JP3263232B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus,
In particular, the present invention relates to an image processing apparatus used for image quality improvement processing related to background removal.

[0002]

2. Description of the Related Art Conventionally, if an image reading apparatus such as a digital copying machine reads an original and attempts to faithfully reproduce the image, the background of the original is also reproduced as it is, making the image hard to see. Further, since the ink is applied to the background, the ink consumption is increased. Therefore, conventionally, for example, a pixel having a relatively low density level is detected as a background before outputting an image, and the background is removed by performing processing such as replacing the pixel value with 0. In a general method of detecting a background area, a threshold corresponding to the density level of the background of the document is set, and pixels equal to or less than the threshold are regarded as the background. As a method of setting the threshold, there are a method of setting by the user and a method of automatically setting based on the characteristics of the input image. In the case of the method set by the user, high-quality background removal can be performed by setting the background removal threshold with high precision, but if the threshold is not set to an appropriate value, for example, when the threshold is too low If the background remains, and the threshold is too high, highlights of the pattern portion other than the background are not reproduced. In the method of automatically setting the threshold, the threshold can be set in real-time processing such as pixel-based processing or block-based processing. It is necessary to analyze the features, which requires a very long processing time. For example, in a background removal method of an image recording apparatus described in Japanese Patent Application Laid-Open No. 4-313744, an attempt is made to effectively separate a background and an image of a document to obtain an optimum background removal level. Has been conducted,
Since the threshold value used for the background determination is determined at the time of the pre-scan and is given to the background removal unit as the threshold value for the background removal at the time of the main scan (copy scan), it still takes a considerable processing time to analyze the characteristics of the entire input image. It takes.

[0003]

SUMMARY OF THE INVENTION It is an object of the present invention to automatically set a threshold value for background removal without performing prescan in a color image output apparatus of a frame sequential type (one drum type). Accordingly, an object of the present invention is to provide an image processing apparatus capable of performing high-quality and high-speed background removal.

[0004]

According to the first aspect of the present invention, an image reading device reads an original with R, G, and B digital signals and obtains K, C, M, and Y recording signals in a frame sequence. In an image processing apparatus having a correction processing unit, at the time of a first scan, a background removal threshold generation unit that detects a background density level of an original by taking a density histogram of an image and stores the detection result as a background removal threshold, and a second scan Thereafter, the background of the image is detected using the stored threshold value, and based on the detection result, the image determined to be the background is detected.
Set the density value of the element to 0 or according to the density value of surrounding pixels
The value that adaptively changes by the darkness of the pixel determined to be the background
A background removing means for reducing the density of the background by subtracting from the degree value .

According to a second aspect of the present invention, there is provided an image processing apparatus having a color correction processing means for reading an original with an R, G, B digital signal by an image reading apparatus and obtaining C, M, Y recording signals in a frame sequential manner. At the time of the first scan,
A background removal threshold generating means for detecting the background density level of the document by taking a density histogram , storing the detection result as a background removal threshold, and after the second scan, using the stored threshold to determine the background of the image. Detected and the detection result
The density value of a pixel determined to be background based on 0
Or a value that changes adaptively according to the density value of surrounding pixels
Is subtracted from the density value of the pixel determined to be the background.
A background removing means for lowering the density of the background.

According to the third aspect of the present invention, prior to the first scan, there is provided a means for setting a predetermined arbitrary threshold value, and the background removing means uses the set threshold value at the time of the first scan to obtain an image. Detects background and detects
Based on the result, set the density value of the pixel determined to be the background to 0
Or adaptively changes according to the density value of surrounding pixels.
Value is subtracted from the density value of the pixel determined to be the background.
This is characterized in that the density of the background is reduced by using.

According to a fourth aspect of the present invention, the background removal threshold generation means determines a background removal threshold in real time according to each pixel value at the time of the first scan, performs the background removal using the threshold, and sets the threshold. It is characterized by memorizing.

[0008] The invention of claim 5 is characterized in that a K-version recording signal is created at the time of the first scan.

