Image forming apparatus capable of recognizing top and bottom of document image

• US 5,889,884 A
• Filed: 05/23/1996
• Issued: 03/30/1999
• Est. Priority Date: 05/23/1995
• Status: Expired due to Term

• Current Assignee: Minolta Corporation Limited (Konica Minolta Inc.)
• Original Assignee: Minolta Corporation Limited (Konica Minolta Inc.)
• Inventors: Imaizumi, Shoji, Sakatani, Kazuomi, Nabeshima, Takayuki, Hashimoto, Hideyuki
• Primary Examiner(s): Shalwala, Bipin
• Assistant Examiner(s): Mariam, Daniel G.
• Application Number:
  US08/652,180
• Time in Patent Office: 1,041 Days
• Field of Search: 382/168, 382/170, 382/171, 382/174, 382/177
• US Class Current: 382/168
• CPC Class Codes: G06V 10/242 by image rotation, e.g. by 90 degrees

Abstract

An image forming apparatus automatically identifying top and bottom of document image based on characteristics with regard to shapes or positions of data, etc. recorded on document. Distribution states of pixel cumulative values in main scanning direction and sub scanning direction of image signals provided by scanning the document, are analyzed and top and bottom of the document image is identified from the result of analysis. Further, it is possible to identify the top and bottom of the document image by analyzing the distribution states the pixel cumulative values in the main scanning direction and the sub scanning direction of the image signals and identifying letter series regions and line segment regions displaced therefrom. In addition thereto, it is possible to automatically identify the top and bottom of the document image based on distribution states in the main scanning direction and the sub scanning direction of specific signs included in the image signals provided by scanning the document or distribution states of pixel cumulative values of specific image regions including the specific signs.

Claims

1. An image processing apparatus for processing pixel image data of a text image having a plurality of letters arranged in a plurality of letter lines, said image processing apparatus comprising:

a first histogram forming means for forming a first histogram indicating a distribution state of the pixel image data in a first direction;

a second histogram forming means for forming a second histogram indicating a distribution state of the pixel image data in a second direction perpendicular to the first direction;

an identifying means for identifying whether the plurality of letters of the text image are arranged in the plurality of letter lines in the first direction or in the second direction based on the first histogram and the second histogram;

a region dividing means for dividing each of the plurality of letter lines into three regions using a histogram identified by the identifying means as being perpendicular to the direction of the plurality of letter lines, the three regions including a first region having a predetermined width, a second region placed on an external side of the first region, and a third region placed on an external side of the first region and opposed to the second region;

a comparing means for comparing a distribution of the pixel image data of the second region with a distribution of the pixel image data of the third region thereof for each of the plurality of letter lines and for determining a top and a bottom of each of the plurality of letter lines; and

a recognizing means for recognizing a top and a bottom of the text image as a whole in accordance with a result of determination of the comparing means.

2. An image processing apparatus according to claim 1,

wherein the identifying means compares a shape of the first histogram and a shape of the second histogram with a predetermined pattern and identifies that the letters of the text image are arranged in the first direction when the shape of the first histogram is similar to the predetermined pattern and identifies that the letters of the text image are arranged in the second direction when the shape of the second histogram is similar to the predetermined pattern.

3. An image processing apparatus according to claim 1,

wherein the identifying means compares a shape of the first histogram and a shape of the second histogram with a predetermined pattern and determines that the direction of the plurality of letter lines cannot be identified when both of the shape of the first histogram and the shape of the second histogram are similar to the predetermined pattern.

4. An image processing apparatus according to claim 1,

wherein the identifying means compares a shape of the first histogram and a shape of the second histogram with a predetermined pattern and determines that the direction of the plurality of letter lines cannot be identified when neither of the shape of the first histogram and the shape of the second histogram is similar to the predetermined pattern.

5. An image processing apparatus according to claim 1,

wherein said region dividing means divides the letter line area by setting boundary lines at positions of peaks value of the histogram of which direction is perpendicular to the direction to the plurality of letter lines.

6. An image processing apparatus according to claim 1,

wherein said comparing means compares a peak value in the second region of the histogram with a peak value in the third region of the histogram, and determines one of the second region and the third region having a larger peak value to be on a top side of the letter line.

7. An image processing apparatus according to claim 1,

wherein said recognizing means includes a first counter which counts a number of letter lines of which the second region is determined to be on the top side by said comparing means and a second counter which counts a number of letter lines of which the third region is determined to be on the top side by said comparing means, and recognizes top and bottom of the text image as a whole based on count values of the first and second counters.

8. An image processing method for recognizing top and bottom of a text image having letters arranged in a plurality of lines in a first direction or in a second direction perpendicular to the first direction, said image processing method comprising steps of:

forming a first histogram indicating a distribution state of the pixel image data in the first direction;

forming means for forming a second histogram indicating a distribution state of the pixel image data in the second direction;

identifying whether the letters of the text image are arranged in the lines in the first direction or in the second direction based on the first histogram and the second histogram;

dividing in three a letter line area, based on one of the first and second histograms perpendicular to the direction of the letter lines identified by the identifying means, into a first region having a predetermined width in a letter line, a second region placed on an external side of the first region, and a third region placed on an external side of the first region and opposed to the second region;

comparing a distribution of the pixel image data of the second region with a distribution of the pixel image data of the third region thereof to determine top and bottom of the letter line;

repeating said dividing and comparing steps for a plurality of letter lines; and

recognizing top and bottom of the text image as a whole in accordance with a result of said repeating step.

9. An image processing method according to claim 8,

wherein said identifying step compares a shape of the first histogram and a shape of the second histogram with a predetermined pattern and identifies that the letters of the text image are arranged in the first direction when the shape of the first histogram is similar to the predetermined pattern and identifies that the letters of the text image are arranged in the second direction when the shape of the second histogram is similar to the predetermined pattern.

10. An image processing method according to claim 8,

wherein said identifying steep compares a shape of the first histogram and a shape of the second histogram with a predetermined pattern and determines that the direction of the letter line cannot be identified when both of the shape of the first histogram and the shape of the second histogram are similar to the predetermined pattern.

11. An image processing method according to claim 8,

wherein said identifying step compares a shape of the first histogram and a shape of the second histogram with a predetermined pattern and determines that the direction of the letter line cannot be identified when neither of the shape of the first histogram and the shape of the second histogram is similar to the predetermined pattern.

12. An image processing method according to claim 8,

wherein said dividing step divides the letter line area by setting boundary lines at positions of peaks value of the histogram of which direction is perpendicular to the direction of the letter lines.

13. An image processing method according to claim 8

wherein said comparing step compares a peak value in the second region of the histogram with a peak value in the third region of the histogram, and determines one of the second region and the third region having a larger peak value to be on a top side of the letter line.

14. (New) An image processing method according to claim 8, further comprising a counting step for counting a number of letter lines of which the second region is determined to be on the top side during said repeating step and for counting a number of letter lines of which the third region is determined to be on the top side during said repeating step, and

wherein said recognizing step recognizes top and bottom of the text image as a whole based on the count values of the counting step.