Method and apparatus for character string recognition

• US 8,200,016 B2
• Filed: 04/28/2008
• Issued: 06/12/2012
• Est. Priority Date: 04/27/2007
• Status: Expired due to Fees

• Current Assignee: Nidec Sankyo Corporation (Nidec Corporation)
• Original Assignee: Nidec Sankyo Corporation (Nidec Corporation)
• Inventors: Nakamura, Hiroshi
• Primary Examiner(s): Liu, Li
• Application Number:
  US12/110,524
• Publication Number:
  US 20080304746A1
• Time in Patent Office: 1,506 Days
• Field of Search: None
• US Class Current: 382/177
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/158 using character size, text spacings or pitch estimation

Abstract

A method for character string recognition may include processing image data into black-and-white binary image data, calculating vertical projection data of the binary image data in a vertical direction perpendicular to a direction of the character string while shifting the binary image data, detecting positions exceeding a prescribed border judgment threshold value in the vertical projection data, judging validity of the border judgment threshold value, and deciding whether to segment characters out of the character string based on whether the border judgment threshold value is valid.

Claims

1. A method for character string recognition by processing image data obtained through picking up a character string placed on a medium comprising:

processing the image data into black-and-white binary image data by using a prescribed reference threshold value for binary conversion;

calculating vertical projection data of the binary image data in a vertical direction perpendicular to a direction of the character string while shifting the binary image data in the direction of the character string;

detecting positions exceeding a prescribed border judgment threshold value in the vertical projection data as segmentation positions of characters included in the character string;

judging validity of the border judgment threshold value according to the number of characters included in the character string;

if the border judgment threshold value is judged to be valid, segmenting the characters out of the character string according to the segmentation positions of the characters, and examining characteristics of each character segmented; and

if the border judgment threshold value is judged to be invalid, repeating the method from the step of processing the image data into black-and-white binary image data without segmenting the characters out of the character string.

2. The method for character string recognition according to claim 1:

wherein the reference threshold value for binary conversion is changed when the border judgment threshold value is judged to be invalid.

3. The method for character string recognition according to claim 2:

wherein the reference threshold value for binary conversion is changed to be smaller than the initial one when the border judgment threshold value is judged to be invalid.

4. The method for character string recognition according to claim 3:

wherein the prescribed border judgment threshold value corresponds to the “

n+1”

-th value of the peak values where the positive peak values included in the vertical projection data are disposed in due order starting from the largest one and the “

n”

represents the number of characters included in the character string.

5. The method for character string recognition according to claim 2:

wherein the prescribed border judgment threshold value corresponds to the “

n+1”

-th value of the peak values where the positive peak values included in the vertical projection data are disposed in due order starting from the largest one and the “

n”

represents the number of characters included in the character string.

6. The method for character string recognition according to claim 1:

wherein the prescribed border judgment threshold value corresponds to the “

n+1”

-th value of the peak values where the positive peak values included in the vertical projection data are disposed in due order starting from the largest one and the “

n”

represents the number of characters included in the character string.

7. The method for character string recognition according to claim 1:

wherein the border judgment threshold value is a positive peak corresponding to the “

n+1”

-th number of the character string that make up the number of positive peaks contained in the vertical projection data, where “

n”

represents the number of characters included in the character string.

8. An apparatus for character string recognition for use with a record data carrier having a character string, the apparatus comprising:

an image pickup device configured to read and photoelectrically convert the character string into image data;

an image memory configured to store the image data;

a binarizinq section configured to process the image data into black-and-white binary image data by using a prescribed reference threshold value for binary conversion;

a projection creating section configured to calculate vertical projection data of the binary image data in a vertical direction perpendicular to a direction of the character string while shifting the binary image data in the direction of the character string;

a threshold value calculating section configured to detect vertical projection data of the binary image data in a vertical direction perpendicular to a direction of the character string while shifting the binary image data in the direction of the character string;

a threshold value evaluating section configured to judge validity of the border judgment threshold value according to the number of characters included in the character string; and

a border defining section configured to segment the characters out of the character string according to the segmentation positions of the characters if the border judgment threshold value is judged to be valid;

wherein, if the border judgment threshold value is judged to be invalid, the apparatus is configured such that the binarizing section repeats processing the image data into black-and-white binary image data by using a prescribed reference threshold value for binary conversion, without segmenting the characters out of the character string.

9. The apparatus according to claim 8,

wherein the reference threshold value for binary conversion is changed when the border judgment threshold value is judged to be invalid.

10. The apparatus according to claim 9,

wherein the reference threshold value for binary conversion is changed to be smaller than the initial one when the border judgment threshold value is judged to be invalid.

11. The apparatus according to claim 8,

wherein the prescribed border judgment threshold value corresponds to the “

n+1”

-th value of the peak values where the positive peak values included in the vertical projection data are disposed in due order starting from the largest one and the “

n”

represents the number of characters included in the character string.