Character segmentation device and character segmentation method

• US 9,158,986 B2
• Filed: 02/06/2014
• Issued: 10/13/2015
• Est. Priority Date: 02/06/2013
• Status: Active Grant

• Current Assignee: Nidec Sankyo Corporation (Nidec Corporation)
• Original Assignee: Nidec Sankyo Corporation (Nidec Corporation)
• Inventors: Nakamura, Hiroshi
• Primary Examiner(s): Allison, Andrae S
• Application Number:
  US14/174,424
• Publication Number:
  US 20140219561A1
• Time in Patent Office: 614 Days
• Field of Search: None
• 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 30/18095 Summing image-intensity values; Projection and histogram analysis
• CPC Class Codes: G06V 30/40 Document-oriented image-based pattern recognition

Abstract

A character segmentation section, for segmenting characters of a character line may include a minimum pixel-value curve creating section configured to extract a smallest pixel value in pixels composing a pixel line arranged in a direction orthogonal to a character line direction in said multi-level image data and create a minimum pixel-value curve, a character partitioning position determining section configured to determine partitioning positions of said characters, based on said minimum pixel value curve, a binarization processing section configured to detect a minimum pixel value indicating said linear drawing from said minimum pixel-value curve, acquires a binarization threshold based on said minimum pixel value, and binarizes said multi-level image data using said binarization threshold, and a character segmentation implementing section configured to extract the image data of each character.

Claims

1. A character segmentation section, for segmenting characters of a character line, for use in a character segmentation device which processes multi-level image data acquired by imaging said character line, in which a linear drawing is present, on an information recording medium said character segmentation section comprising:

a minimum pixel-value curve creating section configured to extract a smallest pixel value in pixels composing a pixel line arranged in a direction orthogonal to a character line direction in said multi-level image data and create a minimum pixel-value curve,

a character partitioning position determining section configured to determine partitioning positions of said characters, based on said minimum pixel value curve,

a binarization processing section configured to detect a minimum pixel value indicating said linear drawing from said minimum pixel-value curve, acquires a binarization threshold based on said minimum pixel value, and binarizes said multi-level image data using said binarization threshold, and

a character segmentation implementing section configured to extract the image data of each character.

2. The character segmentation device as set forth in claim 1, wherein said binarization processing section detects a minimum pixel value in said minimum pixel-value curve for a blank space provided between characters and binarizes said multi-level image data using said minimum pixel value as a binarization threshold.

3. The character segmentation device as set forth in claim 1, wherein said character partitioning position determining section calculates a temporary binarization threshold, binarizes said multi-level image data, and determines the boundary between a character space in which a character is printed and an adjacent blank space provided between characters, based on said binary image data.

4. A character segmentation method for processing multi-level image data acquired by imaging a character line, in which a linear drawing is present, on an information recording medium to segment characters composing said character line, the character segmentation method comprising:

extracting the smallest pixel value in pixels composing a pixel line arranged in the direction orthogonal to the character line direction in said multi-level image data and creating a minimum pixel-value curve;

determining partitioning positions of said characters based on said minimum pixel value curve;

detecting a minimum pixel value indicating said linear drawing from said minimum pixel curve, acquiring a binarization threshold based on said minimum pixel value and binarizing said multi-level image data using said binarization threshold; and

extracting the image data of each character.

5. The character segmentation method as set forth in claim 4, wherein in said detecting a minimum pixel value, a minimum pixel value in said minimum pixel-value curve for a blank space provided between characters is detected, and said multi-level image data is binarized by using said minimum pixel value as binarization threshold.

6. The character segmentation method as set forth in claim 4, wherein in said determining partitioning positions of said characters, a temporary binarization threshold is calculated and said multi-level image data is binarized using said temporary binarization threshold, and a boundary between a character space in which a character is printed and an adjacent blank space provided between characters is determined, based on said binary image data.