Systems and methods for separating ligature characters in digitized document images

Patent metadata

Field	Value
Publication number	US-10878271-B2
Application number	US-201916357878-A
Country	US
Kind code	B2
Filing date	Mar 19, 2019
Priority date	Mar 19, 2019
Publication date	Dec 29, 2020
Grant date	Dec 29, 2020

Abstract

Embodiments disclosed herein provide for systems and methods of separating characters associated with ligatures in digitized documents. The systems and methods provide for a ligature detection engine configured to identify the ligatures, and a ligature processing engine configured to identify and remove the glyphs attaching the separate characters forming the ligature.

First claim

The invention claimed is:

1. A system for separating characters associated with ligatures in digitized document images, the system comprising one or more processors configured to:
receive at least one digitized document image including a plurality of characters;
determine which of the plurality of characters in the at least one digitized document image are associated with ligatures;
generate a contour around each of the ligatures, wherein the contour includes a pixelated version of the ligature, wherein pixels associated with glyphs of the ligature are darkened;
scan each column of the contour;
determine, for each scanned column including at least one darkened pixel, a height of a respective glyph associated with the at least one darkened pixel by scanning from top and bottom of the column of the contour, wherein the scanning determines an imaginary vertical line so as to separate one or more characters from the plurality of characters, the imaginary vertical line based on a transition from a first point belonging to one or more darkened pixels of the column to a second point belonging to one or more darkened pixels of a different column, the transition defining a change from a first slope to a second slope between the column and different column;
identify a pinch point for the ligature based on the imaginary vertical line and a comparison between a plurality of adjacent scanned columns including at least one darkened pixel;
remove the glyph associated with the pinch point; and
separate the one or more characters associated with the ligature based on the removed glyph and the imaginary vertical line.

2. The system of claim 1 , wherein the one or more processors are further configured to:
(i) receive the separated characters and (ii) verify the accuracy of the separated characters.

3. The system of claim 1 , wherein the columns are scanned from at least one selected from the group of: (i) from left to right and (ii) right to left.

4. The system of claim 1 , wherein the contour includes (i) a height of the ligature and (ii) a width of the ligature.

5. The system of claim 1 , wherein the height of the respective glyph is determined based on (i) a first distance from the top of the contour to a topmost darkened pixel in the column and (ii) a second distance from the bottom of the contour to a bottommost darkened pixel in the column.

6. The system of claim 5 , wherein (i) the first distance is determined based on a first scan from the top of the contour to the topmost darkened pixel in the column and (ii) the second distance is determined based on a second scan from the bottom of the contour to the bottommost darkened pixel in the column.

7. The system of claim 1 , wherein the pinch point is identified upon determining (i) a decrease in height of the respective glyph from a first column to a second column and (ii) an increase in height of the respective glyph from a second column to a third column.

8. The system of claim 7 , wherein the one or more processors are further configured to segment the contour vertically at the pinch point and separate the contour into separate contours.

9. The system of claim 1 , wherein the one or more processors are further configured to store a height of a glyph in a previously scanned column upon determining a change in height of the glyph from the previously scanned column to another glyph in a currently scanned column.

10. The system of claim 1 , wherein the one or more processors are further configured to convert the at least one digitized document image into monochrome.

11. A method for separating characters associated with ligatures in digitized document images, the method comprising:
receiving, with a processor, at least one digitized document image including a plurality of characters;
determining, with the processor, which of the plurality of characters in the at least one digitized document image are associated with ligatures;
generating, with the processor, a contour around each of the ligatures, wherein the contour includes a pixelated version of the ligature, wherein pixels associated with glyphs of the ligature are darkened; and
scanning, with the processor, each column of the contour;
determining, with the processor, for each scanned column including at least one darkened pixel, a height of a respective glyph associated with the at least one darkened pixel by scanning from top and bottom of the column of the contour, wherein the scanning determines an imaginary vertical line so as to separate one or more characters from the plurality of characters, the imaginary vertical line based on a transition from a first point belonging to one or more darkened pixels of the column to a second point belonging to one or more darkened pixels of a different column, the transition defining a change from a first slope to a second slope between the column and different column;
identifying, with the processor, a pinch point for the ligature based on the imaginary vertical line and a comparison between a plurality of adjacent scanned columns including at least one darkened pixel;
removing, with the processor, the glyph associated with the pinch point; and
separating, with the processor, the one or more characters associated with the ligature based on the removed glyph and the imaginary vertical line.

12. The method of claim 11 , wherein the columns are scanned from at least one selected from the group of: (i) from left to right and (ii) right to left.

13. The method of claim 11 , wherein the contour includes (i) a height of the ligature and (ii) a width of the ligature.

14. The method of claim 11 , wherein the height of the respective glyph is determined based on (i) a first distance from the top of the contour to a topmost darkened pixel in the column and (ii) a second distance from the bottom of the contour to a bottommost darkened pixel in the column.

15. The method of claim 14 , wherein (i) the first distance is determined based on a first scan from the top of the contour to the topmost darkened pixel in the column and (ii) the second distance is determined based on a second scan from the bottom of the contour to the bottommost darkened pixel in the column.

16. The method of claim 11 , wherein the pinch point is identified upon determining (i) a decrease in height of the respective glyph from a first column to a second column and (ii) an increase in height of the respective glyph from a second column to a third column.

17. The method of claim 16 , further comprising:
segmenting, with the processor, the contour vertically at the pinch point and separating the contour into separate contours.

18. The method of claim 11 , further comprising:
storing, with the processor, a height of a glyph associated with a previously scanned column in a memory database upon determining a change in height of the glyph from the previously scanned column to another glyph in a currently scanned column.

19. A system for separating characters associated with ligatures in digitized document images, the system comprising:
a processor, wherein the processor is configured to:
receive a contour including a ligature, wherein the ligature is pixelated, wherein pixels associated with the ligature are darkened;
scan each column of the contour;
determine, for each scanned column including at least one darkened pixel, a height of a respective glyph associated with the at least one darkened pixel by scanning from top and bottom of the column of the contour, wherein the scanning determines an imaginary vertical line so as to separa

Assignees

• Capital One Services Llc

Inventors

• Slattery Douglas

Classifications

• G06V30/155

  Removing patterns interfering with the pattern to be recognised, such as ruled lines or underlines · CPC title

• G06V30/414 Primary

  Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text · CPC title

• G06V30/158

  using character size, text spacings or pitch estimation · CPC title

• G06V30/153

  using recognition of characters or words · CPC title

• G06V30/10

  Character recognition · CPC title