JP2003051946A - Image processing method, image processing apparatus, imaging device provided with the image processing apparatus, image processing program, and computer- readable recording medium with the program recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing method that can prevent deterioration of image quality, regardless of a type of an original image through accurate detection of a background level, and to provide an image processing apparatus, an imaging device provided with the image processing apparatus, an image processing program, and a computer-readable recording medium for recording the program. SOLUTION: The image processing method using a digital color copying machine generates histograms from input image data (input signal) obtained by preliminary scanning by a CCD from a read original, detects a background level, and performs the image processing according to the background level, and starts searching from a dark side of the histogram, detects a level first which exceeds a prescribed threshold as a background object level, compares a frequency of occurrence of the background object levels with the frequency of occurrence of levels adjacent to the level toward the bright side, and decides a level having higher frequency of occurrence to be the background level.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method in a color or monochrome digital color copying machine or the like, and in particular, it accurately detects the background level to reduce the image quality regardless of the type of the original image. The present invention relates to an image processing method that can be prevented, an image processing apparatus, an image forming apparatus including the image processing apparatus, an image processing program, and a computer-readable recording medium that records the same.

[0002]

2. Description of the Related Art As an image forming apparatus such as a copying machine using an electrophotographic process, a digital type is widely used in addition to the conventional analog type. Under such circumstances, there is known an automatic exposure function which is a general function in analog copying machines, that is, a method of obtaining the optimum image quality by detecting the original density sensor and changing the brightness of the exposure lamp.

As a method of realizing an automatic exposure function in a digital copying machine, Japanese Patent No. 2948280 discloses a method of determining the image type of a color image and removing the background when it is determined that a character image is included. Image processing method is described. Further, as the image type determination, there is a method of determining whether the number of data existing in a predetermined range from the maximum document reflection brightness is a predetermined% or not.

Further, in an image processing apparatus used in a color copying machine or the like, it may be considered to read a document whose background is colored, such as newspaper, for image processing.

However, in the conventional image processing apparatus,
There is a problem in that the image quality is deteriorated by reading the background color.

In order to solve such a problem, for example, in Japanese Unexamined Patent Publication No. 4-313774, a histogram is created based on input data of a read original and the frequencies are sequentially checked from the high density side. First, an image processing apparatus has been proposed which uses a density exceeding a threshold as a base.

[0007]

However, the image processing method by the conventional image processing apparatus as described in the above publication has the following problems.

In the configuration described in the above-mentioned Japanese Patent No. 2948280, since the maximum reflection density is used as the background, for example, when a document having a background or show-through has a white edge or a white portion, a white edge or white is present. It is determined that the removed portion has the background density, and the background or show-through that should not be printed remains.

Further, since the maximum reflection density is used as the base, it is not possible to remove the back slip having a density higher than the maximum reflection density of the original.

Further, in the case of a cut and pasted original such as a newspaper original on a blank sheet, the newspaper background density cannot be detected as the background.

The image processing apparatus disclosed in Japanese Patent Laid-Open No. 4-313774 is intended to improve the legibility of an original document having a background fog or backslip and to obtain an optimum reproduced image. If a proper background level is not detected, it is impossible to prevent image quality deterioration.

In particular, it is difficult to reproduce the highlight portion in a document in which characters and a grayscale image having gradation are mixed, and in the case of a cut and pasted document or a document having a plurality of peak-like peak density, Accurate detection of the background level becomes difficult. Further, in a document containing a color image, there is a color cast, and depending on the processing, a color tone change occurs in the reproduced image.

As described above, with the image processing method by the image processing apparatus of the above publication, it is difficult to detect an accurate background level for various original images and prevent image quality deterioration.

The present invention has been made in view of the above problems, and an object of the present invention is to perform image processing capable of preventing deterioration of image quality regardless of types of original images by more accurate detection of a background level. A method, an image processing apparatus, an image forming apparatus including the image processing apparatus, an image processing program, and a computer-readable recording medium recording the program.

[0015]

In order to solve the above-mentioned problems, an image processing method of the present invention is to create a histogram from an input signal obtained by reading an original to detect a background level, and to detect the background level. In an image processing method for performing image processing according to a level, a search is started from the dark side of the histogram, a level that first exceeds a predetermined threshold is detected as a background candidate level, and the frequency and the bright side of the background candidate level are detected. It is characterized in that the level having a higher frequency is determined as the base level by comparing the frequency with the level adjacent to the.

According to the above arrangement, it is possible to detect the background level more accurately, and it is possible to form an excellent image.

That is, the level that first exceeds the predetermined threshold is set as the background candidate level, and the frequency of the background candidate level is compared with the frequencies of the levels adjacent to the bright side thereof, and the level with the higher frequency is used as the background. By detecting the level, it is possible to detect the background level more accurately. Thus, even with a simple image processing method that does not perform peak search, it is possible to detect a level closer to the peak.

Accordingly, the readability of an original document with a background fog or back-slip is improved, and the background level can be detected even for a cut and pasted original document obtained by scrapping a blank of a newspaper or the like.

Further, by controlling the predetermined threshold value and the like, it is possible to accurately detect the background level not only for cut and pasted originals but also for various original images such as originals having a plurality of peak-like peak densities. .

In order to solve the above-mentioned problems, the image processing method of the present invention creates a histogram from an input signal obtained by reading a document, detects a background level, and performs image processing according to the background level. In the image processing method, the search is started from the dark side of the histogram, a level that first exceeds a predetermined threshold is detected as a background candidate level, and a level adjacent to the bright side with respect to the frequency of the background candidate level is detected. If the frequency ratio is less than a predetermined value set in advance, the base candidate level is determined as the base level, and if the frequency ratio is the predetermined value or more, the bright side The feature is that the level adjacent to is determined as the base level.

According to the above arrangement, it is possible to detect the background level more accurately, and it is possible to form an excellent image.

That is, by detecting the background level in accordance with the frequency of the detection level and the frequency ratio of the frequencies of the levels adjacent to the light side thereof, a more accurate background level of the histogram frequency corresponding to variations and errors in the histogram frequency is detected. It becomes possible to detect.

With this arrangement, the background level closer to the peak can be detected with a simple structure in which the peak search is not performed.

In order to solve the above problems, the image processing method of the present invention creates a histogram from an input signal obtained by reading an original to detect a background level, and performs image processing according to the background level. In the image processing method to be performed, the search is started from the bright side of the histogram, a level that first exceeds a predetermined threshold is detected as a background candidate level, and the frequency of the background candidate level and the frequency of a level adjacent to the dark side are detected. Are compared with each other, and the level having the higher frequency is determined as the background level.

According to the above arrangement, it is possible to detect the background level more accurately, and it is possible to form an excellent image.

That is, the search is started from the bright side of the histogram, and the frequency of the background candidate level detected by the predetermined threshold is compared with the frequency of the level adjacent to the dark side, and the level with the higher frequency is compared. By determining as the background level, it becomes possible to detect the background level more accurately even for a document having a plurality of peak density values that seem to be the background.

With this arrangement, it is possible to detect the background level closer to the peak, even though the peak search is not performed.

When the frequency of the brightest level in the histogram is larger than the predetermined threshold value, the level is excluded from the background candidate level and the predetermined threshold value is set according to the frequency of the background candidate level. It is more preferable to change it.

As a result, when the frequency of the level on the brightest side of the histogram is larger than the predetermined threshold value, the level is excluded from the background candidate levels, and the background level considering the influence of the white level data amount is taken into consideration. It becomes possible to detect.
Therefore, in addition to a manuscript having peak density like multiple backgrounds, even a cut and pasted manuscript with scraps of newspapers cut out on a white paper, etc., without being affected by the white level data,
An accurate background level can be detected.

Further, it is more preferable to perform the same density conversion processing on each of the input signals of the respective colors according to the background level determined above.

As a result, in the background removal performed based on the input signals of the respective colors, when the background removal of only a specific color is performed on an original having a color cast or a tinted background, the change of the color is performed. There is a problem that
By performing the same density conversion processing on each of the GB and CMY colors, it is possible to obtain a reproduced image with no change in color.

Further, it is more preferable that the predetermined threshold value is determined by a ratio to the total frequency of the histogram.

As a result, the predetermined threshold value used for detecting the background level is determined as a ratio to the total histogram frequency, so that it is not necessary to set a plurality of threshold values according to the document size, and various types can be used with a simple structure. An appropriate threshold value can be set for the original document.

It is more preferable that the predetermined threshold value is changed according to the density level of each section of the histogram and the threshold value of each section is determined to detect the background level.

Thus, for example, by setting the threshold value 1 corresponding to the brightest side section 1 in the histogram to be larger than the threshold value 2 corresponding to other sections, the section 1 is erroneously detected as the background candidate level. Can be prevented. Therefore, it is possible to detect the background level more accurately, not only for a document having a plurality of peak densities like a background but also for a cut and pasted document.

