Method and apparatus for form identification and registration employing predefined text grouping

A computer-implemented method of using a machine learning system to identify forms in document and form analysis, the method comprising:

• a) responsive to receiving a form as an input image, retrieving topological information from the form to identify one or more regions in the input image;
• b) in each identified region, identifying one or more fields;
• c) in each identified field, discerning a format of one or more sub-regions;
• d) responsive to the topological information, the identified one or more regions, the identified one or more fields, and the discerned format of the one or more sub-regions, discerning a format each of the one or more fields;
• e) responsive to the discerned format of the one or more fields and of the one or more sub-regions, identifying relative locations of the one or more fields and sub-regions;
• f) responsive to the identification of the relative locations, categorizing the form, wherein a) to f) enable form identification for the document and form analysis without determining precise content of textual information;
• g) responsive to an incorrect categorization of the form, updating the machine learning system;
• and
• h) repeating a) to g) until all forms have been categorized.

A machine learning system to identify forms in document and form analysis, the machine learning system comprising at least one processor and a non-transitory memory that contains instructions that, when executed, enable the machine learning system to perform a method comprising:

• a) responsive to receiving a form as an input image, retrieving topological information from the form to identify one or more regions in the input image;
• b) in each identified region, identifying one or more fields;
• c) in each identified field, discerning a format of one or more sub-regions;
• d) responsive to the topological information, the identified one or more regions, the identified one or more fields, and the discerned format of the one or more sub-regions, discerning a format each of the one or more fields;
• e) responsive to the discerned format of the one or more fields and of the one or more sub-regions, identifying relative locations of the one or more fields and sub-regions;
• f) responsive to the identification of the relative locations, categorizing the form, wherein a) to f) enable form identification for the document and form analysis without determining precise content of textual information;
• g) responsive to an incorrect categorization of the form, updating the machine learning system;
• and
• h) repeating a) to g) until all forms have been categorized.