JP2009231945A - Image processing apparatus, image forming apparatus, control method for the image processing apparatus, and control program for the image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus that can properly discriminate the character properties. <P>SOLUTION: The image processing apparatus includes a character color determining unit 225 which determines the color of a character from input image data; a MIN(RGB) generating unit 203, which generates image data of MIN(RGB) from the input image data; a character property determining unit (MIN(RGB)) 211, which determines properties of the character from the image data of MIN(RGB); a MAX(RGB) generating unit 213, which generates image data of MAX(RGB) from the input image data, a character property determining unit (MAX(RGB)) 221, which determines properties of the character from the image data by MAX(RGB); and a character property selecting unit 227, which selectively outputs the character property determination result of MIN(RGB), or the character property determination result by MAX(RGB). <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image forming apparatus, a control method for the image processing apparatus, and a control program for the image processing apparatus, and more particularly to an image processing apparatus, an image forming apparatus, and an image processing apparatus capable of determining a character attribute in an image. The present invention relates to a control method for a processing apparatus and a control program for an image processing apparatus.

  An image forming apparatus (MFP (Multi Function Peripheral), facsimile apparatus, copier, printer, etc.) includes an image processing apparatus that processes an image read by a scanner and converts it into an image suitable for printing. ing.

  FIG. 8 is a block diagram illustrating a configuration of a character attribute determination processing unit provided in the MFP.

The character attribute discrimination processing unit includes a data input unit 301 that inputs image data to be processed obtained by scanning or the like, and a MIN (RGB) generation unit 303 that creates MIN (RGB) image data from the input image data. An edge discriminating unit 305 that discriminates edges from MIN (RGB) image data, a closing processing unit 307 that performs closing processing on the edge discriminating results, and a negative positive that discriminates negative / positive boundaries from MIN (RGB) image data. A boundary determination unit 309; a lightness / saturation generation unit 323 that creates lightness / saturation data from the input image data; a character color determination unit 325 that determines the color of a character based on the lightness / saturation data;
A character attribute determining unit 327 that determines character attributes based on MIN (RGB) image data, edge data after closing processing, negative / positive boundary data, and character color.

  A known differential filter or the like is used for edge discrimination.

  The negative / positive boundary determination is for determining a dark region (region where positive characters are present) and a thin region (region where negative characters are present) of the image.

  9 and 10 are diagrams showing specific processing of negative / positive boundary determination.

  As shown in FIG. 9, MAX data and MIN data of a pixel in a filter of a certain size are determined, and (MAX + MIN) / 2 is set as reference data REF. As shown in FIG. 10, the positive area and the negative area are determined by comparing the input data DIN with the reference data REF.

  More specifically, a pixel having a density exceeding the reference data REF is determined as a positive area pixel, and a pixel having a density equal to or lower than the reference data REF is determined as a negative area pixel.

  The lightness / saturation generation unit 325 obtains lightness V and saturation W of the input data. The saturation W can be calculated by a known formula such as W = MAX (RGB) −MIN (RGB).

  By comparing the output value of the lightness-saturation reference table with the saturation of the input data, the character color discriminating unit 325 can detect the color region (region with high saturation) and the gray region (region with low saturation). Yes, including black area).

  In character discrimination processing, edge discrimination is generally performed using a brightness signal or a MIN (RGB) signal.

  FIG. 11 is a diagram showing a specific example of character edge determination using a lightness signal or a MIN (RGB) signal.

  Here, the black character (1) on a white background, the red character (2) on a white background, and the black character (3) on a dark background (for example, a dark pink background) are processed. It is shown that. MIN (RGB) images (or brightness images) obtained from the respective data are indicated by (1) -1, (2) -1, and (3) -1.

  A monochrome image formed by selecting data indicating the smallest value of each RGB color for each pixel is a MIN (RGB) image. The closer the original pixel is to white, the closer the value of MIN (RGB) at that pixel is to white. As shown in (3), in a pixel in a dark background (for example, a dark pink background), the value of MIN (RGB) in the pixel is close to black. This is because the G and B components are small on the dark pink background, and these components are selected as MIN (RGB).