In order to solve the above problems, the image processing method of the present invention creates a histogram from an input signal obtained by reading an original to detect a background level, and performs image processing according to the background level. In the image processing method, the histogram is searched in a predetermined direction, and the frequency of the density section on the search start side is compared with the first threshold to detect whether the document is backgroundless or to detect a background candidate later. If it is determined that the first step of determining whether to perform the step and the step of detecting the background candidate in the first step are performed, the frequency of the density classification is compared with the second threshold to determine the background candidate. In the second step of determining whether or not there is a background candidate, and in the second step, when it is determined that there is a candidate for the background, the frequency of the density classification in which the frequency is equal to or higher than the second threshold, and the density A third step of determining the presence or absence of a background by adding the division and the frequencies of adjacent density divisions in the search direction, and comparing the added value with a third threshold value. The feature is that base removal processing is performed based on the result of the third step.

Specifically, in the above image processing method, it is preferable that the search is performed from the low density side to the high density side of the histogram.

According to the above configuration, the case where there is no background is determined in the first step, and when it is not determined that there is no background in the first step, the background level is determined in the second step to select the background candidate. To detect. If the background candidate cannot be detected, it is determined that there is no background, and if the background candidate can be detected, the presence or absence of the background is further determined in the third step.

As a result, the readability of a photographic printing paper, a print photographic original, an original with a background fog or a backslip can be improved even with a simple structure that does not require the processing for determining the type of the original. An optimum reproduced image can be obtained. That is, it is possible to detect the background level more accurately even in the case where there are a plurality of originals having peak density that is similar to the background, that is, white data exists.

Further, by using the same background removal method for a plurality of color components, it is possible to obtain a reproduced image in which no color change occurs in the color image.

In the above image processing method, the histogram is
It is preferably created based on the luminance information obtained by the ratio calculation from the input signal.

As a result, the luminance information (Gray) calculated based on the input signal such as the CMY signal or the RGB signal is obtained.
CMY or RGB by creating a histogram from the
It is not necessary to create and refer to the histograms for all the colors, and the base level can be accurately detected with a simpler configuration.

In the above image processing method, it is preferable that the luminance information is converted such that the ratios of a plurality of color components in the input signal are equal.

According to the above configuration, it is possible to grasp the characteristics of the original such as reddish and to detect a more colored background. Therefore, it is possible to detect the background more appropriately without the problem that the detection accuracy varies depending on the tint of the background.

In the above image processing method, in the third step, when the added value is equal to or more than the third threshold value, the density of the density section whose frequency is equal to or more than the second threshold value, or the direction of searching with the density section. It is preferable to determine the density of the density section adjacent to as the background level.

According to the above arrangement, it is possible to detect a more proper background level in response to variations in histogram frequency and errors. By determining the background level according to the frequency relationship with the adjacent density section, it is possible to detect the level close to the peak while having a simple configuration in which the peak search is not performed.

In the above image processing method, in the first step, it is preferable to determine that the document has no background when the frequency of the density section on the search start side is equal to or higher than the first threshold value.

According to the above-mentioned structure, an accurate background level can be detected with a simple structure even when, for example, a photographic document having a white paper portion and / or a highlight portion at the edge, that is, white data exists. can do.

The image processing method described above uses the first threshold value and the second threshold value.
It is preferable that the threshold value and the third threshold value are independent of each other, and each of them can be arbitrarily changed.

According to the above configuration, it is possible to change one of the first threshold value, the second threshold value, and the third threshold value depending on the purpose.

For example, the first threshold makes it possible to set the amount of white data for the cut and pasted original to a desired setting.

The second threshold makes it possible to arbitrarily set the selection of the background level candidates. This makes it possible to remove show-through.

By the third threshold value, it is possible to set the presence or absence of the background detection for the photographic paper and the print photographic original to a desired state.

Therefore, by controlling the threshold value for determining the background as described above, the background level can be accurately detected not only for the cut and pasted original, but also for an original having a plurality of peak densities like the background.

In order to solve the above-mentioned problems, the image processing apparatus of the present invention is characterized by using the above-described image processing method to determine the background level of the read original and perform background removal processing.

According to the above arrangement, the halftone data, that is, the white data in the original including the print photograph and the photographic paper photograph is not detected as the background. Therefore, it is possible to prevent necessary image information from being lost. Further, it is possible to perform a desired background removal process without causing a change in color.

In order to solve the above-mentioned problems, an image forming apparatus of the present invention includes the above-mentioned image processing apparatus, an image input apparatus for reading a document and inputting it as image data to the image processing apparatus, and the above image processing. An image output device for forming an image on a recording medium based on output image data output by the device.

According to the above arrangement, it is possible to prevent the necessary image information from being lost and to prevent the color image from being lost by not detecting the background of the original including the halftone dot photograph and the photographic paper photograph. A desired underlayer removal process can be performed without causing a change.

Therefore, without judging the type of the original,
It is possible to prevent deterioration of image quality. As a result, a high quality image can be output in the image forming apparatus.

An image processing program of the present invention is characterized by causing a computer to execute the above-described image processing method in order to solve the above problems.

According to the above arrangement, the computer can read and execute the image processing program, so that the same operation and effect as those of the image processing method described above can be obtained. Further, the image processing method described above can be general-purpose.

The computer-readable recording medium of the present invention is characterized by recording the above-mentioned image processing program in order to solve the above problems.

According to the above arrangement, the same operation and effect as those of the image processing method described above can be obtained. Further, the image processing method described above can be general-purpose.

[0064]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [First Embodiment] The following will describe one embodiment of the image processing method of the present invention with reference to FIGS.

As shown in FIG. 2, the image processing apparatus according to the present embodiment is provided in a tandem type digital color copying machine 10 which is becoming the mainstream with the recent downsizing of internal devices and assembly of devices. This digital color copying machine 10
4 color image forming stations 200Bk (black), 210C (cyan), 220M (magenta), 2
It is equipped with 30Y (yellow).

Image forming station 200B for the above four colors
The k, 210C, 220M, and 230Y differ only in the toner color inside the apparatus, and the other configurations are all the same.

In the following, the Bk (black) image forming station 200Bk will be described as an example, but the other image forming stations 210C, 220M and 230Y.
Is also the same. For example, the developing device 204Bk
The rotational speeds of 214C, 224M, and 234Y are also set to the same rotational speed.

The drum-shaped photosensitive member 201Bk of the image forming station 200Bk rotates in the direction of arrow D shown in FIG.
k, the writing optical system 203Bk, the developing device 204Bk,
A cleaning device 205Bk is arranged.

In the digital color copying machine 10,
Laser light is emitted from the writing optical system 203Bk to the surface of the photoconductor 201Bk between the charger 202Bk and the developing device 204Bk, and a Bk (black) electrostatic latent image is formed on the photoconductor 201Bk. The image forming station 21 having the same structure as the image forming station 200Bk.
0C, 220M, and 230Y are provided along the sheet transport belt 250 that is a sheet transport unit.

The sheet conveying belt 250 is in contact with the photoconductor 201Bk between the developing device 204Bk and the cleaning device 205Bk of the image forming station 200Bk. Further, each photoconductor 201B of the paper transport belt 250
A transfer roller 206Bk for applying a transfer bias is arranged on the back surface on the k side.

In the digital color copying machine 10 having the above structure, the image forming operation is performed as follows.

First, in the image forming station 200Bk, the photoconductor 201Bk is charged by the charger 202Bk. Then, an electrostatic latent image corresponding to a black (Bk) image is formed by the laser light emitted from the writing optical system 203Bk.

Next, the developing device 204Bk develops the electrostatic latent image to form a toner image. Developing device 204B
k is developed with black (Bk) toner, and the toner image formed on the photoconductor 201Bk is a sheet (recording medium).
And the image is formed.

When a color image is formed, the toner images of the other three colors formed in the same manner as described above are transferred so as to be superimposed on the paper to form a color image.

The sheet on which the image is formed is transferred from the sheet feeding section 240 by the sheet conveying belt 250 to the photosensitive member 2 described above.
At the timing of toner image formation on 01Bk,
It is conveyed in the direction of arrow E shown in FIG.

The paper held on the paper carrying belt 250 is carried and is brought into contact with the photosensitive member 201Bk (hereinafter, referred to as a contact position).
(Referred to as a transfer unit), the toner images of the respective colors described above are transferred. The toner image on the photoconductor 201Bk is transferred onto the photoconductor 20 with the transfer bias applied to the transfer roller 206Bk.
It is transferred onto the paper by the electric field generated from the potential difference from 1 Bk. Then, the sheet on which the toner images have been transferred so as to overlap each other through the transfer unit is conveyed to the fixing device 260,
The toner image is fixed and discharged to a paper discharge unit (not shown).

