Methods and Systems for Detecting Numerals in a Digital Image

• US 20090290751A1
• Filed: 05/23/2008
• Published: 11/26/2009
• Est. Priority Date: 05/23/2008
• Status: Active Grant

• Current Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Campbell, Richard John, Ferman, Ahmet Mufit
• Granted Patent: US 8,023,741 B2
• Time in Patent Office: Days
• US Class Current: 382/100
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: G06V 30/10 Character recognition

Abstract

Aspects of the present invention are related to systems and methods for determining the location of numerals in an electronic document image.

Claims

1. A method for detecting a numeral connected component in a digital image, said method comprising:

a) receiving a text-line component, wherein said text-line component comprises a plurality of connected components in a digital image;

b) determining a component bounding box for each of said plurality of connected components, wherein each component bounding box comprises a first-side coordinate, a second-side coordinate, a third-side coordinate and a fourth-side coordinate, wherein said first-side coordinate and said second-side coordinate are associated with a first axis of said bounding box and said third-side coordinate and said fourth-side coordinate are associated with a second axis of said bounding box;

c) determining a first variability measure associated with said first-side coordinates;

d) determining a second variability measure associated with said second-side coordinates;

e) determining a third variability measure associated with said third-side coordinates;

f) determining a fourth variability measure associated with said fourth-side coordinates;

g) determining a first accumulation of said first variability measure and said second variability measure;

h) determining a second accumulation of said third variability measure and said fourth variability measure; and

i) when said first accumulation and said second accumulation meet a first criterion;

i) classifying said text-line component as a numeral component when said first variability measure meets a first threshold criterion and said second variability measure meets a second threshold criterion; and

ii) classifying said text-line component as a non-numeral component when either said first variability measure does not meet said first threshold criterion or said second variability measure does not meet said second threshold criterion; and

j) when said first accumulation and said second accumulation do not meet said first criterion;

i) classifying said text-line component as a numeral component when said third variability measure meets a third threshold criterion and said fourth variability measure meets a fourth threshold criterion; and

ii) classifying said text-line component as a non-numeral component when either said third variability measure does not meet said third threshold criterion or said fourth variability measure does not meet said fourth threshold criterion.

2. A method as described in claim 1, wherein:

a) said determining a first variability measure associated with said first-side coordinates comprises calculating a first representative value of said first-side coordinates;

b) said determining a second variability measure associated with said second-side coordinates comprises calculating a second representative of said second-side coordinates;

c) said determining a third variability measure associated with said third-side coordinates comprises calculating a third representative value of said third-side coordinates; and

d) said determining a fourth variability measure associated with said fourth-side coordinates comprises calculating a fourth representative of said fourth-side coordinates.

3. A method as described in claim 2, wherein:

a) said first representative value is a sample mean of said first-side components;

b) said second representative value is a sample mean of said second-side components;

c) said third representative value is a sample mean of said third-side components; and

d) said fourth representative value is a sample mean of said fourth-side components.

4. A method as described in claim 2, wherein:

a) said determining a first variability measure comprises calculating a first error measure between said first representative value and said first-side components;

b) said determining a second variability measure comprises calculating a second error measure between said second representative value and said second-side components;

c) said determining a third variability measure comprises calculating a third error measure between said third representative value and said third-side components; and

d) said determining a fourth variability measure comprises calculating a fourth error measure between said fourth representative value and said fourth-side components.

5. A method as described in claim 4, wherein:

a) said first error measure is an error measure selected from the group consisting of mean absolute error, mean-square error and root mean-square error;

b) said second error measure is an error measure selected from the group consisting of mean absolute error, mean-square error and root mean-square error;

c) said third error measure is an error measure selected from the group consisting of mean absolute error, mean-square error and root mean-square error; and

d) said fourth error measure is an error measure selected from the group consisting of mean absolute error, mean-square error and root mean-square error.

6. A method as described in claim 1 further comprising:

a) when said text-line component is classified as a numeral component;

i) calculating the aspect ratio of each connected component in said plurality of connected components, thereby producing a plurality of aspect ratios;

ii) calculating the mean of said plurality of aspect ratios;

iii) calculating the variance of said plurality of aspect ratios; and

iv) reclassifying said text-line component based on said mean and said variance.

7. A method as described in claim 6, wherein said reclassifying said text-line component comprises:

a) reclassifying said text-line component as numeral when said mean meets a first mean criterion and said variance meets a first variance criterion; and

b) reclassifying said text-line component as non-numeral when either said mean does not meet said first mean criterion or said variance does not meet said first variance criterion.

8. A method as described in claim 7 further comprising, when said text-line component is reclassified as a numeral component, reclassifying said text-line component based on the number of connected components in said plurality of connected components.

9. A method as described in claim 1 further comprising, when said text-line component is classified as a numeral component, reclassifying said text-line component based on the number of connected components in said plurality of connected components.

10. A method as described in claim 9, wherein said reclassifying said text-line component comprises:

a) reclassifying said text-line component as a numeral component when said number of connected components meets a first quantity criterion; and

b) reclassifying said text-line component as a non-numeral component when said number of connect components does not meet said first quantity criterion.

11. A method for detecting a numeral connected component in a digital image, said method comprising:

a) receiving a text-line component, wherein said text-line component comprises a plurality of connected components in a digital image;

b) calculating an aspect ratio for each of said connected components in said plurality of connected components, thereby producing a plurality of aspect ratios;

c) calculating a first characteristic of said plurality of aspect ratios;

d) classifying said text-line component as a numeral component when said first characteristic meets a first criterion; and

e) classifying said text-line component as a non-numeral component when said first characteristic does not meet said first criterion.