  What obtained the edge from each MIN (RGB) image (or brightness image) is shown by (1) -2, (2) -2, and (3) -2.

  In (1) -1 and (2) -1, since the contrast between white and black is high, it is possible to easily determine the edge of the character part. Because there is not much contrast difference, it is difficult to determine the edge of the character part.

  That is, when the lightness or MIN (RGB) signal is used, the density difference between the character and the background becomes small for a character on a dark background, and it is difficult to discriminate the edge. Therefore, it is difficult to discriminate characters in characters on a dark color background.

  FIG. 12 is a diagram showing a specific example of character edge determination using a MAX (RGB) signal.

  Here, as in FIG. 11, the black character (4) on a white background, the red character (5) on a white background, and the black character (6) on a dark background (for example, a dark pink background). It shows that the image is the object of processing. What obtained the MAX (RGB) image from each data is shown by (4) -1, (5) -1, (6) -1.

  A monochrome image formed by selecting data indicating the largest value of each RGB color for each pixel is a MAX (RGB) image. The closer the original pixel is to white, the closer the MAX (RGB) value at that pixel is to white. As shown in (5), in a pixel in a color character, the value of MAX (RGB) in that pixel is close to white. This is because, for example, a red character has a large R component, so that component is selected as MAX (RGB).

  Further, as shown in (6), in a pixel in a dark background (for example, a dark pink background), the value of MAX (RGB) in that pixel is close to white. This is because the component of R is large on a dark pink background, and that component is selected as MAX (RGB).

  Edges obtained from the respective MAX (RGB) images are indicated by (4) -2, (5) -2, and (6) -2.

  In (4) -1 and (6) -1, since the contrast between white and black is high, it is possible to easily determine the edge of the character part. Because there is not much contrast difference, it is difficult to determine the edge of the character part.

  In other words, when the MAX (RGB) signal is used, the density difference between the character and the background becomes relatively large for a character on a dark background, and edge discrimination is possible. However, there is a side effect that it becomes difficult to distinguish color characters on a white background.

  Patent Document 1 below discloses an image processing apparatus in which a background detection unit determines a most suitable determination method according to a background state, and performs character determination using that method.

Japanese Patent Application Laid-Open No. 2004-228561 discloses a method of discriminating a black character on a color background by performing character discrimination using two physical quantities of MIN (RGB) and MAX (RGB) and ORing the discrimination result. That is, character discrimination (edge discrimination) based on MIN (RGB) and MAX (RGB) is performed in parallel, and the discrimination result is ORed.
JP 2002-330285 A JP 2000-358166 A

  As described above, it is conceivable to perform character discrimination by MIN (RGB) and MAX (RGB) in parallel and to OR the discrimination results. However, the accuracy of the result of character discrimination by MIN (RGB) is low, and noise components may be included outside the original character area. For this reason, there has been a problem that a character area cannot be accurately extracted with a simple OR.

  FIG. 13 is a diagram illustrating a state in which black characters are processed on a dark background (for example, a dark pink background).

  As described in (3) -1 of FIG. 11, in the MIN (RGB) of the image (7) in which black characters exist on a dark background (for example, dark pink background), the gradation of characters and background The difference becomes smaller. For this reason, as shown in (8), the edge may or may not be detected in the vicinity of the edge detection limit. Further, as shown in (9), the negative / positive boundary signal may be distorted because the REF value at the negative / positive boundary cannot be set appropriately.

  As shown in (10) and (11), when MAX (RGB) is used, both the edge and the negative / positive boundary can be clearly determined. When two character discrimination results are ORed, as shown in (12) The result is only a character discrimination result with low accuracy.

  The present invention has been made in order to solve such problems, and an image processing apparatus, an image forming apparatus, an image processing apparatus control method, and an image processing apparatus capable of well determining character attributes. The purpose is to provide a control program.

