Image processing apparatus and method for determining whether an image is of a vertical or horizontal writing

• US 7,369,713 B2
• Filed: 07/12/2007
• Issued: 05/06/2008
• Est. Priority Date: 04/10/2002
• Status: Expired due to Fees

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: Suino, Tooru
• Primary Examiner(s): LAROSE, COLIN M
• Application Number:
  US11/777,056
• Publication Number:
  US 20070258660A1
• Time in Patent Office: 299 Days
• Field of Search: 382/173, 382/176, 382/177, 382/185, 382/290, 382/297, 358/462
• US Class Current: 382/290
• CPC Class Codes: G06T 2207/10008 from scanner, fax or copier
• CPC Class Codes: G06T 2207/20064 Wavelet transform [DWT]
• CPC Class Codes: G06T 5/10 using non-spatial domain filtering
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone
• CPC Class Codes: H04N 1/4095 Correction of errors due to scanning a two-sided document, i.e. show-through correction

Abstract

An image processing apparatus, including a transform unit configured to wavelet-transform image data into a set of wavelet coefficients, and a determination unit configured to determine whether the image data is of a vertical writing or of a horizontal writing based on the set of wavelet coefficients.

Claims

1. An image processing apparatus, comprising:

a transform unit configured to wavelet-transform image data into a set of wavelet coefficients; and

a determination unit configured to determine whether the image data is of a vertical writing or of a horizontal writing based on the set of wavelet coefficients,

wherein the determination unit includes

a vertical detection unit configured to detect vertical writing patterns by detecting edge components of a vertical component included in the wavelet coefficients;

a vertical count unit configured to count the vertical writing patterns by counting the edge components of the vertical component detected by the vertical detection unit;

a horizontal detection unit configured to detect horizontal writing patterns by detecting edge components of a horizontal component included in the wavelet coefficients;

a horizontal count unit configured to count the horizontal writing patterns by counting the edge components of the horizontal component detected by the horizontal detection unit; and

a vertical/horizontal determination unit configured to determine whether the image data is of the vertical writing or of the horizontal writing based on counts of the vertical count unit and the horizontal count unit.

2. The image processing apparatus as claimed in claim 1, further comprising an image rotation unit configured to rotate the image data in accordance with the determination made by the determination unit.

3. An image processing apparatus, comprising:

a transform unit configured to wavelet-transform image data into a set of wavelet coefficients; and

a determination unit configured to determine whether the image data is of a vertical writing or of a horizontal writing based on the set of wavelet coefficients, wherein the set of wavelet coefficients comprises “

n”

layers, and the determination unit determines whether the image data is of the vertical writing or of the horizontal writing using upper ones of the “

n”

layers and, if the determination of whether the image data is of the vertical writing or of the horizontal writing can not be made using the upper ones of the “

n”

layers, the determination unit uses lower ones of the “

n”

layers in turn.

4. A method of processing image data, the method comprising:

wavelet-transforming image data into a set of wavelet coefficients; and

determining whether the image data is of a vertical writing or of a horizontal writing based on the set of wavelet coefficients,

wherein the determining comprises

detecting vertical writing patterns by detecting edge components of a vertical component included in the wavelet coefficients;

counting the vertical writing patterns by counting the edge components of the vertical component detected;

detecting horizontal writing patterns by detecting edge components of a horizontal component included in the wavelet coefficients;

counting the horizontal writing patterns by counting the edge components of the horizontal component detected; and

determining whether the image data is of the vertical writing or of the horizontal writing based on counts of the vertical writing patterns and the horizontal writing patterns.

5. The method as claimed in claim 4, further comprising rotating the image data in accordance with the determination made of whether the image data is of the vertical writing or of the horizontal writing.

6. A method of processing image data, the method comprising:

wavelet-transforming image data into a set of wavelet coefficients; and

determining whether the image data is of a vertical writing or of a horizontal writing based on the set of wavelet coefficients,

wherein the set of wavelet coefficients comprises “

n”

layers, and the determination of whether the image data is of the vertical writing or of the horizontal writing is made using upper ones of the “

n”

layers and, if the determination of whether the image data is of the vertical writing or of the horizontal writing cannot be made using the upper ones of the “

n”

layers, lower ones of the “

n”

layers are used in turn.