JP2003125198A - Image processing apparatus and image processing method, and image forming device provided with the same, program, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing apparatus and image processing method that estimate a degree of recognition representing a degree of image type from input image data and performs optimum image processing on the basis of the degree of recognition, and an image forming device provided with the same, a program, and a recording medium. SOLUTION: An original type recognition section 20 extracts a plurality of parameters representing a feature of an image from input image data and uses the parameters to estimate a degree of recognition of an image type such as a degree of character and a degree of photograph. The result of the degree of recognition of the image type is outputted to an input gradation correction section 13, a color correction section 40, a black generation under color elimination section 50, and a spatial filter processing section 14, and each processing section applies proper processing to the input image data.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to an image processing apparatus and an image processing method for estimating a recognition degree indicating the degree of an image type in input image data, and performing optimum image processing based on the recognition degree. The present invention relates to an image forming apparatus, a program, and a recording medium having the same.

[0002]

2. Description of the Related Art Conventionally, in a digital copying machine or a facsimile machine, an original document having a character portion, a photographic paper photograph portion, a halftone dot portion, or an original document having a mixture of these portions is imaged by a scanner or the like. Image processing for improving the image quality of a reproduced image is performed on an image read by a reading device.

For example, in Japanese Unexamined Patent Publication No. 8-251406, first, image data of an input image for one page (hereinafter, page image) is separated into local blocks. Next, for each local block, edge determination and halftone dot determination are performed,
Each local block is classified into one of a line drawing (corresponding to a character image), a halftone image (corresponding to a halftone dot image), a continuous toning (corresponding to a photographic paper photographic image), and a background portion. Then, by using the classification result, the continuity and frequency of the image types of the local blocks in the main scanning direction and the sub scanning direction are counted. From the continuity and frequency data obtained in this way, the page drawing
Continuous tones, halftones, mixed line and halftone images,
It is determined whether the image is a mixed image of halftone and continuous tones, a mixed image of line images and continuous tones, or a mixed image of line images, intermediate tones and continuous tones, and image processing according to the determination result. Is supposed to do.

In Japanese Patent No. 3005003, a membership function used in fuzzy inference is applied to two parameters to identify the characteristics of each pixel. Then, the image processing is performed by performing a weighted average of the edge emphasis output and the smoothing output at an appropriate ratio according to the identification result.

[0005]

However, in the above-mentioned Japanese Patent Application Laid-Open No. 8-251406, when the page images of are classified into individual types such as line drawings, continuous tones, and halftone images, the image is classified according to the image type. Although it is possible to perform accurate image processing, it is possible to use four patterns according to the types of to, without considering the mixedness of the images, for images that are classified as to and in which different types of images are mixed. It is designed to perform uniform image processing. Therefore, there is a problem in that page images classified into can not be subjected to appropriate image processing.

In other words, for example, even if the page drawing has a part of the character with the photographic paper photograph, it is almost the character and there is a small photograph at one place. Is also identified as a mixed image in which continuous tones and line drawings are mixed, and the same image processing that is intermediate between continuous tones and line drawings is performed.

Therefore, the reproduced image of the former is inferior to the image reproduced by the continuous reproduction of the photographic paper photographic portion, which occupies most of the page image, and is inferior in image quality. I cannot deny the decline of. Similarly, in the latter reproduced image, the reproducibility of the characters that occupy most of the image is inferior to that of the image that has been identified as a line drawing and subjected to image processing, and the deterioration in image quality cannot be denied.

Further, in the above publication, edge determination and halftone dot determination are performed for each local block to classify into image types, and the continuity and frequency of image types of local blocks are counted using the classification result. , The image type of the entire page image is determined.

Therefore, the determination of the image type requires complicated processing, which increases the processing time.

On the other hand, in Japanese Patent No. 3005003, the edge amount is used for character detection and the changed pixel count amount is used for halftone dot detection, and the membership function is applied to these to determine the characteristics of the image data of the local block. Have identified. Then, the edge emphasis output and the smoothing output are controlled according to the identification result. Therefore, complicated image processing is required and processing time is increasing.

The present invention has been made to solve the above-mentioned conventional problems, and an object thereof is an image processing apparatus which performs optimum image processing even on input image data in which characters and photographs are mixed. And an image processing method, and an image forming apparatus, a program, and a recording medium having the same.

[0012]

In order to solve the above problems, an image processing apparatus according to the present invention is an image processing apparatus for performing image processing on input image data, and a plurality of image processing apparatuses obtained from the input image data are used. The document type recognizing means for estimating the recognition degree indicating the degree of the image type in the input image data by using the parameter and the image processing means for performing the image processing based on the recognition degree.

Input image data input by a scanner or the like is usually classified into image types (document types) of character images, photographic images, and halftone images. The image processing apparatus recognizes the image type of the input image data, performs image processing according to the recognition result, and outputs output image data.

In the above arrangement, the degree of recognition of the image type of the input image data is estimated from the plurality of parameters of the input image data. The parameter is a characteristic amount of the input image data determined based on the pixel density of the input image data, the signal level of the pixel, and the like. Further, the degree of recognition is a rate that indicates which image type an image is recognized in, and a degree of character that is a rate of being a character image, a degree of photography that is a rate of being a photographic image, and a mesh degree. It is a general term for the degree of halftone dots, which is the ratio of dot images.

Since it is difficult to estimate the degree of recognition of the image type with sufficient accuracy using only one parameter,
By combining a plurality of parameters, it is possible to accurately estimate the degree of recognition of the image type.

As a result, according to the degree of recognition, the image processing apparatus can obtain the optimum image based on the degree of recognition even for input image data in which character images, photographic images and halftone dot images are present in various proportions. Processing can be performed. The image processing includes, for example, input tone correction, color correction, black generation undercolor removal, and spatial filter processing.

