Image processing apparatus and control method

• US 8,401,306 B2
• Filed: 03/31/2010
• Issued: 03/19/2013
• Est. Priority Date: 04/13/2009
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Dai, Xiaoyan
• Primary Examiner(s): Chen, Wenpeng
• Assistant Examiner(s): Niu, Feng
• Application Number:
  US12/751,714
• Publication Number:
  US 20100260417A1
• Time in Patent Office: 1,084 Days
• Field of Search: None
• US Class Current: 382/199
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/162 Quantising the image signal
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

An image processing apparatus includes a block background/foreground determination unit which determines, based on a block attribute and quantized color information, whether each connected area is a foreground or a background, a block background/foreground attribute determination unit which determines, based on the block attribute, the attribute of a connected area determined to be the foreground of the block and the attribute of a connected area determined to be the background of the block, and a block background/foreground attribute information recording unit which records and holds information of the attribute of the block and the attribute of each connected area.

Claims

1. An image processing apparatus comprising:

a block attribute determination unit which determines, for each block including a predetermined number of pixels of a document image, a block attribute based on the number of edge pixels included in the block, the number of connected edges formed from the edge pixels, and the number of edge pixels included in a connected area formed from the edge pixels,

wherein the block attribute determination unit determines the block attribute as a background block, which is formed with background pixels, when the number of edge pixels included in the block is not more than a first threshold, and

wherein the block attribute determination unit determines the block attribute as a character block on background area, which is formed with background pixels and character pixels, when the number of edge pixels included in the block is more than the first threshold and the number of connected edges is less than a second threshold, and

wherein the block attribute determination unit determines the block attribute as a halftone dot block, which is formed from halftone dot pixels, when the number of edge pixels included in the block is more than the first threshold, the number of connected edges is not less than the second threshold, and there is no connected edge whose number of edge pixels included in the connected area is not less than the third threshold, and

wherein the block attribute determination unit determines the block attribute as a character block on halftone dot area, which is formed from halftone dot pixels and character pixels, when the number of edge pixels included in the block is more than the first threshold, the number of connected edges is not less than the second threshold, and there is a connected edge whose number of edge pixels included in the connected area is not less than the third threshold;

a block color quantization unit which quantizes color information of each pixel of the block including the predetermined number of pixels of the document image;

a block connected area extraction unit which extracts, based on the color information of each pixel quantized by the block color quantization unit, connected areas each included in the block and including connected pixels of the same color;

a block background/foreground determination unit which determines whether each connected area is a foreground or a background, based on the block attribute determined by the block attribute determination unit, the color information quantized by the block color quantization unit, and the connected areas acquired by the block connected area extraction unit;

a block background/foreground attribute determination unit which determines, based on the block attribute determined by the block attribute determination unit, an attribute of a connected area determined by the block background/foreground determination unit to be the foreground of the block and an attribute of a connected area determined to be the background of the block; and

a block background/foreground attribute information recording unit which records and holds information of the attribute of the block determined by the block attribute determination unit and the attribute of each connected area determined by the block background/foreground attribute determination unit.

2. The apparatus according to claim 1, wherein the block color quantization unit performs tone reduction to generate one of binary data and multilevel data of the pixels of the block.

3. The apparatus according to claim 1, wherein

the block background/foreground determination unit acquires, as a background color, a color of a block whose attribute has been determined by the block attribute determination unit as one of background and halftone dot, and

if a color of the connected area included in the block whose attribute has been determined as one of character on background area and character on halftone dot area is different from the background color, the block background/foreground determination unit determines the connected area as the foreground.

4. The apparatus according to claim 1, wherein

based on the block attribute determined by the block attribute determination unit and a result of the connected area determined by the block background/foreground determination unit,

the block background/foreground attribute determination unit determines the background of the block as halftone dot if the block attribute is one of halftone dot and character on halftone dot area,

determines the background of the block as background if the block attribute is one of background and character on background area,

determines the connected area determined to be the foreground of the block as character on background area if the block attribute is character on background area, and

determines the connected area determined to be the foreground of the block as character on halftone dot area if the block attribute is character on halftone dot area.

