication US 2022/0318546 A1, Oct. 6, 2022
Int. Cl. G06F 17/00 (2019.01); G06V 30/413 (2022.01); G06F 40/169 (2020.01); G06F 40/177 (2020.01); G06V 10/25 (2022.01); G06V 30/414 (2022.01); G06V 30/10 (2022.01)
CPC G06V 30/413 (2022.01) [G06F 40/169 (2020.01); G06F 40/177 (2020.01); G06V 10/25 (2022.01); G06V 30/10 (2022.01); G06V 30/414 (2022.01)] 19 Claims
OG exemplary drawing
 
1. An automatic image validation system, comprising:
at least one processor;
a non-transitory processor-readable medium storing machine-readable instructions that cause the processor to:
separately identify contents of a document package wherein the contents of the document package include at least a first document with one or more tables and a second document including one or more images of engineering drawings with parts, annotation labels for the parts, and specifications of the parts that are to conform to rules codified in the one or more tables;
automatically extract the rules from the one or more tables;
extract the specifications of the parts from the images by:
automatically identifying annotation labels identifying the parts within the engineering drawings;
detecting zone regions in the images corresponding to the annotation labels wherein each zone region includes in entirety, at least one of the annotation labels and specifications of at least one of the parts identified by the at least one annotation label; and
extracting from the images via optical character recognition (OCR), label data of the annotation labels and data elements corresponding to the label data within each of the zone regions as the specifications of the parts;
determine if each of the specifications of the parts conform to the automatically extracted rules; and
provide as a display on a user interface, a validation document including a list of the annotated parts marked as valid, invalid, or partial match based on confidence scores indicative of the conformance of the specifications of the annotated parts to corresponding rules,
the list of annotated parts includes corresponding image widgets wherein selection of one of the annotated parts from the list of annotated parts activates the corresponding image widget that causes a display of a corresponding portion from the engineering drawings that includes the selected annotated part.