For example, in the case of the above-mentioned two examples, consider the case where image processing is performed on the entire page.
As in Japanese Patent Publication No. 251406, intermediate image processing between a photograph (continuous image) and a character (line drawing) is not performed in the same manner, but the proportion of each photograph and character in the image, That is, image processing is performed based on the degree of characters and the degree of photographs. Therefore, in each of the two cases, the optimum image processing can be performed for the entire page, and the image quality can be improved.

Further, unlike Japanese Patent Application Laid-Open No. 8-251406, the degree of recognition is estimated by combining a plurality of parameters without requiring complicated processing for determining the image type. Will be easier. Therefore, the processing speed is increased and the required processing capacity is reduced.

Further, as disclosed in Japanese Patent No. 3005003, the edge amount and the changed pixel count amount that require complicated processing are not used for identifying the image type, and recognition is performed by combining a plurality of parameters. Since the degree is estimated, the processing becomes very easy. Therefore, the processing speed is increased and the required processing capacity is reduced.

In order to solve the above problems, the image processing apparatus according to the present invention is further characterized in that the degree of recognition is a degree of recognition of the input image data of the entire page area.

With the above arrangement, the image processing of the entire page area is performed based on the degree of recognition of the entire page area. Therefore, the optimum image processing can be performed on the entire page and each local area can be processed. It becomes possible to perform the processing more quickly than the processing of.

In order to solve the above problems, the image processing apparatus according to the present invention further comprises a color correction means for correcting the color space of the input image data based on the recognition degree. It is characterized by including.

With the above arrangement, the character degree, the photograph degree and the halftone dot of the input image data are further reflected in the color correction, so that the input image data can be subjected to the optimum color correction processing. Thereby, the image quality can be improved.

In order to solve the above problems, the image processing apparatus according to the present invention further comprises: the image processing means generating image data representing black from the input image data based on the degree of recognition. , And includes black generation undercolor removal means for performing image processing on the input image data by subtracting generated image data representing black from the input image data.

With the above configuration, since the black generation undercolor removal processing is further performed by reflecting the character degree, photograph degree and halftone dot degree of the input image data, the black generation undercolor removal processing optimal for the input image data is performed. Can be applied. As a result, the reproducibility of thin lines such as characters can be improved.

In order to solve the above-mentioned problems, the image processing apparatus according to the present invention is characterized by further comprising display means for displaying the input image data subjected to image processing based on the degree of recognition. .

Here, the display means is an image display device such as a CRT display or a liquid crystal display.

With the above arrangement, the image display device further displays an image which has been subjected to image processing while reflecting the character degree, the photograph degree and the halftone degree of the input image data.
This allows the user to adjust the image processing so as to obtain the desired image. Further, even when the image processing is erroneous, it can be easily corrected.

An image forming apparatus according to the present invention is characterized by including the above image processing apparatus in order to solve the above problems.

With the above-described structure, image processing is performed in the image forming apparatus such as a digital color copying machine or a facsimile by reflecting the character degree, the photograph degree and the halftone degree of the input image data to form an image. Can be improved.

In order to solve the above problems, the image processing method according to the present invention uses a plurality of parameters obtained from input image data to estimate the recognition degree indicating the degree of image type in the input image data. The image processing is performed based on the degree of recognition.

In the above method, the degree of recognition of the image type of the input image data is estimated from a plurality of parameters of the input image data.

Since it is difficult to estimate the degree of recognition of the image type with sufficient accuracy using only one parameter,
By combining a plurality of parameters, it is possible to accurately estimate the degree of recognition of the image type.

As a result, optimum image processing can be performed based on the degree of recognition even for input image data in which character images, photographic images, and halftone images are present in various proportions according to the degree of recognition. it can. Therefore, it is possible to perform the optimum image processing even on an image having an intermediate characteristic between the image types. Thereby, the image quality can be improved.

Further, unlike Japanese Patent Application Laid-Open No. 8-251406, the degree of recognition is estimated by combining a plurality of parameters without requiring complicated processing for determining the image type. Will be easier. Therefore, the processing speed is increased and the required processing capacity is reduced.

Further, as disclosed in Japanese Patent No. 3005003, the edge amount and the changed pixel count amount that require complicated processing are not used to identify the image type, and recognition is performed by combining a plurality of parameters. Since the degree is estimated, the processing becomes very easy. Therefore, the processing speed is increased and the required processing capacity is reduced.

In order to solve the above problems, the image processing method according to the present invention is further characterized in that the degree of recognition is a degree of recognition of the input image data of the entire page area.

With the above configuration, the image processing of the entire page area is performed based on the degree of recognition of the entire page area. Therefore, the optimum image processing for the entire page is performed as compared with the processing for each local area. It will be possible to do it quickly.

A program according to the present invention is characterized by causing a computer to execute the above image processing method in order to solve the above problems.

With the above program, a computer can read and execute an image processing method for performing image processing by reflecting the character degree, photograph degree, and halftone degree of input image data. It can be general-purpose.

The recording medium according to the present invention is characterized in that, in order to solve the above problems, the above program is stored so that it can be read by a computer.

With the above configuration, the program of the image processing method for performing the image processing by reflecting the character degree, the photograph degree, and the halftone degree of the input image data can be easily supplied to the computer via the recording medium. .

The recording medium is a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a floppy (registered trademark) disk or a hard disk, or a CD-RO.
Disc system of optical discs such as M, MO, MD, DVD,
Card systems such as IC cards and optical cards including memory cards, mask ROM, EPROM (Erasable Pro)
grammable Read Only Memory), EEROM (Electri
Cally Programmable Read Only Memory), a semiconductor memory such as a flash ROM, and the like may be a medium that carries the program in a fixed manner.

[0044]

BEST MODE FOR CARRYING OUT THE INVENTION [Embodiment 1] An embodiment of the present invention will be described with reference to FIGS. 1 to 8.
It is as follows.

The image forming apparatus of this embodiment is a digital color copying machine provided with the color image processing apparatus 10.