Further, the photosensitive member 201 is not transferred at the transfer portion.
The residual toner remaining in Bk is cleaned by the cleaning device 205B.
Recovered at k.

The four color image forming stations 200Bk, 210C, 220M and 230Y of this embodiment are
Bk / C / M from the upstream side to the downstream side in the paper transport direction
Although they are arranged in the order of Y, they are not limited to this order, and the color order can be set arbitrarily.

The color image processing in the digital color copying machine 10 to which the image processing apparatus of this embodiment is applied will be described below with reference to FIG.

In addition to the configuration shown in FIG. 2, the digital color copying machine 10 is provided with an image processing device (not shown) including a photoelectric conversion element (CCD; Charge Coupled Device) as a document reading means.

In this image processing apparatus, as shown in FIG. 3, first, the CCD 11 that has read the image of the original document
The 8-bit data corresponding to each color of RGB is converted into the density conversion unit 1.
Send to 2. Then, in the density conversion unit 12, the CCD 1
The variation correction of 1 and the correction according to the characteristics of human eyes and the conversion to CMY data (data obtained by inverting RGB as complementary colors) are performed.

8 bits corresponding to the converted CMY colors
The area separation unit 13 discriminates the area of the data of 1) for each pixel.

Then, the gamma correction unit 14 corrects the density. Further, the color correction / black generation unit 15 corrects the data to newly generate black data K, and CMYK
It is data of 8 bits each corresponding to 4 colors.

Subsequently, the scaling unit 16 scales the image to a desired scaling ratio, and the filter processing unit 17 performs filter processing for each color.

C processed by the filter processing unit 17
Each of the MYK four-color data is transmitted to the laser unit 18, and the above-described image formation is performed.

Among the above-mentioned members, the density conversion unit 12, the region separation unit 13, the gamma correction unit 14, the color correction / black generation unit 1
5, the variable power unit 16, and the filter processing unit 17 constitute an image processing apparatus.

The present invention relates to the density conversion processing by the gamma correction unit 14 among the image processing steps shown in FIG. Here, the density conversion processing by the gamma correction unit 14 will be described with reference to FIG.

As shown in FIG. 1, the density converting section 12 performs a pre-scan for detecting a background level on a document before a main scan for reading the document (S1).
1) A histogram is created based on the obtained input image data (input signal) (S12).

This histogram is searched from the dark side, that is, the high density side, and the frequency of the histogram is compared with a predetermined threshold, and the level having the frequency that exceeds the predetermined threshold first is set as the background candidate level ( S13).

Since this predetermined threshold value is set from the ratio to the total histogram frequency, it is not necessary to set the threshold value individually for the document image size.

As a result, when this threshold value is set by the number of pixels, it is necessary to prepare the threshold value set according to the image size. However, according to the above threshold setting method, each threshold value can be set with a simple structure. An appropriate threshold can be set for the original.

The prescan is not always necessary, and the read image data may be stored in a storage means such as an image memory and the density conversion process or a series of image processes may be performed using this data. Good.

Further, in the background candidate level detection, as shown in FIG. 4, first, i is initialized to a limiter value (S21), and the frequency of the section i is compared with a predetermined threshold value (S22). If the frequency of the section i is larger than the predetermined threshold value, the section i is set as the background candidate level (S23). On the other hand, if it does not exceed the predetermined threshold value, 1 is subtracted from i (S24), and the frequency of the section i is compared with the predetermined threshold value.
Is repeated until the frequency exceeds a predetermined threshold value or i becomes 0 (S25). It should be noted that if there is no division with a frequency exceeding the predetermined threshold value until i becomes 0, it is determined that there is no background (S26).

Then, the final background level is determined according to the relationship between the detected background candidate level frequency and the light side adjacent level frequency (S14 to S16).

Further, as shown in FIG. 5, a plurality of background removal density conversion tables used for the density conversion processing are set with 8-bit data, and a table corresponding to the detected background level is selected, and the density conversion processing is performed. Is done.

As shown in FIG. 5, when the detected background level is, for example, divisions 1 and 2, a table of ◆, divisions 3 and 4 of a table, and divisions 5 and 6 of a table:
In the case of divisions 7 and 8, the table is set to x and gamma processing is performed.

As described above, in the image processing method applied to the digital color copying machine 10 of this embodiment, the background candidate level that first exceeds the predetermined threshold value is not immediately set as the background level, but is set to the bright side. Since the final background level is determined according to the relationship with the levels of the adjacent sections, it is possible to detect the background level more accurately.

The step of determining the final background level will be described later in detail.

Thereafter, the main scan is started (S17),
A density conversion process is performed according to the determined final background level (S18), and the process proceeds to the area separation processing step by the area separation unit 13 in FIG.

A limiter is provided in the detection of the background candidate level, and the limit of the density to be the background is set. In the image processing method of this embodiment, since the method of searching from the high density side is adopted, the search start level becomes the limiter.

Further, as a result of the prescan shown in FIG.
The histogram created based on the obtained input image data is created from the luminance information obtained by calculating the ratio of the input image data from the CCD, and the ratio is variable.

Specifically, in the image processing method of this embodiment, CMY data is handled as input image data, and G is calculated by the following equation (1) based on the input CMY data.
The ray data (luminance information) is calculated and a histogram is created.

Gray = a / 255 * C + b / 255 * M + c / 255 * Y (1) Note that a, b, and c are coefficients, and when the calculated value exceeds 255, it is saturated to 255.

As shown in FIG. 6, such a histogram is created by dividing the density value into a plurality of sections and counting the frequency. For example, when divided into 32 sections with a threshold value of 31, the histogram frequency h of the corresponding section is determined according to which section the input data has a density value.
The ist (i) is incremented by one.

As described above, the histogram is created from the calculated Gray data and the background level is detected, so that the background level is not detected by referring to the histogram of CMY or all RGB colors, and a simpler structure is obtained. The background level detection process can be performed at.

In the present invention, the histogram is thus represented by the frequency for a plurality of sections (density sections). Therefore, the threshold may be represented by a value (frequency) obtained by multiplying the total frequency of the histogram by a predetermined ratio as described above, but is not limited to this. It may be expressed as a percentage, and the threshold value may be similarly expressed as a ratio.

Here, the step of determining the final background level, which is a characteristic part of this embodiment, will be described below with reference to FIG.

Digital color copying machine 10 of this embodiment
In the image processing method applied to, as shown in FIG. 1, regarding the background level determination method based on the result of the comparison between the detection level frequency and its bright side adjacent level frequency, the background candidate level frequency and its bright side The final background level is determined according to the result of comparison with the frequencies of adjacent levels.

That is, as shown in FIG. 7, the background candidate level frequency is compared with its bright side adjacent level frequency (S3
1) If the background candidate level frequency is equal to or higher than the light side adjacent level frequency, the background candidate level is determined as the final background level (S32), while the light side adjacent level frequency is higher than the background candidate level frequency. In this case, the bright side adjacent level is determined as the final background level (S33).

For example, a histogram obtained by reading a character original document having a background such as a newspaper by a prescan is shown in FIG.
As shown in (a), category 8 is detected as the background candidate level. The frequency of division 8 and the frequency of division 7 are compared,
Since the frequency of category 7 is higher, category 7 is determined as the final background level. Further, as shown in FIG. 8B, in a histogram obtained by reading a cut and pasted document obtained by pasting a character document with a background such as a newspaper on a blank sheet, as shown in FIG. 8B, category 8 is similarly detected as the background candidate level.
The frequency of Category 8 and the frequency of Category 7 were compared, and Category 7
Since the frequency is higher, category 7 is determined as the final background level.

By adopting the method of searching from the high-density side in this way, it is possible to accurately detect the background level even when there is a blank paper portion that is cut and pasted. Further, by determining the background level according to the relationship between the frequencies of the background candidate levels and the frequencies of the levels adjacent to the bright side, it is possible to detect the accurate background level even with a simple configuration. .

Furthermore, in the histogram of the photographic original having a blank sheet portion and / or a highlight portion at the edge, FIG.
As shown in (c), Category 2 is detected as the background candidate level. The frequency of Category 2 and the frequency of Category 1 are compared,
Since the frequency of category 1 is higher, category 1 is determined as the final background level. Category 1 corresponds to the highlight part and the blank part. As a result, the background level can be detected more accurately without the background level being determined to have a higher density than that portion. Furthermore, it is possible to prevent deterioration of the reproduced image in the highlight portion that occurs when a high density is detected as the background level.

As described above, by comparing the background candidate level frequency and the background level frequency adjacent to the bright side and setting the level with the higher frequency as the background level, the peak search is not performed, but the simple structure is not performed. , It becomes possible to detect a level close to the peak, and more accurate detection of the background level can be performed.