  In order to achieve the above object, according to one aspect of the present invention, an image processing apparatus uses a physical quantity extraction unit capable of extracting a plurality of physical quantities from image data, and each of the extracted physical quantities, A character attribute discriminator capable of discriminating a plurality of character attributes, a character color discriminator for discriminating the color of characters in the image data, and a plurality of character attributes based on the result of the character color discrimination And a character attribute selection unit for selecting a discrimination result to be output.

  Preferably, the image processing apparatus further includes an edge determination unit capable of performing a plurality of edge determinations using each of the extracted plurality of physical quantities, and the character attribute determination unit is a result of each of the plurality of edge determinations. Based on the above, a plurality of character attributes are discriminated.

  Preferably, the character attribute determining unit determines a character color attribute indicating whether the character is a gray character or a color character.

  Preferably, the character attribute determining unit determines an edge portion attribute indicating whether or not the character is an edge portion.

  Preferably, the character attribute determining unit determines a negative / positive attribute indicating whether the character is a positive character or a negative character.

  Preferably, the character attribute determining unit determines an attribute indicating an outer edge of a black character.

  Preferably, the character attribute selection unit selects and outputs an optimum discrimination result based on a character color discrimination result which is each of a plurality of character attribute discrimination results.

  According to another aspect of the present invention, an image forming apparatus scans an original image, processes the image scanned by the scan unit, and outputs the image processing apparatus according to any one of the above and a character attribute selection unit A character region discriminating unit that discriminates the character region based on the image data, and an image data correcting unit that corrects the image scanned by the scanning unit based on the discrimination result of the character region discriminating unit.

  According to another aspect of the present invention, a control method for an image processing apparatus includes a physical quantity extraction step capable of extracting a plurality of physical quantities from image data, and a plurality of characters using each of the extracted plurality of physical quantities. Character attribute determination step capable of performing attribute determination, character color determination step for determining character color in image data, and a plurality of character attribute determination results based on the result of character color determination And a character attribute selection step for selecting a discrimination result to be output.

  According to still another aspect of the present invention, a control program for an image processing apparatus uses a physical quantity extraction step capable of extracting a plurality of physical quantities from image data, and a plurality of extracted physical quantities, respectively. Character attribute determination step capable of performing character attribute determination, character color determination step for determining character color in image data, and determination of a plurality of character attributes based on the result of character color determination Of the results, the computer is caused to execute a character attribute selection step of selecting a discrimination result to be output.

  According to these inventions, it is possible to provide an image processing apparatus, an image forming apparatus, a control method for the image processing apparatus, and a control program for the image processing apparatus that can satisfactorily determine character attributes.

  The following describes an image processing apparatus that performs character discrimination in one embodiment of the present invention. The image processing apparatus is provided in the image forming apparatus.

  The image processing apparatus extracts a plurality of physical quantities from the image data, and performs character attribute discrimination on the plurality of physical quantities in parallel. The image processing apparatus selects a character attribute determination result generated from an optimum physical quantity among a plurality of character attribute determination results.

  According to such a configuration, it is possible to select a discrimination result based on an optimum physical quantity for character attributes such as a color / black character region, an edge / non-edge region, and a positive / negative region, thereby improving the character extraction accuracy. .

  FIG. 1 is a diagram showing an overall configuration of a four-cycle image forming apparatus according to one embodiment of the present invention.

  Here, the image forming apparatus is the MFP 1. The MFP 1 is a multifunction device having functions such as a scanner, a copier, a printer, and a fax machine.

  The MFP 1 includes a plurality of keys 11a, an operation unit 11 that accepts input of various instructions by a user's operation on the keys, data such as characters and numbers, a display that displays an instruction menu for the user, information about an acquired image, and the like. 12. A scanner unit 13 that optically reads a document to obtain image data; a printer unit 14 that prints an image on a recording sheet based on the image data; an ADF 17 that transports paper; and a paper feed that stores paper for image formation A tray 18 and a paper discharge tray 19 are provided.