First, the overall structure of the digital color copying machine will be described. FIG. 1 is a block diagram of this digital color copying machine.

In the digital color copying machine, the color image processing apparatus 10 is connected to the color image input apparatus 1, the color image output apparatus 2, and the operation panel 3.

The color image input device 1 includes a CCD (Charge
It has a scanner unit with a d Coupled Device).
The scanner unit reads the reflected light image from the original, which is the input image data, as an analog signal of RGB by the CCD,
Input to the color image processing apparatus 10.

The color image processing apparatus 10 outputs RGB (R: red, G: green, B :) output from the color image input apparatus 1.
A CMYK (C: cyan, M: magenta, Y: yellow, K: black) digital color signal corresponding to the input image data is generated based on the (blue) analog signal.

The color image output device 2 outputs the input image data on a recording medium such as recording paper based on the digital color signal generated by the color image processing device 10. Examples of the color image output device 2 include a color image output device using an electrophotographic method or an inkjet method.

Further, the operation panel 3 is provided with a display screen (display means) formed of a liquid crystal display (not shown), and input means such as setting buttons and ten keys. Then, it has a function of receiving an instruction from the user and showing the image processing result to the user.

In the digital color copying machine described above, the input image data is input to the color image input device 1 to generate RGB signals.
After being read as an analog signal of CMYK, it is sent to the color image processing apparatus 10 and output to the color image output apparatus 2 as a CMYK digital color signal.

That is, the analog signal of the input image data input to the color image input device 1 is sent to the color image processing device 10. The color image processing apparatus 10 is shown in FIG.
As shown in, A / D (analog / digital) converter 1
1, a shading correction unit 12, a document type recognition unit (document type recognition unit) 20, an input gradation correction unit 13, a region separation processing unit (region separation unit) 30, a color correction unit (color correction unit) 4
0, a black generation undercolor removal unit (black generation undercolor removal means) 50, a spatial filter processing unit 14, an output gradation correction unit 15, and a gradation reproduction processing unit 16. The RGB analog signals input to the color image processing apparatus 10 are sent through the color image processing apparatus 10 in the above order. Then, the color image processing apparatus 10 outputs the CMYK digital color signals to the color image output apparatus 2.

Next, the image processing performed inside the color image processing apparatus 10 will be described with reference to FIG.

First, the RGB analog signals input to the color image input device 1 are sent to the A / D converter 11 in the color image processing device 10. The A / D conversion unit 11 uses R
The analog signal of GB is converted into a digital signal.

The shading correction unit 12 removes various distortions generated in the illumination system, the imaging system, and the imaging system of the color image input apparatus 1 from the RGB digital signal sent from the A / D conversion unit 11. Apply processing.

The document type recognizing unit 20 receives the RGB signal (RG
The reflectance signal of B) is subjected to complementary color inversion and converted into a CMY signal, and as will be described in detail later, the RGB signal is converted into a density signal that can be easily handled by each processing unit of the color image processing apparatus 10.

In addition to this conversion, the document type recognition unit 20
Is the document image type (image type) from the input image data
Recognize. For example, if it is a character image, a photographic image, or a mixture thereof, the degree of recognition between the character image and the photographic image, that is, the degree of character and the degree of photograph are estimated. The recognition result is the input gradation correction unit 13,
It is output to the color correction unit 40, the black generation undercolor removal unit 50, and the spatial filter processing unit 14. The document type recognition unit 20
The method of recognizing the document type performed in 1) will be described in detail later.

The input tone correction unit 13 receives the input CMY
Adjust the color balance of the signal. Further, based on the recognition result of the document type recognition unit 20, the background density is removed and image adjustment processing such as contrast is performed.

The area separation processing section 30 classifies the input image data represented by CMY signals into image areas of any of a character area, a halftone dot area and a photographic area for each pixel. The area identification signal indicating the classification result is stored in the spatial filter processing unit 14,
It is output to the gradation reproduction processing unit 16. In addition, the color correction unit 40 in the subsequent stage includes the input gradation correction unit 13 to the area separation processing unit 3.
The CMY signal output to 0 is output as it is.

The color correction section 40 performs a process for removing color turbidity based on the spectral characteristics of the CMY color materials in order to faithfully reproduce the color of the input image data. This color turbidity is C
This is due to the fact that the MY color material contains unnecessary absorption components.

The black generation and undercolor removal section 50 is for generating a black (K) signal from the CMY signals of the three colors after the color correction.
Further, the black generation undercolor removal unit 50 subtracts this K signal from the CMY signal after color correction to generate a new CMY signal. That is, the black generation undercolor removal unit 50 has a function of generating a CMYK signal including a new CMY signal and a new K signal from the CMY signals of the three colors after the color correction.

The spatial filter processing section 14 performs spatial filter processing on the CMYK signals output from the black generation and under color removal section 50. This processing is performed based on the area identification signal generated by the area separation processing unit 30. As a result, the spatial frequency characteristic is corrected, and blurring of the output image and deterioration of graininess are prevented.

The output gradation correction section 15 performs an output gradation correction process for converting a signal such as a density signal into a halftone dot area ratio, which is a characteristic value of the color image output device 2, and finally a gradation is obtained. The reproduction processing unit 16 performs gradation reproduction processing (halftone generation processing) for separating the image into pixels and processing so that each gradation can be reproduced.

Further, the pixels classified into the character area by the area separation processing unit 30 described above have a large amount of high frequency emphasis by the edge emphasis processing in the spatial filter processing unit 14. As a result, the reproducibility of black characters or color characters is improved. At the same time, the gradation reproduction processing unit 16 selects the binarization or multi-value conversion processing on the high resolution screen suitable for reproducing the high frequency.

On the other hand, the pixels classified into the halftone dot area by the area separation processing section 30 are low-pass filtered by the spatial filter processing section 14 to remove the input halftone dot component. At the same time, the gradation reproduction processing unit 16 performs binarization or multi-value conversion processing on the screen with emphasis on gradation reproducibility.

