System and method for identification and separation of form and feature elements from handwritten and other user supplied elements

• US 9,262,679 B2
• Filed: 07/22/2015
• Issued: 02/16/2016
• Est. Priority Date: 01/10/2012
• Status: Active Grant

• Current Assignee: DST Technologies, Inc. (SS&C Technologies Holdings Incorporated)
• Original Assignee: DST Technologies, Inc. (SS&C Technologies Holdings Incorporated)
• Inventors: Stokes, Kevin W.
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): RIVERA-MARTINEZ, GUILLERMO M
• Application Number:
  US14/805,789
• Publication Number:
  US 20150324638A1
• Time in Patent Office: 209 Days
• US Class Current: 1/1
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/153 using recognition of characters or words
• CPC Class Codes: G06V 30/155 Removing patterns interfering with the pattern to be recognised, such as ruled lines or underlines
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

A system and methods for progressive feature evaluation of an electronic document image to identify user supplied elements is disclosed. The system includes a controller in communication with a storage device configured to receive and accessibly store a generated plurality of candidate images. The controller is operable to analyze the electronic document image to identify a first feature set and a second feature set, wherein each of the first and second feature sets represent a different form feature, compare the first feature set to the second feature set, and define a third feature set based on the intersection of the first and second feature sets, wherein the third feature sets represents the user provided elements.

Claims

1. A computer implemented method for conducting progressive feature evaluation to identify a feature of interest in a document, the method comprising:

receiving, by a processor, a document image and storing the document image in a memory coupled with the processor;

identifying, by the processor, a first pixel of the stored document image having a first characteristic;

defining, by the processor, a neighborhood of pixels of the stored document image, wherein the neighborhood of pixels includes one or more additional pixels having the first characteristic and being within a threshold distance of the first pixel or any of the one or more additional pixels;

defining, by the processor, an area including the defined neighborhood of pixels;

identifying, by the processor, an aspect ratio of the defined area; and

comparing the identified aspect ratio to one or more defined aspect ratios stored in a memory coupled with the processor, wherein the one or more defined aspect ratios correspond to one or more features, and determining the feature of the one or more features corresponding to the identified aspect ratio.

2. The computer implemented method of claim 1, wherein the first characteristic comprises that the pixel is not white.

3. The computer implemented method of claim 1, wherein the threshold distance comprises two pixels.

4. The computer implemented method of claim 1, wherein defining an area comprises:

identifying, by the processor, one or more pixels having the first characteristic inside the neighborhood of pixels; and

defining, by the processor, a height and a width for the defined area which includes all of the one or more pixels having the first characteristic.

5. The computer implemented method of claim 4, wherein the defined area is at least twenty pixels wide.

6. The computer implemented method of claim 4, wherein the defined area is at least twenty pixels high.

7. The computer implemented method of claim 4, wherein the defined area has an area of at least 2000 pixels.

8. The computer implemented method of claim 1, wherein the identified aspect ratio of the defined area is derived from the outermost points of the defined area.

9. The computer implemented method of claim 1, wherein the defined neighborhood of pixels excludes pixels that are included within a vertical line of greater than 75 pixels.

10. The computer implemented method of claim 1, further comprising:

combining, by the processor, more than one neighborhood of pixels to define the area.

11. The computer implemented method of claim 1, further comprising:

identifying, by the processor, a pixel density of the pixels having the first characteristic which are included in the area, and

comparing, the processor, the pixel density to one or more defined pixel densities;

wherein the one or more defined pixel densities correspond to one or more features.

12. The computer implemented method of claim 11, wherein the pixel density is a ratio of pixels having the first characteristic to pixels not having the first characteristic within the defined area.

13. A system for conducting progressive feature evaluation to identify a feature of interest in a document, the system comprising:

a controller in communication with a storage device configured to receive and accessibly store a generated plurality of document images, wherein the controller is operable to;

receive a document image from the storage device;

identify a first pixel of the received document image having a first characteristic;

define a neighborhood of pixels of the received document image, wherein the neighborhood of pixels includes one or more additional pixels having the first characteristic and being within a threshold distance of the first pixel or any of the one or more additional pixels;

define an area including the defined neighborhood of pixels;

identify an aspect ratio of the defined area; and

compare the identified aspect ratio to one or more defined aspect ratios, wherein the one or more defined aspect ratios correspond to one or more features, and determine the feature of the one or more features corresponding to the identified aspect ratio.

14. The system of claim 13, wherein the controller is further operable to:

identify a pattern of one or more defined areas in the stored document image.

15. The system of claim 13, wherein the controller is further operable to:

identify a pixel density of the pixels included in the defined area; and

compare the pixel density to one or more defined pixel densities, wherein the one or more defined pixel densities correspond to one or more features.

16. The system of claim 15, wherein the one or more features are checkboxes.

17. The system of claim 15, wherein the one or more features are alphanumeric characters.

18. A computer implemented method for conducting progressive feature evaluation to identify a feature of interest in a document, the method comprising:

receiving, by a processor, a document image and storing the document image in a memory coupled with the processor;

identifying, by the processor, a first pixel of the stored document image having a first characteristic;

identifying, by the processor, a second pixel of the stored document image having the first characteristic;

defining, by the processor, a first neighborhood of pixels of the stored document image, wherein the first neighborhood of pixels includes one or more additional pixels having the first characteristic and being within a threshold distance of the first pixel or any of the one or more additional pixels.

defining, by the processor, a first area including the first defined neighborhood of pixels;

defining, by the processor, a second neighborhood of pixels of the stored document image, wherein the second neighborhood of pixels includes one or more second additional pixels having the first characteristic and being within a threshold distance of the second pixel or any of the one or more second additional pixels;

defining, by the processor, a second area including the second defined neighborhood of pixels;

identifying, by the processor, on the document image, a position of the first defined area and a position of the second defined area; and

determining, by the processor, a feature of the document image using the identified position of the first defined area and the identified position of the second defined area.

19. The method of claim 18, further comprising:

identifying a pixel density of the pixels included in the first and second defined areas; and

comparing the pixel density to one or more defined pixel densities, wherein the one or more defined pixel densities correspond to one or more features.

20. The method of claim 18, wherein identifying, by the processor, a feature of the document image comprises:

identifying a proximity of the first defined area to the second defined area based on the identified positions of the first and second areas.