  In addition, the machine is provided with a power supply unit 16 that converts a DC power supply voltage to be supplied to each element from an AC power supply, and a controller unit 23 that is equipped with arithmetic and storage functions for controlling the machine.

  FIG. 2 is a block diagram of an image processing apparatus (copy processing unit) included in the image forming apparatus of FIG.

  Referring to the drawing, an image processing apparatus includes a scanning unit 101 that optically reads a document, an image data adjustment unit 103 that adjusts data obtained by scanning, an edge that performs edge determination of image data and negative / positive boundary determination. A discriminating / negative / positive discriminating unit 105, a character color discriminating unit 107 for discriminating the color of the character in the image data, a character region discriminating unit 109 for discriminating the character region in the image data based on various attributes of the character, and An isolated point discriminating unit 111 for discriminating isolated points, a periodicity determining unit 113 for discriminating periodicity of isolated points, a halftone dot region discriminating unit 115 for discriminating halftone dot regions based on isolated points, and a color space of image data A color space conversion unit 117 that performs conversion, an image data correction unit 119 that corrects image data based on the character region determination result and the halftone dot region determination result, and a screen It comprises a screen / error diffusion processing unit 121 for performing management and error diffusion processing, and a printed-out portion 123 for printing out the data after processing.

  The document is scanned by the scanning unit 101. The image data is captured as RGB data. The image data adjustment unit 103 performs background removal processing and color adjustment on RGB data.

  The color space conversion unit 117 converts the RGB data into YMCK data that can be printed out. For the image data, the processing in the edge determination / negative / positive boundary determination unit 105 for determining the edge and the negative / positive boundary and the processing in the isolated point determination unit 111 for determining the isolated point of the image data are executed in parallel.

  In the image data correction unit 119, the character portion performs edge enhancement processing to reproduce more clearly, and the halftone dot portion performs smoothing to prevent the occurrence of moire due to isolated points in the halftone dot. Processing is performed.

  The corrected image data is subjected to screen processing and error diffusion processing, and converted into data to be printed by an actual engine. Finally, an image is printed out by the engine in the printout unit 123. In this way, the image processing apparatus functions as a copy processing unit.

  FIG. 3 is a diagram illustrating details (character attribute determination processing unit) of the edge determination / negative / positive boundary determination unit 105 and the character color determination unit 107 of FIG.

  The character attribute determination processing unit includes a lightness / saturation generation unit 223 that creates lightness / saturation data from the input image data, a character color determination unit 225 that determines the color of the character based on the lightness / saturation data, and A data input unit 201 that inputs image data to be processed, a MIN (RGB) generation unit 203 that creates MIN (RGB) image data from the input image data, and an edge from the MIN (RGB) image data An edge discriminating unit 205 that discriminates, a closing processing unit 207 that performs a closing process on the edge discrimination result, a negative / positive border discriminating unit 209 that discriminates a negative / positive border from MIN (RGB) image data, and an MIN (RGB) image Statement that determines character attributes based on data, edge data after closing, negative / positive boundary data, and character color An attribute determination unit (MIN (RGB)) 211, a MAX (RGB) generation unit 213 that generates MAX (RGB) image data from input image data, and an edge that determines an edge from MAX (RGB) image data A determination unit 215; a closing processing unit 217 that performs a closing process on the edge determination result; a negative / positive boundary determination unit 219 that determines a negative / positive boundary from MAX (RGB) image data; and MAX (RGB) image data and closing Based on the processed edge data, negative / positive boundary data, and character color, a character attribute determination unit (MAX (RGB)) 221 for determining the attribute of the character, a character attribute determination result by MIN (RGB), or MAX (RGB ) And a character attribute selection unit 227 for selectively outputting the character attribute discrimination result by It is.

  Each of the character attribute determining unit (MIN (RGB)) 211 and the character attribute determining unit (MAX (RGB)) 221 determines the following attributes.