Further, the pixels classified into the photographic area by the area separation processing section 30 are subjected to binarization or multi-value conversion processing on the screen with an emphasis on gradation reproducibility.

As described above, the image data processed by each part of the color image processing apparatus 10 is temporarily stored in the storage means (not shown), read at a predetermined timing, and input to the color image output apparatus 2. To be done.

Next, the method of recognizing the degree of characters and the degree of photographs performed by the document type recognition unit 20 in the color image processing apparatus 10 will be described in detail.

As shown in FIG. 2, the document type recognizing section 20 includes a signal converting section 21, a histogram creating section 22, a first maximum frequency density category extracting section 23, and a second maximum frequency density category extracting section 24. The storage unit 25 and the recognition unit 26 are provided. FIG. 2 is a diagram showing the configuration of the document type recognition unit 20.

The signal conversion unit 21 inverts the RGB signals output from the shading correction unit 12 shown in FIG. 1 by complementary colors and converts them into CMY signals.

The histogram creating section 22 is the signal converting section 2
Based on the input image data composed of CMY signals input from 1, a density histogram is created as shown in FIGS. 3 and 4. Here, the number of pixels in each density section is represented as a frequency value. FIG. 3 shows an example of a histogram of a text original, and FIG. 4 shows an example of a histogram of a photo original.

When creating the density histogram, the histogram creating section 22 determines that, in the input image data, if there are more pixels (number of pixels) than the predetermined threshold value in the low density area, the low density area is generated. All the pixels are determined to be the background area, which is the background area of the document, and are removed from the density histogram.
Here, as shown by the dotted line in FIG. 3, when the frequency of the low density region exceeds a predetermined threshold value, the density segment and its adjacent density segment are removed as the base region. In FIG. 3, even if the pixels of the density section removed as the background area are not more than the threshold value, the pixel is not judged to be the background area as shown in FIG. , Are not removed from the density histogram.

The first maximum frequency density classification extraction unit (hereinafter, referred to as the first
An abbreviation "extracting unit" 23 is for obtaining a first ratio which is a first parameter. The first extraction unit 23 selects the maximum frequency density category based on the density histogram created by the histogram creation unit 22, and sets the frequency value of the selected maximum frequency density class to the “first maximum frequency value”. . Then, the first maximum power value is divided by the total number of pixels of the page image to obtain the ratio of the first maximum power value to the total number of pixels (first maximum power value / total pixel count). This ratio is the first ratio (first parameter). The total number of pixels of the page image is the total frequency forming the density histogram of the page image. Therefore, if the page image has a background region, the number of pixels of the background region is removed from the total number of pixels. ing.

The second maximum frequency density category extraction unit (hereinafter referred to as the second frequency
An abbreviation "extraction unit" 24 is for obtaining a second ratio which is a second parameter. The second extracting unit 24 sets the frequency value of the density category having the second highest frequency value as the “second maximum frequency value” based on the density histogram created by the histogram creating unit 22. At this time, the first extraction unit 23
Concentration division of the first maximum frequency value extracted in step 1 and the concentration division adjacent to this concentration division (shaded areas in FIGS. 3 and 4)
Is excluded. Next, the second extraction unit 24 calculates the difference between the first maximum power value and the second maximum power value, divides the total number of pixels of the page image by the calculated difference, and the ratio of the total number of pixels to the difference. (Total number of pixels / (first maximum power value-second maximum power value)) is calculated.
This ratio is the second ratio.

In the storage unit 25, the first ratio and the degree of character x ₁ obtained in advance by using a character original, a photographic original, and a character photographic original composed of characters and photographs as image samples.
And the relationship between the second ratio and x ₂ representing the degree of photograph are stored in the form of a conversion table representing a function or a relationship. FIG. 5A shows the relationship between the first ratio stored in the storage unit 25 and the character degree x ₁ . Further, FIG. 5B shows the relationship between the second ratio and the photograph degree x ₂ . As can be seen from FIG. 5A, the first ratio increases as the character degree x ₁ increases. In addition, FIG.
As can be seen, the second ratio increases as the photographic degree x ₂ increases.

The recognizing unit 26 uses the first ratio obtained by the first extracting unit 23 and the second ratio obtained by the second extracting unit 24 as a function (or a function stored in the storage unit 25). The conversion table) is used to determine the character level x ₁ and the photo level x ₂ of the page drawing. In this way, the determined character degree x ₁ and photograph degree x ₂ are output from the recognition unit 26 to each processing unit in the subsequent stage.

In the flow chart of FIG.
Character degree x in the separate recognition unit 20 ₁And photo degree x₂
To determine the document type and show the processing procedure for recognizing the document type.
You

In S1, the color image input device 1 shown in FIG. 1 reads a page image as an RGB signal by prescanning. For the input image data of the read page image, S2
Then, the signal conversion unit 21 in the document type recognition unit 20
Convert RGB signals to CMY signals. In S3, the histogram creation unit 22 represents each pixel of the input image data as a density histogram. Based on this density histogram, in S4, the first extraction unit 23 selects the density category of the first maximum frequency value and obtains the first ratio. Furthermore, in S5,
The second extraction unit 24 selects the density category of the second maximum power value,
Find the second ratio.

In S6, the recognizing unit 26 uses the function stored in the storage unit 25 or the conversion table thereof, based on the first ratio and the second ratio obtained in S4 and S5.
The character degree x ₁ and the photograph degree x _{2 of the} input image data are respectively determined (estimated). This recognition result is output, and the recognition of the document type by the document type recognition unit 20 ends.

Next, a color correction method performed by the color correction section 40 and a black generation undercolor removal method performed by the black generation undercolor removal section 50 based on the recognition result performed by the document type recognition section 20. ,explain.