The density conversion processing for the input image data of each color during the main scan is performed by performing density conversion processing for all RGB or CMY colors in accordance with the background level determined by the above steps. Configured to be removed.

From the Gray data, one of the density conversion tables shown in FIG. 5 is selected according to the final background level determined by the histogram information, and the density conversion is performed on the input image data obtained by the main scan. Perform processing. When performing background removal on color originals, color cast,
Alternatively, when a document having a colored background is read, if a process of removing only a specific color is performed, a change in color is caused in a reproduced image. Therefore, based on the Gray data obtained from the color image data, C
By performing density conversion processing on all the MY colors in the same table and determining the background level, even in the case of a document having a color cast or a background with a tint, a reproduced image in which the color tone does not change Is obtained.

In the image processing method of this embodiment, an example in which the final background level is determined by comparing the sizes of the background candidate level and its bright side adjacent level has been described, but the present invention is not limited to this. Not a thing.

For example, as shown in FIG. 9, the ratio (frequency ratio) of the light side adjacent level frequency to the background candidate level frequency is calculated (S41), and the frequency ratio is set to a predetermined value, for example, 0. If it is less than 0.9, the background candidate level is determined as the final background level (S42). If the frequency ratio is equal to or more than the predetermined value, for example, 0.9 or more, the bright side adjacent level is determined as the final background level (S42). The S43) method may be adopted.

As the predetermined value, an appropriate value may be set in advance based on many images.

Accordingly, the ratio of the light side adjacent level frequency to the background candidate level frequency is used instead of simply comparing the frequencies of the two levels, so that even if the histogram frequency varies or the error is large, It is possible to detect various background levels.

Further, in addition to the above method, the present invention performs finer control in accordance with the ratio between the background candidate level frequency and the bright side adjacent level frequency, or the background candidate level and the bright side as the background level. The configuration may be such that a level between adjacent levels is provided. Even with such a configuration, it is possible to detect the background level more accurately,
By determining the background level in accordance with the frequency ratio between the background candidate level and the bright side adjacent level, it is possible to detect a level close to the peak because of a simple configuration that does not perform peak search.

Regarding the processing order of the image processing steps,
The process order is not limited to that shown in FIG. 3, and the order of each process may be changed.

Further, the background removal density conversion table used in the density conversion processing of the present embodiment is not limited to the method shown in FIG. 5, and the detected background can be compared with the basic background removal density conversion table. The configuration may be such that the tilt and the shift amount are calculated according to the level. The method of searching the background level from the high-density side is effective for detecting the background level of a cut and pasted original obtained by scrapping a blank such as a newspaper from a scrap.

[Embodiment 2] Another embodiment of the image processing method of the present invention will be described below with reference to FIGS.

Contrary to the first embodiment, the image processing method according to the present embodiment starts the search from the bright side, first detects a level exceeding a predetermined threshold, and then detects the frequency of the level and its dark side. The feature is that the background level is determined according to the relationship with the adjacent level frequency.

The image processing process in this embodiment is as shown in FIG.
As shown in 0, prior to the main scan for reading the document, a pre-scan for detecting the background level is performed (S5
1) A histogram is created for the obtained input image data (S52).

This histogram is searched from the light side, that is, the low density side, and the frequency is compared with a predetermined threshold value,
First, a level having a frequency exceeding a predetermined threshold is set as a background candidate level (S53).

The background candidate level detection is performed in the first embodiment.
Similarly, a limiter is provided to set the limit of the density to be the base. Since the method is to search from the bright side (low density side), the search is performed up to the density classification that should be the base.

That is, the background candidate level in FIG. 10 is detected by the following processing.

First, as shown in FIG. 11, i is initialized to 1 (S61), the frequency of the section i is compared with a predetermined threshold value (S62), and if it is larger than the predetermined threshold value, the section i is the background candidate. The level is set (S63). If it does not exceed the predetermined threshold value, 1 is added to i (S64), and if i is not the limiter value (NO in S65), the frequency of the section i is compared with the predetermined threshold value. If there is no division that exceeds the threshold value until i reaches the limiter value (YES in S65), it is determined that there is no background (S66).

The method of searching the background level from the light side (low density side) is a detection method suitable for a document having a plurality of peak density that seems to be the background. Then, the final background level is determined according to the relationship between the detected background candidate level frequency and the frequency of the dark side adjacent level (S54-).
S56).

Then, the main scan is started (S57),
A density conversion process according to the determined final background level is performed (S58), and the density conversion process by the density conversion unit 12 ends.

Also in this case, as in the first embodiment, the prescan need not necessarily be performed.

In the image processing method of this embodiment, as in the first embodiment, the detected background candidate level is not immediately determined as the background level, but the background level is further determined according to the relationship with the adjacent dark side level. Finalize the level. This makes it possible to detect the background level more appropriately.

For example, when reading a document having a plurality of peak density values that seem to be the background, the histogram will be as shown in FIG.

According to the above-described background candidate level detection method, section 4 is detected as the background candidate level. Then, when the frequency of the section 4 is compared with the frequency of the section 5, the frequency of the section 5 is larger, and thus the section 5 is determined as the final base level.

On the contrary, when the search is performed from the high density direction, the section 10 is detected as the background candidate level.
The frequency of the section 10 is compared with the frequency of the section 9, and since the frequency of the section 10 of the frequency is larger, the section 10 is determined as the final base level.

However, in the case of a manuscript having a plurality of peak density like a background as shown in FIG. 12A, if the higher density category 10 is determined as the background level and the background removal density conversion processing is performed, There is a problem that the reproduction deterioration of the portion including the background candidate level of low density, that is, the highlight portion is caused.

Therefore, in the image processing method of this embodiment, since the search is performed from the bright side (low density side), the classification 5
Is determined as the background level, and appropriate background removal density conversion processing can be performed. As described above, according to the method of searching the background level from the light side (low density side), the accurate detection of the background level is performed corresponding to a document having a plurality of peak density like the background shown in FIG. It can be performed.

Further, as described above, by determining the background level according to the relationship between the frequency of the background candidate level and the frequency of the adjacent dark side level, a more accurate background level can be obtained with a simple structure. It becomes possible to detect.

If the frequency of the brightest level in the histogram created based on the input image data is the largest, the predetermined threshold value is the total frequency minus the level frequency of the brightest side. The determined value is set as the total number of pixels, and the ratio to the total number of pixels is set as a threshold.

Such a threshold value determination method can detect not only an original having a plurality of peak-like peak densities, but also a cut-and-pasted original made by scraping a cutout such as a newspaper on a blank sheet and accurately detecting the background level. Becomes

Further, the background candidate level detection method of FIG. 10 may be configured as shown in FIG.

That is, the frequency of the division 1 which is the level on the brightest side is compared with the predetermined threshold value determined in the above step (S71). If it is smaller than the predetermined threshold value, the background candidate level is detected in order from the section 1 with the detection threshold value (threshold value) 1 (S72). If it is equal to or more than the predetermined threshold value, the category 1 is excluded from the background candidate level, and the background candidate level is detected by the detection threshold value (threshold) 2 in order from the section 2 (S73).

The predetermined threshold value in this step is set to a value at which the original is assumed to be a cut and pasted original. The detection threshold 1 is a normal detection threshold, and is set as a ratio to the total histogram count. The detection threshold 2 is
It is set according to the frequency of category 1.

For example, the number of pixels obtained by subtracting the frequency of category 1 from the total frequency of the histogram is set as the total number of pixels, and the threshold value is set in proportion to this. When a cut and pasted document obtained by pasting images having a plurality of peak density like a background on a blank sheet is read, a histogram as shown in FIG. 12B is formed.

According to the image processing method of this embodiment, since the frequency of the section 1 which is the brightest side is higher than the predetermined threshold value, the section 1 is excluded from the background candidates. And
The detection threshold is set based on the ratio of the number of histograms obtained by subtracting the frequency of category 1 from the histogram total frequency, and category 4 is detected as the background candidate level. Then, the frequency of the section 4 and the frequency of the section 5 adjacent to the dark side are compared, and the frequency of the section 5 is larger, so
Is determined as the final background level.

As described above, since it is constructed so as to subtract the influence of the white level data, it is applicable not only to a manuscript having a plurality of peak density like a background but also to a cut-and-pasted manuscript obtained by scraping a cutout such as a newspaper on a blank paper. It is possible to detect the background level more accurately.

Further, the predetermined threshold value for detecting the background level according to the present embodiment is changed according to the density level for each section of the histogram like the detection threshold value 1 and the detection threshold value 2 described above, and the threshold value for each section is changed. May be provided to detect the background level.

For example, in the case of individually setting the threshold values corresponding to the respective sections of the histogram, the detection threshold value 1 corresponding to the section 1 which is the brightest side is assumed to correspond to the cut and pasted original, and the other sections. The background detection threshold value may be set higher than the detection threshold value 2 corresponding to.

