US 6,400,845 B1Granted

System and method for data extraction from digital images

AssigneeComputer Services Corporation

InventorGary Volino

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system and method of the extraction of textual data from a digital image using a data pattern comprised of visible and invisible characters to locate the data to be extracted and upon find such data populating the fields of an associated data base with the extracted visible data. The digital image to be processed is first compared against master document images contained in a database. Upon determining the proper master document image, a template having predefined data zone is applied to the image to create zone images. The zone images are optically read and converted into a character file which is then parsed with the pattern to locate the text to be extracted. Upon finding data matching the pattern, that data is extracted and the visible portions used to populate data fields in a database record associated with the digital image. In an alternate embodiment, if the extracted data cannot be successfully matched, a validation file of the unmatched data is created for review by an operator. In a further embodiment, if the scanned digital image cannot be matched with an existing master document image, a new master document image can be created from the unmatched digital image. In another alternate embodiment, alternate patterns can be used to search the data files allowing for variation in format of the data being extracted.

Claims13

Dependent claims

A system for the extraction of textual data from a digital image using predefined patterns based on visible and invisible characters contained in the textual data, comprising:

database means for storing data base records comprising:
a master document image database comprised of at least one table containing at least one master document image;
a template database comprised of at least one table comprising at least one template associated with the master document image, the template having at least one zone, the zone associated with a unique pattern comprised of one or more nonoverlapping segments, each segment containing one or more characters, with selected ones of the segments being associated with a data field in an extracted data base record;
an extracted data database comprised of at least one table of extracted data base records, each record comprised of at least one data field for storing textual information extracted from the digital image;
an image comparator in communication with the database means and receiving therefrom the master document image, the image comparator having an input for receiving the digital image, the image comparator comparing the master document image to the digital image and providing an output indicative of the success of the comparison;
a template mapper in communication with the database means and the output of the image comparator and having an input for receiving the digital image and, on receiving the image comparator output indicating a successful comparison, retrieving the template from the template database associated with the successfully compared master document image and applying the template to the digital image, the template mapper providing as an output an image of each zone associated with the applied template;
a zone optical character reader (OCR) in communication with the template mapper and receiving the output thereof, the zone OCR creating a zone data file of the characters in each zone image and providing the zone data file as an output;
a zone pattern comparator in communication with the database means and the output of the zone OCR, the zone pattern comparator retrieving from template database the pattern associated with the zone and comparing the pattern to the zone data file, and, in the event that the pattern is found, extracting the data matching the pattern digital into an extracted data file, the zone pattern comparator providing the extracted data file as an output;
and
an extracted data parser in communication with the database means and the output of the zone pattern comparator, the parser parsing the data in the extracted data file for populating the data field of the database record associated with the pattern, the parser providing as an output the populated database record to the extracted data database for storage therein.

7 more elements

A method for the extraction of textual data from a digital image containing character images using predefined patterns based on visible and invisible characters contained in the textual data, comprising:

a) selecting from a database a master document image having associated therewith a template, the template having a zone, the zone having associated therewith one pattern comprised of at least one data segment each containing a data sequence of one or more characters;
b) creating an unpopulated database table having one or more data records, each data record having one or more data fields for containing visible character data extracted from the digital image and associating the database table with the master document image and the database record with the digital image, and, for at least one of the data segments containing visible data associating it with a database field in the database;
c) comparing the digital image to the master document image and upon an successful match occurring:
applying the associated template and the zone therein to the digital image, performing optical character recognition on the character images within the zone, creating a zone data file containing the characters optically read from the zone;
comparing the zone data file with the pattern associated with the zone;
extracting the data in the zone data file that matches the pattern, and, for each data segment associated with a data field, populating the data field with the visible data extracted from the zone data file corresponding to that data segment.

3 more elements

A method for the extraction of textual data from a digital image using predefined patterns based on visible and invisible characters contained in the textual data, comprising:

a) creating at least one master document image having associated therewith at least one template, the template having at least one zone, each zone having associated with it one predefined pattern comprised of one or more data segments containing a data sequence of one or more characters;
b) creating an unpopulated database table having one or more data records, each data record having one or more data fields for containing visible character data extracted from the digital image and associating the database table with the master document image and the database record with the digital image, and, for at least one of the data segments containing visible data associating it with a database field;
c) storing the database record, master document image and associated template, zone and pattern in a database;
d) comparing the digital image to a master document image retrieved from the database;
e) upon an unsuccessful match:
1) selecting a new master document image from the database for comparison until a match is found;
and
2) in the event no match is found, creating a file for the unmatched image and alerting the operator that no match has been found;
f) upon an successful match:
1) applying the associated template and each zone therein to the digital image,
2) performing optical character recognition on the characters images within each zone,
3) creating, for each zone, a zone data file containing the characters read from the zone;
4) selecting a zone data file;
5) comparing the selected zone data file with the pattern associated with the selected zone;
6) in the event no data matching the pattern is found, creating a validation file containing the zone data file for operator review;
and
7) in the event data matching the pattern is found, extracting the data in the zone data file that matches the pattern, and, for each data segment associated with a data field, populating the data field with the visible data extracted from the zone data file corresponding to that data segment.
g) selecting, if additional zones are present, the next zone and repeating steps d-f;
and
h) selecting, if additional digital images are present, the next digital image to be processed and repeating steps d-g.

17 more elements

Dependencies are inferred from the claim text.