  (1) Character color attribute (whether the character part is gray (achromatic) or color (chromatic))

  (2) Edge part attribute (whether it is the edge part of a character)

  (3) Negative / positive attribute (whether the character part is positive or negative)

  (4) Whether it is the outer edge of the character

  The character attribute selection unit 227 refers to the character color discrimination result for each of the plurality of character attributes, selects an optimum character attribute, and outputs it.

  FIG. 4 is a flowchart showing processing executed by the character attribute selection unit 227.

  In step S101, it is determined whether the color of the character is black (or an achromatic color such as gray). If YES, in step S103, the character attribute discrimination result by MAX (RGB) is selectively output. If NO, in step S105, a character attribute discrimination result based on MIN (RGB) is selectively output.

  FIG. 5 is a diagram illustrating a specific example of image data input by the data input unit.

  Here, it is assumed that two characters “A” are arranged in the vertical direction, the upper character is white and black, and the lower character is red on white.

  FIG. 6 is a diagram illustrating a configuration of image data of black characters in white characters in FIG.

  The black character edge surrounds the inside of the black character. The periphery is the outer edge. Around the outer edge is a base.

  FIG. 7 is a diagram illustrating a configuration of image data of red characters (color characters) in white characters in FIG.

  The edge of the color character surrounds the inside of the color character. The periphery is the outer edge. Around the outer edge is a base.

  The character discrimination result by MAX (RGB) is used in the area inside the black character and the area of the black character edge, and the character discrimination result by MIN (RGB) is used in the other areas. Thereby, the character on a color ground can be discriminated.

  Further, if the portion of the black character edge attribute is dilated (extended) and the outer edge attribute included in the area is obtained, the outer edge attribute around the black character can be determined. That is, the outer edge portion around the true black character can be determined by performing dilation processing on the edge portion determined to be a black character and taking an AND with the region determined as the outer edge by MAX (RGB).

  If the outer edge portion is determined to be a color character attribute derived from MIN (RGB), it can be said that it is noise. Such color character attributes can be denoised by replacing them with outer edges.

  That is, the outer edge attribute portion around the black character obtained from MAX (RGB) may include noise derived from MIN (RGB). For this reason, the attribute of MAX (RGB) is used in the outer edge attribute portion around the black character obtained from MAX (RGB). Thereby, the location becomes an outer edge, and noise can be excluded.

  [Effects of the embodiment]

  According to the above configuration, the character attribute can be determined using two kinds of physical quantities, MAX (RGB) and MIN (RGB). Thereby, the character extraction accuracy can be improved.

  [Others]

  The present invention can be implemented for an image forming apparatus such as an MFP, a facsimile machine, and a copying machine.

  Further, the processing in the above-described embodiment may be performed by software or by using a hardware circuit.

  In addition, a program for executing the processing in the above-described embodiment can be provided, and the program is recorded on a recording medium such as a CD-ROM, a flexible disk, a hard disk, a ROM, a RAM, and a memory card and provided to the user. You may decide to do it. The program may be downloaded to the apparatus via a communication line such as the Internet.

  In addition, it should be thought that the said embodiment is an illustration and restrictive at no points. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