This makes it possible to prevent category 1 from being erroneously detected as the background candidate level even when there is white data such as a cut and pasted original. Therefore, it is possible to detect the background level more accurately not only for a document having a plurality of peak density values like a background but also for a cut and pasted document.

In the image processing method of the present invention, an image processing apparatus that detects a background level from a read input signal and performs processing according to the detected background level from the input signal Creating a histogram, searching from the dark side, first detecting the level exceeding a predetermined threshold, and determining the base level according to the relationship between the frequency of the level and the frequency of the level adjacent to the bright side. The image processing apparatus may be characterized by:

[Embodiment 3] Still another embodiment of the image processing method of the present invention will be described below with reference to FIGS. 14 to 17.

FIG. 14 is a block diagram showing the arrangement of a digital color copying machine as an image forming apparatus equipped with the image processing apparatus of the present invention.

As shown in FIG. 14, the image processing device 41
Is a density conversion unit 51, a region separation processing unit 52, a γ correction unit 5
3, color correction unit 54, black generation undercolor removal unit 55, scaling unit 5
6, a spatial filter processing unit 57, an output gradation correction unit 58, and a gradation reproduction processing unit 59.

An image input device (image input means) 42 and an image output device (image output means) 43 are further connected to the image processing device 41.

The image input device (image reading means) 42 is composed of, for example, a scanner section having a CCD (Charge Coupled Device), and an image of reflected light from a document is converted into RGB.
The analog signals (image data) of (R: red, G: green, B: blue) are read by the CCD and input to the image processing device 41.

The analog signal read by the image input device 42 is transmitted in the image processing device 41 by the density conversion portion 51, the area separation processing portion 52, the γ (gamma) correction portion 53, the color correction portion 54, and the black generation background. The color removal unit 55, the scaling unit 56, the spatial filter processing unit 57, the output gradation correction unit 58, and the gradation reproduction processing unit 59 are sent in this order, and an image is output as a CMYK digital color signal (output image data). It is output to the device 43.

RGB read by the image input device 42
The analog signal of is subjected to A / D (analog / digital) conversion processing and converted into a digital signal. Density converter 51
Is the image input device 42 for digital RGB signals.
Image used in the image processing device 41, such as processing for removing various distortions generated in the illumination system, the imaging system, and the imaging system, and adjusting the color balance with respect to the reflectance signals of RGB. The processing system converts the signal into a signal that can be easily handled by the processing system.

The area separation processing section 52 determines from the RGB signals that
Each pixel in the input image is divided into a character area, a halftone dot area, and a photographic area. The area separation processing unit 52
Based on the separation result, an area identification signal indicating which area the pixel belongs to is output to the black generation undercolor removal unit 55, the spatial filter processing unit 57, and the gradation reproduction processing unit 59, and the density conversion unit. The input signal output from 51 is output as it is to the γ correction unit 53 in the subsequent stage.

The γ correction unit 53 is for performing the background determination process. The configuration of the γ correction unit 53 and the background determination process will be described in detail later.

The color correction section 54 removes color turbidity based on the spectral characteristics of the CMY (C: cyan, M: magenta, Y: yellow) color material containing unnecessary absorption components in order to realize faithful color reproduction. The processing is performed.

The black generation and undercolor removal unit 55 uses the CM after the color correction.
CMY is a process for generating a black (K) signal from Y three-color signals and a process for generating a new CMY signal by subtracting the K signal obtained by black generation from the original CMY signal.
3 color signals are converted into CMYK 4 color signals.

As an example of black generation processing, there is a method (general method) for generating black by skeleton black. In this method, the input / output characteristics of the skeleton curve are y = f
(X), input data is C, M, Y, output data is C ', M', Y ', K', UCR (Under Color Remo
If the val) ratio is α (0 <α <1), the black generation undercolor removal processing is performed by the following equations (2) to (5) K ′ = f {min (C, M, Y)} (2) C ′ = C−αK ′ (3) M ′ = M−αK ′ (4) Y ′ = Y−αK ′ (5)

The magnification changing unit 56 enlarges or reduces the image based on a signal input from the operation panel provided in the image forming apparatus.

The spatial filter processing unit 57 performs a spatial filter process by a digital filter on the image data of the CMYK signals input from the scaling unit 56 based on the region identification signal to correct the spatial frequency characteristic. The processing is performed so as to prevent blurring of the output image and deterioration of graininess.

The tone reproduction processing unit 59, like the spatial filter processing unit 57, also performs a predetermined process on the image data of the CMYK signals based on the area identification signal.

For example, in order to improve the reproducibility of black characters or color characters, the areas separated into characters by the area separation processing unit 52 are processed by the spatial filter processing unit 57 in the sharpness enhancement process in the spatial filtering process to have a high frequency. The amount of emphasis is increased. At the same time, in the gradation reproduction processing unit 59, binarization or multi-value conversion processing on a high resolution screen suitable for reproduction of high frequency is selected.

The area separated into halftone dots by the area separation processing unit 52 is subjected to low-pass filter processing for removing the input halftone dot component in the spatial filter processing unit 57. Then, the output gradation correction unit 58 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 image output device 43, and then a gradation reproduction processing unit 59. Finally, gradation reproduction processing (halftone generation) is performed to separate the image into pixels and process so that each gradation can be reproduced.

The areas separated into photographs by the area separation processing section 52 are subjected to binarization or multi-value processing on the screen with emphasis on gradation reproducibility.

The image data that has been subjected to the above-mentioned processing is
The image is once stored in the storage means, read at a predetermined timing, and input to the image output device 43.

The image output device 43 outputs the image data onto a recording medium (for example, paper). For example, a color image forming device using an electrophotographic method or an ink jet method can be used. It is not limited. The above processing is performed by a CPU (Central Pr
Ocessing Unit).

Next, the configuration of the γ correction section 53 will be described with reference to FIG.
It will be described based on 5.

The γ correction section 53 performs pre-scan (for background level detection) for determining whether or not there is a background,
A histogram is created based on the obtained input image data (input signal).

The γ correction section 53 includes a signal conversion section 61, a histogram creation section 62, a density classification extraction section 64, and a threshold setting section 6.
3, density correction unit 65, CPU 66, and ROM (Re
Ad Only Memory) 67.

The signal conversion section 61 converts the RGB signal (input signal) into luminance data (luminance information) L. The luminance data L (Gray data) is represented by the following equation (6) L = a · R + b · G + c · B (6) using RGB data.

Generally, a, b, and c are values that are not the same ratio for converting RGB data to luminance data (for example, a: b: c = 1.000: 4.5907: 0.0601), but here a = B = c (= 1/3). a = b =
By setting c, it is possible to grasp the characteristic of the document (reddish or the like), and it is possible to detect a more colored background.

Further, when the histogram is created, the value L'as the brightness data which is the complementary color inversion of the brightness data L.
(L '= 255-L) is used.

The histogram creating section creates a histogram based on the brightness data (brightness information) L '. The density division is, for example, 32 divisions (see FIGS. 16A to 16C).

The density classification extraction unit 64 searches the histogram from the low density side, that is, the high brightness (bright) side to the high density side, that is, the low brightness (dark) side, and the frequency of the histogram is predetermined. The threshold value is compared with the threshold value and the background is extracted. The method of searching for the background level from the low density side is a detection method corresponding to a document having a plurality of peak density that seems to be the background. The background extraction processing method here will be described later.

The threshold setting unit 63 uses the predetermined threshold A
・ B and C are stored. These thresholds A, B, C are
They are independent of each other and can be set individually. The thresholds A to C are represented by values or ratios that the total frequency of the histogram should be multiplied by a predetermined ratio (for example, a ratio such as 15% or 20%).

The threshold value A (first threshold value) is a value for judging whether or not there is a lot of white (white without tint) as a background. Controls whether to remove.

The threshold value B (second threshold value) is a value for determining the presence / absence of a background candidate.

Here, there is a candidate for the background (frequency ≧ threshold value B).
In this case, the density classification extraction unit 64 adds the frequency of the density classification equal to or higher than the threshold value B and the frequency of the density classification on the high density side.

For example, when the frequency in the section 6 is greater than or equal to the threshold value B, the frequency in the section 6 and the frequency in the section 7 are added.

The threshold value C (third threshold value) is a value for judging the presence or absence of a background (judging as a background).

If the added frequency ≧ threshold value C, the density category equal to or higher than the threshold value B or the density category on the high density side is used as the base.

For example, in the case of the sum of the frequencies of the categories 6 and 7 (the added frequencies) ≧ the threshold value C, it is either the category 6, the category 7, or the concentration between the categories 6 and 7. It is determined whether the area is used as a base.

On the other hand, when the added frequency is less than the threshold value C, there is no background.

The thresholds A, B and C are set in advance to optimum values based on many images.

