Systems and methods for identification document processing and business workflow integration

• US 9,058,580 B1
• Filed: 03/19/2014
• Issued: 06/16/2015
• Est. Priority Date: 01/12/2012
• Status: Active Grant

• Current Assignee: Tungsten Automation Corp.
• Original Assignee: Kofax Incorporated
• Inventors: Amtrup, Jan W., Thompson, Stephen Michael, Kilby, Steven, Macciola, Anthony
• Primary Examiner(s): Mariam, Daniel
• Application Number:
  US14/220,029
• Publication Number:
  US 20150170085A1
• Time in Patent Office: 454 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/06316 Sequencing of tasks or work
• CPC Class Codes: H04N 1/387 Composing, repositioning or otherwise geometrically modifying originals
• CPC Class Codes: H04N 1/3872 Repositioning or masking
• CPC Class Codes: H04N 1/3878 Skew detection or correction

Abstract

A method includes receiving or capturing an image comprising an identity document (ID) using a mobile device; classifying the ID; building an extraction model based on the ID classification; extracting data from the ID based on the extraction model; building an ID profile based on the extracted data; storing the ID profile to a memory of the mobile device; detecting a predetermined stimulus in a workflow; identifying workflow-relevant data in the stored ID profile at least partially in response to detecting the predetermined stimulus; providing the workflow-relevant data from the stored ID profile to the workflow; and driving at least a portion of the workflow using the workflow-relevant data. Related systems and computer program products are also disclosed.

Claims

1. A method, comprising:

receiving or capturing an image comprising an identity document (ID) using a mobile device;

classifying the ID;

wherein the classifying comprises;

generating a reduced-resolution representation of the ID,

determining the ID corresponds to a particular state; and

identifying the particular state to which the ID corresponds; and

extracting data from the ID based at least in part on the ID classification; and

driving at least a portion of a workflow based on the extracted data,

wherein the classifying is based on the image characteristics corresponding to the reduced-resolution representation of the identity document.

2. The method as recited in claim 1, further comprising building an extraction model based on the ID classification;

and

extracting the data from the ID using the extraction model.

3. The method as recited in claim 1, further comprising determining the extracted data are relevant to the workflow, wherein the determining is based at least in part on one or more of:

the ID classification;

determining the portion of the extracted data comprises one or more predetermined strings; and

determining the portion of the extracted data comprises one or more predetermined images.

4. The method as recited in claim 1,

wherein the reduced resolution representation of the identity document is a grayscale image.

5. The method as recited in claim 1, wherein the image is a single image comprising multiple documents, the method further comprising:

classifying at least two of the multiple documents;

building an extraction model based on each document classification; and

extracting the data from at least some of the classified documents based on the extraction model.

6. The method as recited in claim 1, wherein the ID comprises a multi-paged document, and the method further comprising:

receiving or capturing one or more additional images using the mobile device, each additional image comprising at least a portion of at least one page of the ID,

classifying each page of the ID;

comparing each page classification to a classification of one or more other pages; and

classifying the ID based on the comparison

wherein the data are extracted from at least two pages of the ID.

7. The method as recited in claim 1, wherein the workflow comprises activities relating to a business process selected from:

a tracking workflow;

applying for employment;

applying for an entitlement;

a financial reporting workflow;

a scheduling workflow

wherein one or more activities comprising the workflow rely on some or all of the extracted data.

8. The method as recited in claim 1, further comprising synthesizing a plurality of identity sub-profiles based on at least some of the extracted data,

wherein each identity sub-profile comprises identifying information pertaining to one or more of a predetermined type of business process and a unique entity.

9. The method as recited in claim 1, the classifying further comprising:

building a feature vector corresponding to the ID based on the image characteristics;

comparing the feature vector to at least two reference feature matrices, each reference feature matrix corresponding to a unique class of identity document.

10. The method as recited in claim 1, the classifying further comprising defining a hyper-plane in an at least N-dimensional feature space, wherein N is greater than or equal to a number of features represented in a feature vector corresponding to the ID.

11. The method as recited in claim 1, further comprising locating workflow-relevant data using the extracted data, wherein the workflow-relevant data are not depicted on the ID.

12. The method as recited in claim 1, wherein the classifying further comprises:

determining whether the document is an ID; and

in response to determining the document is an ID, determining one or more ID characteristics selected from;

an ID class;

an ID subclass

an issuing authority corresponding to the ID; and

a jurisdiction corresponding to the ID.

14. The method as recited in claim 1, further comprising:

determining the financial transaction cannot be completed utilizing the extracted data;

determining one or more alternative data sources suitable for use in connection with the financial transaction;

designating one of the alternative data sources for use in connection with the financial transaction; and

completing the financial transaction based at least in part on data obtained from one or more of the designated alternative data sources, and

wherein the workflow comprises a financial transaction reliant at least in part on either the extracted data or data determined using the extracted data.

15. The method as recited in claim 14, further comprising:

displaying an indication to a user that the financial transaction cannot be completed utilizing the extracted data;

requesting user input designating one or more of the alternative data sources;

receiving the user input; and

completing the financial transaction using the designated alternative data source based at least in part on the user input.

16. The method as recited in claim 1, wherein the classification further comprises a weighted feature vector comparison.

17. The method as recited in claim 1, further comprising:

receiving user input relating to the workflow,

wherein the user input is received at least in part via a mobile device display, and

wherein at least one of the extracting and the driving are further based at least in part on the received user input.

18. The method as recited in claim 1, further comprising validating the ID prior to extracting the data.

13. The method as recited 12, further comprising:

requesting user input relating to financial transaction;

receiving the user input; and

wherein the financial transaction relies at least in part on the received user input,

wherein the user input is received via a mobile device display, and wherein the user input comprises identifying information.

19. A method, comprising

receiving or capturing an image comprising an identity document (ID) using a mobile device;

classifying the ID, the classifying further comprising;

generating a reduced-resolution representation of the ID, wherein the classifying is based on the reduced-resolution representation of the ID, and wherein the reduced resolution representation of the identity document is a grayscale image; and

defining a hyper-plane in an at least N-dimensional feature space, wherein N is greater than or equal to a number of features represented in a feature vector corresponding to the ID; and

building an extraction model based on the ID classification;

extracting data from the ID based on the extraction model;

building an ID profile based on the extracted data;

storing the ID profile to a memory of the mobile device;

detecting a predetermined stimulus in a workflow;

identifying workflow-relevant data in the stored ID profile at least partially in response to detecting the predetermined stimulus;

providing the workflow-relevant data from the stored ID profile to the workflow; and

driving at least a portion of the workflow using the workflow-relevant data.