5. A control method comprising:

a block attribute determination step of causing a block attribute determination unit of an image processing apparatus to determine, for each block including a predetermined number of pixels of a document image, a block attribute based on the number of edge pixels included in the block, the number of connected edges formed from the edge pixels, and the number of edge pixels included in a connected area formed from the edge pixels,

wherein the block attribute determination step determines the block attribute as a background block, which is formed with background pixels, when the number of edge pixels included in the block is not more than a first threshold, and

wherein the block attribute determination step determines the block attribute as a character block on background area, which is formed with background pixels and character pixels, when the number of edge pixels included in the block is more than the first threshold and the number of connected edges is less than a second threshold, and

wherein the block attribute determination step determines the block attribute as a halftone dot block, which is formed from halftone dot pixels, when the number of edge pixels included in the block is more than the first threshold, the number of connected edges is not less than the second threshold, and there is no connected edge whose number of edge pixels included in the connected area is not less than the third threshold, and

wherein the block attribute determination step determines the block attribute as a character block on halftone dot area, which is formed from halftone dot pixels and character pixels, when the number of edge pixels included in the block is more than the first threshold, the number of connected edges is not less than the second threshold, and there is a connected edge whose number of edge pixels included in the connected area is not less than the third threshold;

a block color quantization step of causing a block color quantization unit of the image processing apparatus to quantize color information of each pixel of the block including the predetermined number of pixels of the document image;

a block connected area extraction step of causing a block connected area extraction unit of the image processing apparatus to extract, based on the color information of each pixel quantized in the block color quantization step, connected areas each included in the block and including connected pixels of the same color;

a block background/foreground determination step of causing a block background/foreground determination unit of the image processing apparatus to determine whether each connected area is a foreground or a background, based on the block attribute determined in the block attribute determination step, the color information quantized in the block color quantization step, and the connected areas acquired in the block connected area extraction step;

a block background/foreground attribute determination step of causing a block background/foreground attribute determination unit of the image processing apparatus to determine, based on the block attribute determined in the block attribute determination step, an attribute of a connected area determined in the block background/foreground determination step to be the foreground of the block and an attribute of a connected area determined to be the background of the block; and

a block background/foreground attribute information recording step of causing a block background/foreground attribute information recording unit of the image processing apparatus to record and hold information of the attribute of the block determined in the block attribute determination step and the attribute of each connected area determined in the block background/foreground attribute determination step.

6. A non-transitory computer-readable medium storing a program which causes a computer to function as:

a block attribute determination unit which determines, for each block including a predetermined number of pixels of a document image, a block attribute based on the number of edge pixels included in the block, the number of connected edges formed from the edge pixels, and the number of edge pixels included in a connected area formed from the edge pixels,

wherein the block attribute determination unit determines the block attribute as a background block, which is formed with background pixels, when the number of edge pixels included in the block is not more than a first threshold, and

wherein the block attribute determination unit determines the block attribute as a character block on background area, which is formed with background pixels and character pixels, when the number of edge pixels included in the block is more than the first threshold and the number of connected edges is less than a second threshold, and

wherein the block attribute determination unit determines the block attribute as a halftone dot block, which is formed from halftone dot pixels, when the number of edge pixels included in the block is more than the first threshold, the number of connected edges is not less than the second threshold, and there is no connected edge whose number of edge pixels included in the connected area is not less than the third threshold, and

wherein the block attribute determination unit determines the block attribute as a character block on halftone dot area, which is formed from halftone dot pixels and character pixels, when the number of edge pixels included in the block is more than the first threshold, the number of connected edges is not less than the second threshold, and there is a connected edge whose number of edge pixels included in the connected area is not less than the third threshold;

a block color quantization unit which quantizes color information of each pixel of the block including the predetermined number of pixels of the document image;

a block connected area extraction unit which extracts, based on the color information of each pixel quantized by the block color quantization unit, connected areas each included in the block and including connected pixels of the same color;

a block background/foreground determination unit which determines whether each connected area is a foreground or a background, based on the block attribute determined by the block attribute determination unit, the color information quantized by the block color quantization unit, and the connected areas acquired by the block connected area extraction unit;

a block background/foreground attribute determination unit which determines, based on the block attribute determined by the block attribute determination unit, an attribute of a connected area determined by the block background/foreground determination unit to be the foreground of the block and an attribute of a connected area determined to be the background of the block; and

a block background/foreground attribute information recording unit which records and holds information of the attribute of the block determined by the block attribute determination unit and the attribute of each connected area determined by the block background/foreground attribute determination unit.