1 is a diagram illustrating an overall configuration of a four-cycle image forming apparatus according to one embodiment of the present invention. FIG. 2 is a block diagram of a copy processing unit provided in the image forming apparatus of FIG. 1. It is a figure which shows the detail (character attribute discrimination | determination processing part) of the edge discrimination | determination / negative / positive boundary discrimination | determination part 105 and the character color discrimination | determination part 107 of FIG. It is a flowchart which shows the process which the character attribute selection part 227 performs. It is a figure which shows the specific example of the image data which a data input part inputs. It is a figure which shows the structure of the image data of the black character in the white character of FIG. It is a figure which shows the structure of the image data of the red character (color character) in the white character of FIG. 2 is a block diagram showing a configuration of a character attribute determination processing unit provided in the MFP. FIG. It is a figure which shows the specific process of negative / positive boundary discrimination | determination. It is a figure which shows the specific process of negative / positive boundary discrimination | determination. It is a figure which shows the specific example of the edge determination of the character using a brightness signal and a MIN (RGB) signal. It is a figure which shows the specific example of the edge determination of the character using a MAX (RGB) signal. It is a figure which shows the state which processes the black character in the background (for example, dark pink background) of a dark density.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 4-cycle color image forming apparatus 11 Operation part 11a Panel key 12 Display 13 Scanner part 14 Printer part 16 Power supply part 23 Controller part 101 Scan part 103 Image data adjustment part 105 Edge discrimination | determination / negative / positive discrimination part 107 Character color discrimination part 109 Character area Discriminating unit 111 Isolated point discriminating unit 113 Periodicity discriminating unit 115 Halftone dot region discriminating unit 117 Color space conversion unit 119 Image data correction unit 121 Screen / error diffusion processing unit 123 Printout unit 201 Data input unit 203 MIN (RGB) generation unit 205 Edge discriminating unit 207 Closing processing unit 209 Negative / positive boundary discriminating unit 211 Character attribute discriminating unit (MIN (RGB))
213 MAX (RGB) generation unit 215 Edge determination unit 217 Closing processing unit 219 Negative / positive boundary determination unit 221 Character attribute determination unit (MAX (RGB))
223 brightness / saturation generation unit 225 character color discrimination unit 227 character attribute selection unit

Claims (10)

A physical quantity extraction unit capable of extracting a plurality of physical quantities from image data;
A character attribute determination unit capable of determining a plurality of character attributes using each of the extracted physical quantities;
A character color discriminating unit for discriminating the color of the character in the image data;
An image processing apparatus comprising: a character attribute selection unit that selects a determination result to be output among the determination results of the plurality of character attributes based on a determination result of the character color. An edge discriminator capable of performing a plurality of edge discriminations using each of the plurality of extracted physical quantities,
The image processing apparatus according to claim 1, wherein the character attribute determination unit determines a plurality of character attributes based on a result of each of the plurality of edge determinations.   The image processing apparatus according to claim 1, wherein the character attribute determination unit determines a character color attribute indicating whether the character is a gray character or a color character.   The image processing apparatus according to claim 1, wherein the character attribute determination unit determines an edge part attribute indicating whether or not the character is an edge part.   The image processing apparatus according to claim 1, wherein the character attribute determination unit determines a negative / positive attribute indicating whether the character is a positive character or a negative character.   The image processing apparatus according to claim 1, wherein the character attribute determination unit determines an attribute indicating an outer edge of a black character.   The image processing according to claim 1, wherein the character attribute selection unit selects and outputs an optimum discrimination result based on a character color discrimination result that is each of a plurality of character attribute discrimination results. apparatus. A scanning unit for scanning an original image;
The image processing apparatus according to any one of claims 1 to 7, which processes an image scanned by the scanning unit;
A character area determination unit for determining a character area based on the output of the character attribute selection unit;
An image forming apparatus comprising: an image data correction unit that corrects an image scanned by the scan unit based on a determination result of the character region determination unit. A physical quantity extraction step capable of extracting a plurality of physical quantities from image data;
Character attribute determination step capable of determining a plurality of character attributes using each of the extracted physical quantities,
A character color determining step for determining a color of a character in the image data;
A control method for an image processing apparatus, comprising: a character attribute selection step of selecting a determination result to be output from among the plurality of character attribute determination results based on the character color determination result. A physical quantity extraction step capable of extracting a plurality of physical quantities from image data;
Character attribute determination step capable of determining a plurality of character attributes using each of the extracted physical quantities,
A character color determining step for determining a color of a character in the image data;
A control program for an image processing apparatus that causes a computer to execute a character attribute selection step of selecting a determination result to be output from among the plurality of character attribute determination results based on the character color determination result.

Images