First, the processing method in the color correction section 40 will be described. FIG. 7 is a diagram showing the configuration of the color correction unit 40 in FIG. As shown in FIG. 7, the color correction unit 40 includes a character manuscript table access processing unit (hereinafter abbreviated as character manuscript processing unit) 41 and a photo manuscript table access processing unit (hereinafter abbreviated as photo manuscript processing unit) 41. 42,
Signal processing means 43.

The character manuscript processing means 41 and the photo manuscript processing means 42 use the CMY signals of the input image data as address data, and color correction values C ₁ , M ₁ , Y ₁ , and C ₂ of the character manuscript and the photo manuscript, respectively. , M ₂ and Y ₂ are read from LUTs (LookUp Tables) provided in the respective units, and output.

The signal processing means 43 is a character manuscript processing means 4.
1 and the CMY signals are corrected based on the color correction values read by the photo original processing unit 42. That is, the signal processing unit 43 weights the color correction value by using the character degree x ₁ and the photograph degree x ₂ recognized by the document type recognition unit 20 shown in FIG. 1 as weighting factors when correcting the CMY signal. To do.

Specifically, the character original process of the color correction unit 40
The means 41 uses the input image data Cin / Min / Y.
The color correction value for the in signal is as shown in Table 1,
C₁, M ₁, Y₁To decide. Further, the document type recognition unit 2
At 0, the character degree is x₁Is recognized. Well
Further, the photo original processing means 42 sets the color correction value to C₂, M₂, Y
₂Then, the document type recognition unit 20 determines that the photograph degree is x.
₂Is recognized. In this case, the color correction unit 40
The Cout / Mout / Yout signals output from
Is corrected by.

Cout = (x ₁ C ₁ + x ₂ C ₂ ) / (x ₁ + x ₂ ) Mout = (x ₁ M ₁ + x ₂ M ₂ ) / (x ₁ + x ₂ ) Yout = (x ₁ Y ₁ + x _2). Y ₂ ) / (x ₁ + x ₂ )

[0087]

[Table 1]

In the flow chart of FIG.
The procedure of the color correction process at 0 is shown.

Cin / Min / Yin of input image data
When a signal is input, in S11, the text original shown in FIG.
The processing means 41 addresses the Cin / Min / Yin signals.
Input the data by reading the value from the LUT
Sentences corresponding to Cin / Min / Yin signals of image data
Color correction value C for text originals₁, M₁, Y₁Is acquired (S1
2). Similarly, in S13, the photographic document processing means 42
L as Cin / Min / Yin signals as address data
By reading the value from U, the Cin.
Color correction for photo originals corresponding to Min / Yin signals
Value C₂, M₂, Y ₂Is acquired (S14).

As described above, the color correction value C ₁ of the text original is
When M ₁ , Y ₁ and the color correction values C ₂ , M ₂ , Y _{2 of} the photo original are read from the LUT, the signal processing means 43 causes the original type recognition unit 20 shown in FIG. As described above, the color correction value is corrected based on the recognized character degree x ₁ and photograph degree x ₂ . Corrected Cou
The t · Mout · Yout signal is output, and the color correction process ends.

Next, the processing in the black generation undercolor removal processing section 50 will be described. In the black generation undercolor removal processing unit 50, the character degree x ₁ (0 ≦ x ₁ ≦ recognized by the document type recognition unit 20 is recognized.
Black generation and under color removal processing is performed with 1) as a black generation rate (UCR (Under Color Rmoval) rate). here,
When the input image data is C, M, Y, and the output data is C ', M', Y ', the black generation and undercolor removal processing corrects C, M, Y by the following equation.

C '= C-x ₁ min (C, M, Y) M' = M-x ₁ min (C, M, Y) Y '= Y-x ₁ min (C, M, Y) K = x ₁ min (C, M, Y) As described above, in the image processing apparatus according to the present embodiment, the color correction unit 40 performs the color correction processing based on the character degree and the photograph degree of the input image data. The black generation / undercolor removal processing unit 50 is configured to perform the black generation / undercolor removal processing of the input image data by reflecting the character degree.

Although the image processing apparatus according to the present embodiment performs the processing in the spatial filter processing unit 14 for each pixel based on the result of separation by the area separation processing unit 30, the document type recognition unit Character degree recognized at 20 x ₁
And the degree of photography x ₂ may be used.

The processing in the spatial filter processing section 14 in this case may be performed by the following method.

The spatial filter coefficient f is determined based on the character degree x ₁ and the photograph degree x ₂ recognized by the document type recognition unit 20. The spatial filter coefficient f is, for example,
It is calculated by the following formula.

[0096]

[Equation 1]

Regarding the spatial filter coefficient f, the edge emphasis degree becomes stronger as the character degree x ₁ becomes larger, and the smoothness degree becomes stronger as the photograph degree x ₂ becomes larger. Spatial filter coefficient f and black generation undercolor removal unit 50
By the convolution operation with the CMYK signal input from
Spatial filtering is performed.

As described above, by processing the input image data by using the character degree and the photograph degree of the page drawing, it is possible to deal with the image processing of the input image data including a plurality of image types.

Further, in the present embodiment, since the density histogram is created for the input image data to determine the character degree and the photograph degree, it is not necessary to perform image processing using a complicated function.

Note that the above-mentioned processing for each area is performed by CP
It is controlled by a control means such as U (Central Processing Unit).

A program for causing an image forming apparatus or a computer system to execute the image processing method of the color image processing apparatus as described above is recorded in a computer-readable recording medium (not shown). The program reading device provided in the image forming apparatus or the computer system reads and executes the program recorded in the recording medium.

The computer system displays an image input device such as a flat head scanner, a film scanner or a digital camera, a computer for executing a processing such as an image processing method by loading a predetermined program, and a processing result of the computer. An image display device such as a CRT display or a liquid crystal display, and an image output device such as a printer that outputs a processing result of a computer are provided. Further, it is provided with a communication means such as a modem for communicating via the network.

