IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND COMPUTER-READABLE MEDIUM

• US 20110158517A1
• Filed: 12/07/2010
• Published: 06/30/2011
• Est. Priority Date: 12/28/2009
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Dai, Xiaoyan
• Granted Patent: US 8,401,303 B2
• Time in Patent Office: Days
• US Class Current: 382/165
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/146 Aligning or centring of the image pick-up or image-field
• CPC Class Codes: G06V 30/148 Segmentation of character regions
• CPC Class Codes: G06V 30/287 of Kanji, Hiragana or Katakana characters

Abstract

An image processing apparatus comprises an attribute determination unit that divides image data into a plurality of blocks each having a predetermined number of pixels and determines an attribute of each of the divided blocks that indicates whether or not the block includes a character; a connected area extraction unit that extracts a connected area in which pixels having the same pixel characteristic are connected sequentially, from each of the divided blocks; and a foreground/background determination unit that selects a foreground/background determination method to be used for a processing target block, based on the attribute of the processing target block, the attribute of a first adjacent block that is adjacent to the processing target block, and the extracted connected areas, and determines whether a connected area of the processing target block among the extracted connected areas is the foreground or the background, using the selected foreground/background determination method.

Claims

1. An image processing apparatus comprising:

an attribute determination unit that divides image data into a plurality of blocks each having a predetermined number of pixels and determines an attribute of each of the divided blocks that indicates whether or not the block includes a character;

a connected area extraction unit that extracts a connected area in which pixels having the same pixel characteristic are connected sequentially, from each of the divided blocks; and

a foreground/background determination unit that selects a foreground/background determination method to be used for a processing target block, based on the attribute of the processing target block, the attribute of a first adjacent block that is adjacent to the processing target block, and the extracted connected areas, and determines whether a connected area of the processing target block among the extracted connected areas is the foreground or the background, using the selected foreground/background determination method.

2. The image processing apparatus according to claim 1, wherein the foreground/background determination unit makes a determination such that:

in a case (i) where the attribute of the processing target block is a non-character attribute, the processing target block is determined as the background;

in a case (ii) where the attribute of the processing target block is a character attribute, and the attribute of the first adjacent block is the character attribute, and a foreground connected area of the first adjacent block is present on a boundary between the processing target block and the first adjacent block, each connected area of the processing target block is determined as either the foreground or the background based on a connection relation between the foreground connected area of the first adjacent block and the connected area of the processing target block;

in a case (iii) where the attribute of the processing target block is the character attribute, and the attribute of the first adjacent block is the character attribute, and no foreground connected area of the first adjacent block is present on the boundary, and a surrounding relation is established between connected areas of the processing target block, or in a case (iv) where the attribute of the processing target block is the character attribute, and the attribute of the first adjacent block is the non-character attribute, and a surrounding relation is established between connected areas of the processing target block, each connected area of the processing target block is determined as either the foreground or the background based on the surrounding relation;

in a case (v) where the attribute of the processing target block is the character attribute, and the attribute of the first adjacent block is the character attribute, and no foreground connected area of the first adjacent block is present on the boundary, and no surrounding relation is established between connected areas of the processing target block, or in a case where (vi) the attribute of the processing target block is the character attribute, and the attribute of the first adjacent block is the non-character attribute, and no surrounding relation is established between connected areas of the processing target block, each connected area of the processing target block is determined as either the foreground or the background based on similarity in color between a background connected area of the first adjacent block and the connected area of the processing target block.

3. The image processing apparatus according to claim 2, wherein, in the case (ii), the foreground/background determination unit determines that a color of a connected area of the processing target block that is connected to the foreground connected area of the first adjacent block that is present on the boundary is a foreground color, a connected area of the processing target block that has the same color as the foreground color is the foreground, and a connected area of the processing target block that has a color different from the foreground color is the background.

4. The image processing apparatus according to claim 2, wherein, in the case (iii) or the case (iv), the foreground/background determination unit determines that a color of a connected area of the processing target block that surrounds another connected area is a background color, a connected area of the processing target block that has a color different from the background color is the foreground, and a connected area of the processing target block that has the same color as the background color is the background.

5. The image processing apparatus according to claim 2, wherein, in the case (v) or the case (vi), the foreground/background determination unit determines that a connected area of the processing target block, that has a color similar to a color of a connected area of the first adjacent block determined as the background, is the background, and a connected area of the processing target block, that has a color dissimilar to the color of the connected area of the first adjacent block determined as the background, is the foreground.

6. The image processing apparatus according to claim 1, wherein the first adjacent block is a left-side block or an upper-side block of the processing target block.

7. The image processing apparatus according to claim 1, wherein the foreground/background determination unit further comprises an information holding unit that holds information as to determination results indicating whether each connected area of the processing target block is the foreground or the background.

8. The image processing apparatus according to claim 7, wherein

if the attribute of the processing target block is the non-character attribute and the processing target block is constituted by a connected area having a single color, the information holding unit holds the single color as a background color, and

if the attribute of the processing target block is the non-character attribute and the processing target block is constituted by a connected area having two colors, the information holding unit holds, as the background color, one of the two colors that abuts a second adjacent block at the boundary.

9. The image processing apparatus according to claim 8, wherein the second adjacent block is either a right-side block or a lower-side block of the processing target block.

10. The image processing apparatus according to claim 1, wherein

the attribute determined by the attribute determination unit is one of a character on base area, a character on halftone-dot area, a base area, and a halftone dot area, the character on base area and the character on halftone-dot area being the character attribute, and the base area and the halftone-dot area being the non-character attribute,

the image processing apparatus further comprising:

an attribute identification unit that identifies the attribute of each connected area extracted from the processing target block as one of a base area, a halftone-dot area, a character on base area, and a character on halftone-dot area, based on the attribute of the processing target block determined by the attribute determination unit and a determination result obtained by the foreground/background determination unit and indicating whether the connected area is the foreground or the background.

11. The image processing apparatus according to claim 1, further comprising a foreground integration unit that integrates a connected area of the processing target block determined as the foreground with a connected area of an adjacent block determined as the foreground, when the connected area of the processing target block determined as the foreground abuts the connected area of the adjacent block determined as the foreground.

12. An image processing method comprising:

an attribute determination step of dividing image data into a plurality of blocks each having a predetermined number of pixels, and determining an attribute of each of the divided blocks that indicates whether or not the block includes a character;

a connected area extraction step of extracting a connected area in which pixels having the same pixel characteristic are connected sequentially, from each of the divided blocks; and

a foreground/background determination step of selecting a foreground/background determination method to be used for a processing target block, based on the attribute of the processing target block, the attribute of a first adjacent block that is adjacent to the processing target block, and the extracted connected areas and determining whether a connected area of the processing target block among the extracted connected areas is a foreground or a background, using the selected foreground/background determination method.

13. A computer-readable medium storing a program for causing a computer to perform the image processing method according to claim 12.