The threshold values A, B, and C can be arbitrarily changed. For example, in an image forming apparatus, an operation panel (for example, a display unit including a liquid crystal display,
Configuration button), press the setting button,
Alternatively, by moving the indicator, the threshold value A ~
It suffices to change C, display the result on the display unit, and prompt the user as to which level to set.

Also, in the computer system,
The thresholds A to C are set by directly inputting a numerical value using a keyboard or a mouse, or by dragging a symbol indicating the threshold, and the user is preferable to see the image displayed on the image display device such as a liquid crystal display. The thresholds A to C may be set.

In this case, as the threshold values A to C, a frequency or a ratio with respect to the total frequency may be input, but the latter is preferable in that it can be objectively grasped.

For example, as shown in FIG. 5 described in the first embodiment, the density correction section 65 sets a plurality of density conversion tables and selects a table corresponding to the detected background level (density classification).

For example, when there is no background level, a through table (* mark) for which no correction is made is selected. Further, when the background level is the density categories 1 and 2, ◆, when the background level is the density categories 3 and 4, ■, and the background level is the density categories 5 and 6
In the case of, the value is set as ▲, and when the background level is classified into categories 7 and 8, it is set as ×, and a value obtained by correcting the input according to the background level is output.

For density correction, a plurality of density conversion tables are not set from the beginning as shown in FIG. 5, but the basic background removal density conversion table (through table) is used to detect the detected background level. Alternatively, the inclination and shift amount may be calculated according to

During the main scan, the signal conversion unit 61.
The processes of the histogram creation unit 62 and the density classification extraction unit 64 are skipped, and only the process of the density correction unit 65 is performed.

A ROM (Read Only Memory) 67 stores a program for performing processing of each unit of the γ correction unit 53. Further, the CPU 66 executes the program.

The background judgment method will be described below with reference to FIG.
A specific description will be given based on (a) to (c). Here, the threshold value C is, for example, C = 10 from the ratio to the total frequency.
It is assumed that x10 ⁴ pixels (hereinafter, threshold values A to C are shown as values normalized by 10 ⁴ ).

FIG. 16A is a histogram of a text original having a background such as a newspaper. First, the threshold value A is used to determine the low density section.

In the figure, since the threshold value A does not exceed the threshold value A in the category 1, the background level determination is subsequently made in the threshold value B, and the category 6 is detected as the background candidate. Then, the frequency of the background candidate section 6 and the frequency of the section 7 adjacent to the high density side are added. In this case, the added value is 17 (= 8 + 9)
Since the threshold value is equal to or higher than the threshold value C (= 10), the category 6 is determined as the background level.

Here, whether or not the threshold value A is exceeded is determined by the division (density division on the search start side) 1, but this is limited to this if the density division on the search start side (low density side) is used. However, it may be performed in the division 1 and the division 2. For example, the background determination of the low density section may be performed by comparing the threshold value A with a value obtained by adding the frequency of the section 1 and the frequency of the section 2.

FIG. 16B is a histogram of a cut-and-pasted document obtained by pasting a text document having a base such as a newspaper on a blank sheet.

First, the threshold A is used to determine the low-density category. In this case, however, the frequency of category 1 exceeds the threshold A, so the background is not detected. That is, it is determined that there is no background.

By changing the set value of the threshold value A, it is possible to set how much white data amount should be used to determine the background.

FIG. 16 (c) is a histogram of a photographic original having a white paper portion and / or a highlight portion at the edge.

First, the threshold value A is used to determine the low-density category. In this case, however, the frequency of category 1 does not exceed the threshold value A, so the threshold value B is subsequently used to determine the background level. Figure 1
In the case of 6 (c), Category 2 is detected as a background candidate.
Next, the frequency of the background candidate section 2 and the frequency of the section 2 which is adjacent to the high density side of the section 2 are added. In this case, the added value is 5 (= 3 + 2), which is smaller than the threshold value C (= 10), so it is determined that there is no background.

[0207] The background determination method (image processing method) described above
Will be described based on the flowchart shown in FIG.

First, a prescan is performed (S81), and then a histogram is created (S82). Then, the background determination of the low density section is performed based on the histogram.

At this time, if the frequency of the low density section is equal to or higher than the predetermined value, it is determined that there is no background.

That is, assuming that the histogram frequency of the density category i is hist (i) and the predetermined value is the threshold value A, hist
If (i) <threshold value A (YES in S83), it is determined that there is a background candidate, and the process proceeds to the background candidate detection process. On the other hand, hist
If (i) ≧ threshold value A (NO in S83), it is determined that there is no background (S84), and the process ends.

Next, the high density side is searched, and first, a segment having a frequency equal to or higher than a predetermined value, that is, the threshold value B is detected as a background candidate (a background level is determined). At this time, if no background candidate is detected by the predetermined section, it is determined that there is no background.

For example, if hist (i) ≧ threshold value B (YES in S85), the density division i satisfying the left condition is first set as the background candidate, and the process proceeds to the background level determination process (S86). If the above condition is not satisfied until i reaches the predetermined density category (NO in S85), it is determined that there is no background (S84), and the process is ended.

After that, the background level is determined from the detected background candidates. Here, the background level is determined by the sum of the frequencies of the detected background candidate density categories and the frequencies of the high-density adjacent density categories. Here, the background candidate density segment is segment i ′, the histogram frequency of segment i ′ is hist (i ′),
If the predetermined value is the threshold value C, hist (i ′) + hist
If (i ′ + 1) ≧ threshold value C (YES in S86), the density of the section i ′ is determined as the background level (S87). On the other hand, hist (i ′) + hist (i ′ + 1) <threshold value C
If (NO in S86), it is determined that there is no background (S8).
4).

Also in the case of the present embodiment, the above-mentioned first embodiment
-Similar to 2, prescan is not always necessary. In this case, the image data output from the area separation processing unit 52 may be stored in a storage unit such as an image memory, a background determination process may be performed using this image data, and then a series of image processes may be performed.

The image processing method of this embodiment may be recorded in a computer-readable recording medium in which a program to be executed by a computer is recorded.

As a result, it is possible to freely provide a recording medium on which a program for performing the image processing method is recorded.

In the present embodiment, the recording medium may be a program medium such as a memory (not shown) such as a ROM itself because the microcomputer performs the process. Although not shown, a program reading device may be provided as an external storage device, and the program medium may be readable by inserting a recording medium therein.

In any case, the stored program may be accessed and executed by a microprocessor, or in any case, the program is read and the read program is read by a microcomputer. The program may be downloaded to a program storage area (not shown) and the program may be executed. It is assumed that this download program is stored in the main body device in advance.

Here, the program medium is a recording medium which can be separated from the main body, and is a tape system such as a magnetic tape or a cassette tape, or a floppy (registered trademark).
Magnetic disks such as disks and hard disks, and CD-Rs
Disk systems for optical disks such as OM / MO / MD / DVD, card systems such as IC cards (including memory cards) / optical cards, mask ROM, EPROM (Erasab
le Programmable Read Only Memory), EEPROM
(Electrically Erasable Programmable Read Only Mem
ory), a flash ROM, or a semiconductor memory including a semiconductor memory that holds the program in a fixed manner.

Further, in the present embodiment, since the system configuration is such that a communication network including the Internet can be connected, it may be a medium that carries the program fluidly so as to download the program from the communication network. . When the program is downloaded from the communication network in this way, the program for downloading may be stored in the main body device in advance, or may be installed from another recording medium.

The above-mentioned image processing method is executed by the recording medium being read by a program reading device provided in a digital color image forming apparatus or a computer system.

The computer system displays an image input device such as a flatbed scanner, a film scanner or a digital camera, a computer on which various processes such as the above image processing method are performed 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 a printer for outputting the processing result of the computer to paper or the like. Further, a modem or the like is provided as a communication means for connecting to a server or the like via a network.

As described above, according to the image processing method of the present embodiment, the histogram is searched from the low density side to the high density side, and the frequency of the density division on the search start side, that is, the low density side and the threshold value A are set. If it is determined to perform the step of detecting the background candidate in the first step and the first step of determining whether the original is made to have no background or the step of detecting the background candidate after that, The second step of determining whether there is a background candidate by comparing the frequency of the division with the threshold B, and the frequency is equal to or more than the threshold B when it is determined that there is a background candidate in the second step. A third step of determining the presence / absence of a background by adding the frequency of the density section and the frequency of the density section adjacent to the high density side and comparing the added value with the threshold value C. Having the first to third steps Performing under color removal process based on the results.

Specifically, in the first step, when the frequency of the density section on the low density side (density section 1) is equal to or higher than the threshold value A, it is determined that the original has no background. On the other hand, if the frequency of the density section on the low density side (density section 1) is smaller than the threshold value A, the second
Move to the process.