On the other hand, the recording medium is a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a floppy disk or a hard disk, a CD-ROM, an MO or an M.
Disk systems of optical disks such as D and DVD, card systems such as IC cards and optical cards including memory cards, mask ROM, EPROM, EEROM, flash RO
It may be a medium including a semiconductor memory such as M that holds the program fixedly.

The program stored in the recording medium is
It may be configured to be accessed and executed by a computer, or may be configured to read a program stored in a recording medium, download the read program to a program recording area of the computer, and execute the program. May be.

The image processing method according to this embodiment can also be executed by connecting to a communication network including the Internet. Therefore, it may be a recording medium that fluidly carries the program such as downloading the program from the communication network.

[Second Embodiment] The following will describe another embodiment of the present invention in reference to FIGS. 9 to 13. For convenience of explanation, members having the same functions as those shown in the drawings of the first embodiment will be designated by the same reference numerals, and the description thereof will be omitted.

FIG. 9 is a block diagram showing the color image processing device 60 in this embodiment. The configuration of the color image processing device 60 is different from the color image processing device 10 shown in FIG. 1 described in the first embodiment in that the document type recognition unit 20 is not provided. Therefore, the input tone correction unit 5 is provided.
3 and the image processing method in the area separation processing unit (area separation means) 80 are also different. Therefore, these members will be described below with reference numerals different from those in FIG.

In addition to the function of the input gradation correction unit 13 in FIG. 1, the input gradation correction unit 53 also has a function of complementing the RGB reflection signals output from the shading correction unit 12 with complementary colors to make CMY signals. I have it.

The CMY signal subjected to the image adjustment processing by the input tone correction section 53 is input to the area separation processing section 80. The area separation processing unit 80 extracts a local block from the input image data represented by the CMY signal, and recognizes the degree of character, the degree of halftone dot, and the degree of photograph of photographic paper for each pixel of interest. Here, the local block includes a target pixel and a plurality of peripheral pixels in the vicinity thereof (for example, 5 pixels ×
5 pixels, see FIG. 13). The recognition result is the color correction unit 4
0, black generation undercolor removal unit 50, spatial filter processing unit 14,
It is output to the gradation reproduction processing unit 16. Further, the CMY signal input from the input tone correction unit 53 is output to the color correction unit 40 in the subsequent stage as it is.

Next, a method of recognizing an image area in the area separation processing section 80 will be described in detail with reference to FIGS. 10 to 13.

FIG. 10 shows the area separation processing section 8 in FIG.
It is a figure which shows the structure of 0. The area separation processing unit 80 includes a background determination unit 81, a parameter calculation unit 82, and a storage unit 83.
And a recognition unit 84. The parameter calculation unit 82 includes a first characteristic parameter calculation unit (hereinafter abbreviated as a first calculation unit) 85 and a second characteristic parameter calculation unit (hereinafter abbreviated as a second calculation unit) 86. There is.

The background determining section 81 determines the background area of the input image data and the photographic paper photograph area based on the CMY signal which is the input image data of the local block. Specifically, the signal levels of the C signal, M signal, and Y signal of the input CMY signal are DC, DM, and DY, and a preset threshold value is TH. At this time, DC
If <TH and DM <TH and DY <TH, it is determined that the pixel of interest is the base region.

By detecting the background area in this way, the background area and the photographic paper photograph area can be distinguished from each other, and the background area is prevented from being discriminated from the photographic paper photograph area.

The CMY signal whose background area has been determined by the background determining section 81 is sent to the parameter calculating section 82.

The first calculator 85 of the parameter calculator 82.
Is the signal of each pixel in the local block,
Find the maximum and minimum signal levels. Furthermore, by finding the difference between the maximum signal level and the minimum signal level,
This difference is set as the first characteristic parameter P ₀ .

The second calculating section 86 sequentially adds the absolute values of the difference values between two pixels which are continuous in the main scanning direction within the local block, and obtains the total sum of the difference values. Also in the sub-scanning direction within the local block, the absolute values of the difference values between two consecutive pixels are sequentially added, and the total sum of the difference values is obtained.
The sum of the difference values in the main scanning direction thus obtained is compared with the sum of the difference values in the sub-scanning direction, and the smaller value is set as the second characteristic parameter P ₁ .

The storage unit 33 stores the distribution state of the first characteristic parameter P ₀ and the second characteristic parameter P ₁ and the character area, the halftone dot area, and the photographic paper photograph area, which are obtained in advance by using a plurality of image samples. A function that represents the function and a conversion table that represents these relationships are stored.

The region-likeness (recognition degree) of each pixel of the input image data is recognized based on the characteristic parameters P ₀ and P ₁ of the image sample stored in the storage unit 83. Below, it demonstrates using FIG. 11 and FIG. Figure 11
Is the first characteristic parameter P ₀ and the second characteristic parameter P ₁
FIG. 12 is a diagram showing the distribution of each region when the region separation processing is performed by using FIG. 12, and FIG.
It is an A'sectional view.

In the character area, the difference between the maximum signal level and the minimum signal level of the pixels in the local block is large. In addition, the total sum of the difference values between two pixels that are continuous in the main scanning direction or the sub scanning direction within the local block is small. Therefore, the character area is, as shown in FIG.
The first characteristic parameter P ₀ corresponds to a large area and the second characteristic parameter P ₁ corresponds to a small area.

In the halftone dot area, the difference between the maximum signal level and the minimum signal level of the pixels in the local block is large. Further, since the image signal in the halftone dot area fluctuates in a short cycle, the total sum of the difference values between two consecutive pixels in the local block in the main scanning direction or the sub scanning direction also becomes large. Therefore, in the halftone dot area, as shown in FIG. 11, the first feature parameter P ₀ and the second feature parameter P ₁
Both correspond to a large area.

Further, in the photographic paper photograph area, since the pixels in the local block have gradation, the difference between the maximum signal level and the minimum signal level of the pixel becomes small. In addition, the total sum of the difference values between two pixels that are continuous in the main scanning direction or the sub scanning direction within the local block is also small. That is, the photographic paper photograph area corresponds to an area in which both the first characteristic parameter P ₀ and the second characteristic parameter P ₁ are small, as shown in FIG.