12. A method as described in claim 11, wherein said first characteristic is an average of said plurality of aspect ratios.

13. A method as described in claim 11, wherein said first characteristic is a standard deviation of said plurality of aspect ratios.

14. A method as described in claim 11 further comprising:

a) calculating a second characteristic of said plurality of aspect ratios; and

b) wherein, said first criterion comprises said first characteristic meeting a second criterion and said second characteristic meeting a third criterion.

15. A method as described in claim 14, wherein said second characteristic is selected from the group consisting of mean and variance.

16. A method as described in claim 11, further comprising:

a) determining the number of connected components in said plurality of connected components;

b) calculating a second characteristic of said plurality of aspect ratios; and

c) wherein, said first criterion comprises said first characteristic meeting a second criterion and said second characteristic meeting a third criterion and said number of connect components meeting a quantity criterion.

17. A method as described in claim 16, further comprising:

a) determining a component bounding box for each of said plurality of connected components, wherein each component bounding box comprises a first-side coordinate, a second-side coordinate, a third-side coordinate and a fourth-side coordinate, wherein said first-side coordinate and said second-side coordinate are associated with a first axis of said bounding box and said third-side coordinate and said fourth-side coordinate are associated with a second axis of said bounding box;

b) determining a first variability measure associated with said first-side coordinates;

c) determining a second variability measure associated with said second-side coordinates;

d) determining a third variability measure associated with said third-side coordinates;

e) determining a fourth variability measure associated with said fourth-side coordinates;

f) determining a first accumulation of said first variability measure and said second variability measure;

g) determining a second accumulation of said third variability measure and said fourth variability measure; and

h) when said first accumulation and said second accumulation meet a first accumulation criterion;

i) setting a first variability characteristic equal to said first variability measure; and

ii) setting a second variability characteristic equal to said second variability measure;

i) when said first accumulation and said second accumulation do meet said first accumulation criterion;

i) setting said first variability characteristic equal to said third variability measure; and

ii) setting said second variability characteristic equal to said fourth variability measure; and

j) wherein, said first criterion comprises said first characteristic meeting said second criterion and said second characteristic meeting said third criterion and said number of connect components meeting said quantity criterion and said first variability characteristic meeting a fourth criterion and said second variability characteristic meeting a fifth criterion.

18. A method as described in claim 17, wherein:

a) said determining a first variability measure comprises;

i) calculating a first representative value of said first-side components; and

ii) calculating a first error measure between said first representative value and said first-side components;

b) said determining a second variability measure comprises;

i) calculating a second representative value of said second-side components; and

ii) calculating a second error measure between said second representative value and said second-side components;

c) said determining a third variability measure comprises;

i) calculating a third representative value of said third-side components; and

ii) calculating a third error measure between said third representative value and said third-side components; and

d) said determining a fourth variability measure comprises;

i) calculating a fourth representative value of said fourth-side components; and

ii) calculating a fourth error measure between said fourth representative value and said fourth-side components.

19. A method as described in claim 11, further comprising:

a) determining a component bounding box for each of said plurality of connected components, wherein each component bounding box comprises a first-side coordinate, a second-side coordinate, a third-side coordinate and a fourth-side coordinate, wherein said first-side coordinate and said second-side coordinate are associated with a first axis of said bounding box and said third-side coordinate and said fourth-side coordinate are associated with a second axis of said bounding box;

b) determining a first variability measure associated with said first-side coordinates;

c) determining a second variability measure associated with said second-side coordinates;

d) determining a third variability measure associated with said third-side coordinates;

e) determining a fourth variability measure associated with said fourth-side coordinates;

f) determining a first accumulation of said first variability measure and said second variability measure;

g) determining a second accumulation of said third variability measure and said fourth variability measure; and

h) when said first accumulation and said second accumulation meet a first accumulation criterion;

i) setting a first variability characteristic equal to said first variability measure; and

ii) setting a second variability characteristic equal to said second variability measure;

i) when said first accumulation and said second accumulation do meet said first accumulation criterion;

i) setting said first variability characteristic equal to said third variability measure; and

ii) setting said second variability characteristic equal to said fourth variability measure; and

j) wherein, said first criterion comprises said first characteristic meeting a second criterion and said second characteristic meeting a third criterion and said first variability characteristic meeting a fourth criterion and said second variability characteristic meeting a fifth criterion.

20. A method as described in claim 19, wherein:

a) said determining a first variability measure comprises;

i) calculating a first representative value of said first-side components; and

ii) calculating a first error measure between said first representative value and said first-side components;

b) said determining a second variability measure comprises;

i) calculating a second representative value of said second-side components; and

ii) calculating a second error measure between said second representative value and said second-side components;

c) said determining a third variability measure comprises;

i) calculating a third representative value of said third-side components; and

ii) calculating a third error measure between said third representative value and said third-side components; and

d) said determining a fourth variability measure comprises;

i) calculating a fourth representative value of said fourth-side components; and

ii) calculating a fourth error measure between said fourth representative value and said fourth-side components.

21. A method for detecting a numeral connected component in a digital image, said method comprising:

a) receiving a text-line component, wherein said text-line component comprises a plurality of connected components in a digital image;

b) determining the number of connected components in said plurality of connected components;

c) classifying said text-line component as a numeral component when said number of connected components meets a quantity criterion; and

d) classifying said text-line component as a non-numeral component when said number of connected components does not meet said quantity criterion.