In the second step, the frequency of the density classification is set to the threshold value B.
The background level is determined by comparing with.
Specifically, if a density section having a frequency equal to or higher than the threshold value B is found by the search from the low density side up to a predetermined density section, the density in the density section is detected as a background candidate. On the other hand, if there is no density category having a frequency equal to or higher than the threshold value B up to the predetermined density category, it is determined that there is no background.

In the third step, the frequency in the density category selected as the background candidate in the second step (the frequency of the density category in which the frequency becomes equal to or higher than the threshold value B after searching from the low density side) and the density category Determines the presence / absence of the background by adding the frequency of the density section adjacent to the high density side and comparing with the threshold value C. Specifically, when the added value is equal to or more than the threshold value C, the density of the density segment selected as the background candidate in the second step, or
With this density section, the density of the density section adjacent to the high density side is determined as the background level. On the other hand, when the added value is smaller than the threshold value C, it is determined that there is no background.

As a result, the readability of a photographic printing paper, a printed photographic original, an original with a background fog or a backslip can be improved even with a simple structure that does not require the process of determining the type of the original. An optimum reproduced image can be obtained. Further, for example, even in the case of a photographic document having a blank sheet portion and / or a highlight portion on the edge, it is possible to detect an accurate background level.

That is, even in the case of a document having a plurality of peak densities, which is the peak density, that is, white data, it is possible to detect the background level more accurately.

Also, by using the same background removal method for a plurality of color components, it is possible to obtain a reproduced image in which no color change occurs in the color image.

Further, by carrying out the above-mentioned first to third steps, it becomes possible to detect a more appropriate background level corresponding to variations in histogram frequency and errors. By determining the background level according to the frequency relationship with the adjacent density section, it is possible to detect the level close to the peak while having a simple configuration in which the peak search is not performed.

[0231]

As described above, the image processing method of the present invention starts the search from the dark side of the histogram, detects the level that first exceeds the predetermined threshold as the background candidate level, and detects the background candidate level. The frequency is compared with the frequency of the level adjacent to the bright side, and the level with the higher frequency is determined as the base level.

Therefore, it is possible to more accurately detect the background level, and it is possible to form an excellent image.

That is, the level that first exceeds the predetermined threshold value is set as the background candidate level, and the frequency of the background candidate level is compared with the frequencies of the levels adjacent to the bright side thereof, and the level having the larger frequency is set as the background level. By detecting
It is possible to detect the background level more accurately. Thus, even with a simple image processing method that does not perform peak search, it is possible to detect a level closer to the peak.

Therefore, the readability of an original document with a background fog or back-slip is improved, and the background level can be detected even for a cut and pasted document in which a blank such as a newspaper is cut out and scrapped.

Further, by controlling the predetermined threshold value and the like, it is possible to accurately detect the background level corresponding to not only the cut and pasted original but also various original images such as an original having a plurality of peak-like peak densities. .

The image processing method of the present invention is as described above.
The search is started from the dark side of the histogram, and the level that first exceeds a predetermined threshold is detected as the background candidate level, and the frequency ratio of the frequency of the level adjacent to the bright side with respect to the frequency of the background candidate level is obtained. When the frequency ratio is less than a predetermined value, the background candidate level is determined as the background level, and when the frequency ratio is the predetermined value or more, the level adjacent to the bright side is set as the background level. This is the configuration to be decided.

Therefore, it is possible to detect the background level more accurately and to form an excellent image.

That is, by detecting the background level in accordance with the frequency of the detection level and the frequency ratio of the frequencies of the levels adjacent to the light side, a more accurate background level corresponding to the variation in the histogram frequency and the error is obtained. It becomes possible to detect.

As a result, the background level closer to the peak can be detected with a simple structure in which the peak search is not performed.

The image processing method of the present invention is as described above.
The search is started from the bright side of the histogram, and the level that first exceeds the predetermined threshold is detected as the background candidate level, and the frequency of the background candidate level and the frequency of the level adjacent to the dark side are compared to determine the frequency In this configuration, the larger level is determined as the base level.

Therefore, it is possible to detect the background level more accurately and to form an excellent image.

That is, the search is started from the bright side of the histogram, the frequencies of the background candidate levels detected by the predetermined threshold are compared with the frequencies of the levels adjacent to the dark side, and the level with the higher frequency is the background. By determining the level, it is possible to more accurately detect the background level even for a document having a plurality of peak density values that seem to be the background.

As a result, the background level closer to the peak can be detected with a simple structure in which the peak search is not performed.

Further, according to the image processing method of the present invention, when the frequency of the brightest side level in the histogram is larger than the predetermined threshold value, the level is excluded from the background candidate level and the predetermined threshold value is set to the predetermined threshold value. This is a configuration that is changed according to the frequency of the background candidate level.

Therefore, when the frequency of the level on the brightest side of the histogram is larger than the predetermined threshold value, the level is excluded from the background candidate levels, and the background level considering the influence of the white level data amount is taken into consideration. It becomes possible to detect. Therefore, in addition to a manuscript having peak density like multiple backgrounds, even a cut and pasted manuscript with scraps of newspapers cut out on a white paper, etc., without being affected by the white level data,
The effect is that an accurate background level can be detected.

Further, the image processing method of the present invention is so constructed as to perform the same density conversion processing on each of the input signals of the respective colors in accordance with the background level determined above.

Therefore, in the background removal performed based on the input signals of the respective colors, when the background removal of only a specific color is performed on an original having a background with color cast or tint, the change of the color tone is caused. There is a problem that
By performing the same density conversion processing for each of B and CMY colors, it is possible to obtain a reproduced image with no change in color.

In the image processing method of the present invention, the predetermined threshold value is determined by the ratio to the total frequency of the histogram.

Therefore, since the predetermined threshold value used for the background level detection is determined as a ratio to the total histogram frequency, it is not necessary to set a plurality of threshold values according to the document size, and various types can be used with a simple structure. The effect is that an appropriate threshold can be set for the original document.

Further, the image processing method of the present invention has a structure in which the predetermined threshold value is changed according to the density level for each section of the histogram, and the threshold value for each section is determined to detect the background level.

Therefore, for example, by setting the threshold value 1 corresponding to the brightest section 1 in the above histogram to be larger than the threshold value 2 corresponding to the other sections,
Can be prevented from being erroneously detected as the background candidate level. Therefore, there is an effect that more accurate detection of the background level is possible, which corresponds not only to a document having a plurality of peak-like peak densities but also to a cut and pasted document.

The image processing method of the present invention is as described above.
By searching the histogram in a predetermined direction and comparing the frequency of the density classification on the search start side with the first threshold value, it is determined whether the original is set to have no background or the step of detecting the background candidate is performed. If it is determined that the step 1 and the step of detecting the background candidate in the first step are performed, it is determined whether there is a background candidate by comparing the frequency of the density classification with a second threshold. Process and the second
In the step of, if it is determined that there is a candidate for the background,
By adding the frequency of the density section having the frequency equal to or higher than the second threshold and the frequency of the density section adjacent to the density section in the search direction, and comparing the added value with the third threshold, A third step of determining the presence or absence, and the first to third
The background removal process is performed based on the result of the process.

Specifically, the search is performed from the low density side to the high density side of the histogram.

As a result, the readability of a photographic printing paper, a print photographic original, an original with a background fog or a backslip can be improved, even with a simple structure that does not require the processing for determining the type of the original. An optimum reproduced image can be obtained. That is, it is possible to detect the background level more accurately even in the case where there are a plurality of originals having peak density that is similar to the background, that is, white data exists.

Further, by using the same background removal method for a plurality of color components, it is possible to obtain a reproduced image in which no color change occurs in the color image.

The image processing method of the present invention has a structure in which the histogram is created based on the luminance information obtained from the ratio calculation from the input signal.

Thus, the brightness information (Gray) calculated based on the input signal such as the CMY signal or the RGB signal is obtained.
CMY or RGB by creating a histogram from the
It is not necessary to create and refer to histograms for all the colors, and there is an effect that the base level can be accurately detected with a simpler configuration.

The image processing method of the present invention has a structure in which the luminance information is converted with the same ratio of the plurality of color components in the input signal.

As a result, the characteristics of the document such as reddish can be grasped, and the more colored background can be detected. Therefore, there is an effect that it is possible to detect a more appropriate background without the problem that the detection accuracy varies depending on the tint of the background.

In the image processing method of the present invention, in the third step,
When the added value is equal to or greater than the third threshold value, the density of the density segment whose frequency is equal to or greater than the second threshold value, or the density of the density segment adjacent to the density segment in the search direction is determined as the background level. .

As a result, it becomes possible to detect the background level more appropriately in accordance with the variation in the histogram frequency and the error. By determining the background level according to the frequency relationship with the adjacent density section, it is possible to detect the level close to the peak while having a simple configuration in which the peak search is not performed.