When the recognizing unit 84 receives the characteristic parameters P ₀ and P ₁ obtained from the CMY signal of the input image data from the parameter calculating unit 82, the function (or conversion table) stored in the storage unit 83 is obtained. By referring to, the character degree of the pixel of the input image data,
Recognize the degree of halftone dots and the degree of photographic printing on photographs. That is,
As shown in FIG. 12, if the first characteristic parameter P ₀ and the second characteristic parameter P ₁ are set, the recognition degree of the character degree of the pixel of the input image data, the halftone dot degree, and the photographic paper photograph degree are unique. To decide. The recognition result is output to each of the processing units described above.

The color correction section 40 will be described in the first embodiment.
As is clear, the signal processing means 43 shown in FIG.
The CMY signal of the image data is used as address data
Area, photographic paper photograph area, halftone dot area color correction value
(C₁, M₁, Y₁), (C₂, M ₂, Y₂),
(C₃, M₃, Y₃) Is determined.

The signal processing means 43 corrects the CMY signal based on these color correction values. Further, the signal processing unit 43 inputs the character degree, the photographic paper photograph degree, and the halftone dot degree, which are recognized by the area separation processing unit 80 shown in FIG. 10, as weighting factors when correcting the CMY signals.

In the area separation processing section 80, the character degree is a
_1, photographic paper photograph degree is to be recognized as a _2, a halftone degree a _3. In this case, the CMY signal output from the color correction unit 40 is corrected by the following formula.

C = (a ₁ C ₁ + a ₂ C ₂ + a ₃ C ₃ ) / (a ₁ + a ₂ + a ₃ ) M = (a ₁ M ₁ + a ₂ M ₂ + a ₃ M ₃ ) / (a ₁ + a _2). + A ₃ ) Y = (a ₁ Y ₁ + a ₂ Y ₂ + a ₃ Y ₃ ) / (a ₁ + a ₂ + a ₃ ) The black generation undercolor removal unit 50 and the spatial filter processing unit 14 are also described in the first embodiment. As described above, similar processing is performed (in the spatial filter processing unit 14, the character degree is a
₁ , halftone dot a ₃ is used).

As described above, the image processing apparatus according to the present embodiment determines the character degree, the photographic paper photograph degree, and the halftone dot degree for each pixel. Further, since the degree is determined based on the signal level of each pixel and image processing is performed, it is not necessary to perform image processing using a complicated function.

The above processing may be performed based on the recognition degree obtained not for each pixel but for a block composed of a plurality of pixels.

As described in the first embodiment, the program for causing the image forming apparatus or the computer system to execute the image processing method of the color image processing apparatus as described above is a computer-readable recording (not shown). It is recorded on the medium. The program recorded on the recording medium is read and executed by a program reading device provided in the image forming apparatus or the computer system.

In the above description, the photograph means a photographic paper photograph.

[0131]

As described above, the image processing apparatus according to the present invention uses the plurality of parameters obtained from the input image data in the image processing apparatus for performing the image processing on the input image data. It is characterized in that it includes original document type recognition means for estimating a recognition degree indicating the degree of image type in image data, and image processing means for performing image processing based on the recognition degree.

As a result, the same processing is not performed on an intermediate image in which photographs (continuous tone images) and characters (line drawings) are mixed, regardless of the degree of mixing. The image processing is performed based on the ratio that is considered to occupy in the image, that is, the degree of character and the degree of photograph, and as a result, the optimum image processing can be performed on each page as a whole, improving the image quality You can

Further, according to this, since the recognition degree is estimated by combining a plurality of parameters without requiring complicated processing for determining the image type, the processing becomes very simple. Therefore, the processing speed is increased and the required processing capacity is reduced.

Further, in order to identify the image type, the edge amount and the changed pixel count amount which require complicated processing are not used, and the recognition degree is estimated by combining a plurality of parameters. Very easy to process. Therefore, the processing speed is increased and the required processing capacity is reduced.

In order to solve the above problems, the image processing apparatus according to the present invention is further characterized in that the degree of recognition is a degree of recognition of the input image data of the entire page area.

With the above arrangement, the image processing of the entire page area is further performed based on the degree of recognition of the entire page area. Therefore, the optimum image processing can be performed on the entire page and the local area It becomes possible to perform the processing more quickly than the processing of.

As described above, the image processing apparatus according to the present invention is characterized in that the image processing means includes color correction means for correcting the color space of the input image data based on the degree of recognition. I am trying.

With the above arrangement, the character degree, the photograph degree and the halftone degree of the input image data are further reflected in the color correction, so that the optimum color correction processing can be performed on the input image data. Thereby, the image quality can be improved.

As described above, in the image processing apparatus according to the present invention, the image processing means generates image data representing black from the input image data based on the degree of recognition, and the input image It is characterized by including black generation undercolor removal means for performing image processing on the input image data by subtracting the generated image data representing black from the data.

With the above configuration, since the black generation undercolor removal processing is further performed by reflecting the character degree, photograph degree and halftone dot degree of the input image data, the black generation undercolor removal processing most suitable for the input image data is performed. Can be applied. As a result, the reproducibility of thin lines such as characters can be improved.

As described above, the image processing apparatus according to the present invention is characterized by further comprising display means for displaying the input image data which has been subjected to image processing based on the degree of recognition.

As a result, the image display device further displays an image which has undergone image processing while reflecting the character degree, the photograph degree and the halftone degree of the input image data. This allows the user to adjust the image processing so as to obtain the desired image. Further, even when the image processing is erroneous, it can be easily corrected.

The image forming apparatus according to the present invention is characterized by including the above-mentioned image processing apparatus as described above.

