Ruled line extracting apparatus for extracting ruled line from normal document image and method thereof

• US 6,917,706 B2
• Filed: 01/08/2001
• Issued: 07/12/2005
• Est. Priority Date: 12/20/1996
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Katsuyama, Yutaka
• Primary Examiner(s): Johnson, Timothy M.
• Application Number:
  US09/755,030
• Publication Number:
  US 20010002215A1
• Time in Patent Office: 1,646 Days
• Field of Search: 382/165, 382/170, 382/171, 382/173, 382/176, 382/175, 382/177, 382/199, 382/202, 382/224, 382/282, 382/284, 382/286, 382/290, 358/453, 358/462, 358/464, 358/538, 715/509, 715/510
• US Class Current: 382/173
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/155 Removing patterns interfering with the pattern to be recognised, such as ruled lines or underlines

Abstract

A ruled line extracting apparatus obtains circumscribed rectangles of pixel concatenation regions included in an input pattern, and calculates the most frequent value of their heights. Additionally, the apparatus integrates segments by ignoring a wild card segment, and calculates the most frequent value of height/width of extracted straight lines and segments structuring the straight line. Next, it performs a process for integrating/deleting straight lines using each threshold value based on the highest frequency value. Then, it checks/deletes a straight line according to a distribution of black pixels around the straight line, and recognizes the remaining straight lines as ruled line candidates.

Claims

1. A ruled line extracting apparatus, comprising:

segment extracting means for extracting information of one or more segment patterns from an input image;

storing means for classifying the information of the one or more segment patterns into information of a large segment pattern and information of a small segment pattern; and

straight line extracting means for examining a link state of the one or more segment, patterns, and when a large segment pattern is linked to small segment patterns, extracting a straight line pattern composed of the small segment patterns regardless of a size of the large segment pattern.

2. A ruled line extracting apparatus, comprising:

a segment extracting unit extracting black pixel concatenation regions from an input image and extracting one or more segment patterns using the black pixel concatenation regions;

a straight line extracting unit extracting information of one or more straight line patterns each of which includes the extracted segment patterns; and

a straight line integrating unit integrating two straight line patterns included in the one or more straight line patterns if the two straight line patterns overlap and a size of a non-overlapping portion of one of the two straight lines is within a particular value.

3. A ruled line extracting apparatus, comprising:

a segment extracting unit extracting black pixel concatenation regions from an input image and extracting one or more segment patterns using the black pixel concatenation regions;

a straight line extracting unit extracting information of one or more straight line patterns each of which includes the extracted segment patterns; and

a straight line integrating unit recognizing a straight line pattern after integration as a ruled line candidate, if a size of the straight line pattern generated by integrating two straight line patterns which partially overlap among the one or more straight line patterns, is equal to or lower than a threshold value.

4. A computer-readable storage medium, when used by a computer, to direct the computer to perform the functions of:

extracting information of one or more segment patterns from an input image;

classifying the information of the one or more segment patterns into information of a large segment pattern and information of a small segment pattern; and

examining a link state of one or more segment patterns, and when a large segment pattern is linked to small segment patterns, extracting a straight line pattern composed of the small segment patterns regardless of a size of the large segment pattern.

5. A computer-readable storage medium, when used by a computer, to direct the computer to perform the functions of:

extracting black pixel concatenation regions from an input image and extracting one or more segment patterns using the black pixel concatenation regions;

extracting information of one or more straight line patterns each of which includes the extracted segment patterns; and

integrating two straight line patterns included in the one or more straight line patterns into one, if the two straight line patterns overlap and a size of a non-overlapping portion of one of the two straight lines is within a particular value.

6. A computer-readable storage medium, when used by a computer, to direct the computer to perform the functions of:

extracting black pixel concatenation regions from an input image and extracting one or more segment patterns using the black pixel concatenation regions;

extracting information of one or more straight line patterns each of which includes the extracted segment patterns; and

recognizing a straight line pattern after integration as a ruled line candidate, if a size of the straight line pattern generated by integrating two straight line patterns which partially overlap among the one or more straight line patterns is equal to or lower than a threshold value.

7. A ruled line extracting method, comprising the steps of:

extracting information of one or more segment patterns from an input image;

classifying the information of the one or more segment patterns into information of a large segment pattern and information of a small segment pattern; and

examining a link state of one or more segment patterns, and when a large segment pattern is linked to small segment patterns, extracting a straight line pattern composed of the small segment patterns regardless of a size of the large segment pattern.

8. A ruled line extracting method, comprising the steps of:

extracting black pixel concatenation regions from an input image and extracting one or more segment patterns using the black pixel concatenation regions;

extracting information of one or more straight line patterns each of which includes the extracted segment patterns; and

integrating two straight line patterns included in the one or more straight line patterns, if the two straight line patterns overlap and a size of a non-overlapping portion of one of the two straight lines is within a particular value.

9. A ruled line extracting method, comprising the steps of:

extracting black pixel concatenation regions from an input image and extracting one or more segment patterns using the black pixel concatenation regions;

extracting information of one or more straight line patterns each of which includes the extracted segment patterns;

recognizing a straight line pattern after integration as a ruled line candidate, if a size of the straight line pattern generated by integrating two straight line patterns which partially overlap among the one or more straight line patterns, is equal to or lower than a threshold value.

10. A ruled line extracting apparatus, comprising:

a segment extracting unit extracting at least one black pixel concatenation region from an input image and extracting at least one segment pattern using the black pixel concatenation region; and

a straight line integrating unit integrating at least two straight line patterns included the extracted segment pattern if the two straight line patterns overlap and a size of a non-overlapping portion of one of the two straight lines is within a threshold,

wherein the straight line integrating unit integrates the straight line patterns after the segment extracting unit extracts the black pixel concatenation region.