The image processing method of the present invention comprises, in the first step,
When the frequency of the density classification on the search start side is equal to or higher than the first threshold value, the original is determined to have no background.

With this arrangement, an accurate background level can be detected with a simple structure even when, for example, a photographic document having a blank sheet portion and / or a highlight portion at the edge, that is, white data exists. .

In the image processing method of the present invention, the first threshold value, the second threshold value, and the third threshold value are independent of each other, and each of them can be arbitrarily changed.

According to the above configuration, it is possible to change the first threshold value, the second threshold value, and the third threshold value that are suitable for the purpose.

Therefore, by controlling the threshold value for determining the background in this way, it is possible to accurately detect the background level not only for the cut and pasted original but also for an original having a plurality of peak-like peak densities. Produce an effect.

The image processing apparatus of the present invention, as described above,
Using the image processing method described above, the background level of the read original is determined and the background removal processing is performed.

As a result, white data in a document including a halftone picture, that is, a print picture or a photographic paper picture is not detected as a background. Therefore, it is possible to prevent necessary image information from being lost. Further, there is an effect that a desired background removal process can be performed without causing a change in color.

The image forming apparatus of the present invention, as described above,
An image processing device described above; an image input device that reads a document and inputs image data to the image processing device;
An image output device that forms an image on a recording medium based on output image data output by the image processing device.

As a result, it is possible to prevent the necessary image information from being lost and to change the color tone by not detecting the background of the original including the halftone dot photograph and the photographic paper photograph, for example. It is possible to carry out a desired underlayer removal process without the need.

Therefore, without judging the type of the original,
It is possible to prevent deterioration of image quality. As a result, there is an effect that a high quality image can be output in the image forming apparatus.

The image processing program of the present invention has a configuration for causing a computer to execute the above-described image processing method.

As a result, the same operational effects as the above-described image processing method can be obtained. Further, the image processing method described above can be general-purpose.

The computer-readable recording medium of the present invention has a structure in which the above-mentioned image processing program is recorded.

As a result, the same operational effects as the above-described image processing method can be obtained. Further, the image processing method described above can be general-purpose.

[Brief description of drawings]

FIG. 1 is a flowchart showing an embodiment of an image processing method of the present invention.

FIG. 2 is a schematic diagram showing an internal configuration of a digital color copying machine according to an embodiment of the present invention.

3 is a block diagram showing an image processing step by the digital color copying machine of FIG.

4 is a flowchart showing a method of determining a background candidate level in the flowchart of FIG.

5 is a density conversion table used in a density conversion processing step in the flowchart of FIG.

FIG. 6 is a histogram showing an example of reading a document by prescanning.

FIG. 7 is a flowchart showing an example of a method of determining a background level in the flowchart of FIG.

FIGS. 8A to 8C are histograms showing another example in which a document is read by prescanning.

9 is a flowchart showing another example of the method of determining the background level in the flowchart of FIG.

FIG. 10 is a flowchart showing another embodiment of the image processing method of the present invention.

11 is a flowchart showing an example of a method of determining a background candidate level in the flowchart of FIG.

12A and 12B are histograms showing an example in which a document is read by prescanning in the flowchart of FIG.

13 is a flowchart showing another example of the method for determining the background candidate level in the flowchart of FIG.

FIG. 14 is an explanatory diagram showing a configuration of an image processing apparatus in the digital color copying machine shown in FIG.

15 is an explanatory diagram showing a configuration of a γ correction unit shown in FIG.

16A to 16C are histograms showing still another example of reading an original by prescanning.

FIG. 17 is a flowchart showing an example of a background level determination method in still another embodiment of the image processing method of the present invention.

[Explanation of symbols]

11 CCD 12 Density converter 13 Area separation unit 14 Gamma correction section 15 color correction / black generation part 16 Variable magnification section 17 Filter processing unit 18 Laser unit 41 Image processing device 42 Image input device 53 γ correction unit 62 Histogram creation section 63 Threshold setting section 65 Density correction unit

Continued front page    (72) Inventor Mitsuru Tokuyama             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Masaaki Otsuki             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Eiji Nishimitsu             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Norioka Yasuoka             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Shinji Imagawa             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Tatsuya Tanaka             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Hiroshi Kono             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company (72) Inventor Tomoe Matsuoka             22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka             Inside the company F term (reference) 2H027 DA44 DB01 DE07 EC06 EC07                       EC09 EC10 EC11 EC18 FA28                       FC02                 5B057 AA11 BA02 CA02 CA08 CA12                       CA16 CB02 CB08 CB12 CB16                       CC01 CE02 CH08                 5C077 LL02 MP01 PP10 PP25 PQ12                       PQ19 PQ20 SS01 TT06

Claims (18)

[Claims] 1. An input signal obtained by scanning an original,
In the image processing method of creating a histogram, detecting a background level, and performing image processing according to the background level, a search is started from the dark side of the histogram, and a level that first exceeds a predetermined threshold is set as a background candidate level. An image processing method, characterized by detecting and comparing the frequency of the background candidate level with the frequency of a level adjacent to the bright side, and determining the level with the higher frequency as the background level. 2. An input signal obtained by scanning an original,
In the image processing method of creating a histogram, detecting a background level, and performing image processing according to the background level, a search is started from the dark side of the histogram, and a level that first exceeds a predetermined threshold is set as a background candidate level. Upon detection, the frequency ratio of the frequency of the level adjacent to the bright side to the frequency of the background candidate level is obtained, and when the frequency ratio is less than a predetermined value, the background candidate level is set to the background level. And the level adjacent to the bright side is determined as the background level when the frequency ratio is equal to or higher than the predetermined value. 3. An input signal obtained by scanning a document,
In the image processing method of creating a histogram, detecting a background level, and performing image processing according to the background level, a search is started from the light side of the histogram, and a level that first exceeds a predetermined threshold is set as a background candidate level. An image processing method which detects and compares the frequency of the background candidate level with the frequency of a level adjacent to the dark side, and determines the level with the higher frequency as the background level. 4. When the frequency of the brightest level in the histogram is larger than the predetermined threshold value, the level is excluded from the background candidate level and the predetermined threshold value is set according to the frequency of the background candidate level. The image processing method according to claim 3, wherein the image processing method is changed. 5. The image according to claim 1, wherein the same density conversion processing is performed on each of the input signals of each color according to the determined background level. Processing method. 6. The image processing method according to claim 1, wherein the predetermined threshold value is determined by a ratio to the total frequency of the histogram. 7. The background level is detected by changing the predetermined threshold value according to the density level of each section of the histogram and determining the threshold value of each section.
5. The image processing method according to any one of 5 above. 8. From an input signal obtained by reading a document,
In an image processing method for creating a histogram, detecting a background level, and performing image processing according to the background level, the histogram is searched in a predetermined direction, the frequency of the density division on the search start side and the first threshold value. And a first step of determining whether the original is to have no background or a step of detecting a background candidate after that, and a step of detecting a background candidate in the first step. In the second step, it is determined that there is a background candidate in the second step of determining whether there is a background candidate by comparing the frequency of the density classification with a second threshold value. In this case, the frequency of the density section having the frequency equal to or higher than the second threshold value and the frequency of the density section adjacent to the density section in the search direction are added, and the added value is compared with the third threshold value. Especially Ri and a third step of determining the presence or absence of base, an image processing method and performing background removal process based on the results of the third step from the first. 9. The search is performed from the low density side to the high density side of the histogram.
The image processing method described in. 10. The image processing method according to claim 1, wherein the histogram is created based on luminance information obtained by calculating a ratio from the input signal. 11. The image processing method according to claim 10, wherein the luminance information is converted such that the ratios of a plurality of color components in the input signal are equalized. 12. In the third step, the added value is a third value.
When the density is equal to or more than the threshold value of, the density of the density section having the frequency equal to or higher than the second threshold value, or the density of the density section adjacent to the density section in the search direction is determined as the background level. 8. The image processing method according to 8 or 9. 13. The method according to claim 8, wherein in the first step, it is determined that the original has no background when the frequency of the density division on the search start side is equal to or higher than a first threshold value. Image processing method. 14. The first threshold value, the second threshold value, and the third threshold value are independent of each other, and each of them is configured to be arbitrarily changeable.
The image processing method according to any one of 2 and 13. 15. An image processing apparatus, wherein the image processing method according to claim 1 is used to determine a background level of a read document and perform background removal processing. 16. An image processing apparatus according to claim 15, an image input apparatus for reading a document and inputting it as image data to the image processing apparatus, and output image data output by the image processing apparatus, An image forming apparatus, comprising: an image output device that forms an image on a recording medium. 17. An image processing program that causes a computer to execute the image processing method according to claim 1. Description: 18. A computer-readable recording medium on which the image processing program according to claim 17 is recorded.