With the above-described structure, the image forming apparatus, such as a digital color copying machine or a facsimile machine, reflects the character degree, the photograph degree, and the halftone degree of the input image data to perform image processing and image formation. Can be improved.

In order to solve the above problems, the image processing method according to the present invention uses a plurality of parameters obtained from input image data to estimate the degree of recognition indicating the degree of image type in the input image data. The image processing is performed based on the degree of recognition.

In the above method, the degree of recognition of the image type of the input image data is estimated from a plurality of parameters of the input image data.

Since it is difficult to estimate the degree of recognition of the image type with sufficient accuracy using only one parameter,
By combining a plurality of parameters, it is possible to accurately estimate the degree of recognition of the image type.

As a result, optimum image processing can be performed based on the degree of recognition even for input image data in which character images, photographic images and halftone images are present in various proportions according to the degree of recognition. it can. Therefore, it is possible to perform the optimum image processing even on an image having an intermediate characteristic between the image types. Thereby, the image quality can be improved.

Further, since the degree of recognition is estimated by combining a plurality of parameters without requiring complicated processing for determining the image type, the processing becomes very simple. Therefore, the processing speed is increased and the required processing capacity is reduced.

Further, in order to identify the image type, the edge amount and the changed pixel count amount which require complicated processing are not used, and the recognition degree is estimated by combining a plurality of parameters. Very easy to process. Therefore, the processing speed is increased and the required processing capacity is reduced.

As described above, the image processing method according to the present invention is further characterized in that the degree of recognition is the degree of recognition of the input image data of the entire page area.

With the above configuration, the image processing of the entire page area is performed based on the degree of recognition of the entire page area. Therefore, the optimum image processing for the entire page is performed as compared with the processing for each local area. It will be possible to do it quickly.

As described above, the program according to the present invention can be characterized by causing a computer to execute the above image processing method.

With the above program, the computer can read and execute the image processing method for performing the image processing by reflecting the character degree, the photograph degree, and the halftone degree of the input image data. It can be general-purpose.

The recording medium according to the present invention is as described above.
The above program may be stored so that it can be read by a computer.

With the above arrangement, the program of the image processing method for performing the image processing by reflecting the character degree, the photograph degree, and the halftone degree of the input image data can be easily supplied to the computer through the recording medium. .

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a digital color image forming apparatus to which an image processing apparatus of the present invention is applied.

FIG. 2 is a block diagram showing a configuration of a document type recognition unit shown in FIG.

FIG. 3 is a diagram showing an example of a histogram of a character original in a method of estimating the degree of recognition of an original type using a histogram.

FIG. 4 is a diagram showing an example of a histogram of a photo original in a method of estimating the degree of recognition of an original type using a histogram.

5A is a diagram showing a relationship between a first ratio and a character degree x _1, and FIG. 5B is a diagram showing a relationship between a second ratio and a photograph degree x ₂ .

FIG. 6 is a flowchart showing a method of estimating the degree of recognition of a document type.

FIG. 7 is a diagram showing a configuration of a color correction unit shown in FIG.

FIG. 8 is a flowchart showing a processing method when the image processing method of the present invention is applied to color correction.

FIG. 9 is a block diagram showing another configuration of a digital color image forming apparatus to which the image processing apparatus of the present invention is applied.

10 is a diagram showing a configuration of a region separation processing unit shown in FIG.

FIG. 11 is a diagram showing a distribution of each region when the region separation processing is performed using the first feature parameter and the second feature parameter.

12 is a cross-sectional view taken along the line A-A ′ shown in FIG.

FIG. 13 is a diagram illustrating local blocks used when extracting a characteristic parameter of input image data in the area separation processing unit illustrated in FIG. 9.

[Explanation of symbols]

1 Color image input device 2 Color image output device 3 Operation panel 10 Color image processor 11 A / D converter 12 Shading correction section 13 Input tone correction section 14 Spatial filter processing unit 15 Output gradation correction unit 16 gradation reproduction processing section 20 Document Type Recognition Unit (Document Type Recognition Unit) 30 area separation processing unit (area separation means) 40 color correction unit (color correction means) 50 Black generation undercolor removal unit (black generation undercolor removal means) 53 Input tone correction unit 60 color image processor 80 area separation processing unit (area separation means)

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5B057 AA11 CA01 CA08 CB01 CB08                       CE17 CH01 CH11 DB06 DB09                 5C077 MP06 MP08 PP27 PP28 PP32                       PP33 PP37 PP38 PQ12 PQ22                 5C079 HB01 HB03 HB12 LA06 LA21                       MA01 MA11

Claims (10)

[Claims] 1. An image processing apparatus for performing image processing on input image data, using a plurality of parameters obtained from the input image data, to estimate a recognition degree indicating a degree of an image type in the input image data. An image processing apparatus comprising: a document type recognition unit; and an image processing unit that performs image processing based on the degree of recognition. 2. The image processing apparatus according to claim 1, wherein the degree of recognition is a degree of recognition of the input image data of the entire page area. 3. The image processing apparatus according to claim 1, wherein the image processing means includes color correction means for correcting the color space of the input image data based on the degree of recognition. 4. The image processing means generates image data representing black from the input image data based on the degree of recognition, and subtracts image data representing black generated from the input image data, The image processing apparatus according to claim 1, further comprising a black generation undercolor removal unit that performs image processing on the input image data. 5. The image processing apparatus according to claim 1, further comprising display means for displaying the input image data which has been subjected to image processing based on the degree of recognition. 6. An image forming apparatus comprising the image processing apparatus according to claim 1. Description: 7. A plurality of parameters obtained from the input image data are used to estimate the degree of recognition indicating the degree of image type in the input image data, and image processing is performed based on the degree of recognition. Image processing method. 8. The image processing method according to claim 7, wherein the degree of recognition is a degree of recognition of the input image data of the entire page area. 9. A program for causing a computer to execute the image processing method according to claim 7 or 8. 10. A recording medium in which the program according to claim 9 is stored so that it can be read by a computer.