According to a sixth aspect of the present invention, in an image processing apparatus having a color correction processing means for reading an original with an R, G, B digital signal by an image reading apparatus and obtaining a recording signal in a frame sequential manner, Image density histogram
Detecting the background density level of the original I, by calculation of the background density level obtained when scanning the previous background density level obtained during scanning to determine the background removal threshold of the next scan, and the determined A background removal threshold generation unit that stores a background removal threshold; and, after the second scan, using the stored background removal threshold for the scan, the background of the image is detected, and the background is detected based on the detection result.
Set the density value of the pixel determined to be skin to 0 or
The value that changes adaptively according to the density of the element
Reduce the background density by subtracting from the specified pixel density value.
And means for removing scalp background.

[0010]

Embodiments of the present invention will be described below. The present invention enables high-quality, high-speed background removal by automatically setting a threshold value with high accuracy without performing prescan. FIG. 1 shows a functional block diagram of an embodiment of a color image processing apparatus to which the present invention is applied. The R, G, and B signals, which are linear with respect to the reflectance, input from the image input device 1 such as a scanner, are logarithmically converted by the scanner γ conversion unit 2 and are linearly density R, G, and B.
Generate a signal. Next, in the color correction unit 3, R, G, B
The signal is converted into complementary colors of Y, M, and C. Undercolor removal (UCR / UCA) is performed in the undercolor removal unit 4 from the Y, M, and C signals to determine the amount of black ink, and Y, M, C, and K signals are generated. In the processing subsequent to the under color removing unit 4, for example, the signal of the K plane is processed in the first scan, the signal of the C plane in the second scan, the signal of the M plane in the third scan, and the signal of the Y plane in the force scan. The background removal unit 5 performs a background removal process using the threshold value set by the background removal threshold value generation unit 8 to reduce the background density, as described later. The printer γ conversion unit 6 performs density conversion processing suitable for the characteristics of the output device, and sends the data to the image output device 7. In the case where the recording signals of only the Y, M, and C planes are obtained in the frame sequence, the generation of the K signal by the under color removing unit 4 becomes unnecessary.

First, the present invention will be described with reference to FIGS. 1, 2, 3 and 4. FIG.
2 shows a block diagram of the background removing unit 5 of FIG. The background removal unit 5 includes a background determination unit 9 and a background density conversion unit 10.
Surface color removal unit 5 lowers the background density level by using the threshold value set in the background removal threshold generator 8 of Fig. 1 (Example
For example, 0 or a value close to 0) . The background removal threshold generation unit 8 automatically detects the background density level and sends the result to the background removal unit 5 as a background removal threshold. In the background removal unit 5, the background determination unit 9 uses the threshold value determined by the background removal threshold generation unit 8, considers the background to be a background when the input pixel data is equal to or less than the threshold value, and sends the determination result to the background density conversion unit 10. send. The background density conversion unit 10 converts the density of the pixel determined as the background by the background determination unit 9 to 0 or a value close to it and outputs it. If the density conversion method of the background is a method of lowering the density, for example, a method of subtracting a value adaptively changing according to the density value of the peripheral pixel from the density value of the pixel determined to be the background or other methods But it doesn't matter.

FIG. 3 is a block diagram of the background removal threshold generator 8 shown in FIG. The background removal threshold generation unit 8 uses Max
(R, G, B) holding unit 15, histogram generation unit 11,
It comprises a background removal threshold register 12. Max (R, G,
B) The holding unit 15 stores Max (R, G,
Hold B). Max (R, G, B) means a value with the highest density among the R, G, B data.
If R has the highest density among the G and B data, Max
(R, G, B) = R (density value). In addition, M described later
conversely, in (R, G, B) means the lowest density value among the R, G, B data. The histogram generation unit 11 first sets the M of the R, G, and B data during the first scan.
For ax (R, G, B), a histogram is taken for the density of all pixels. FIG. 4 shows the result of taking a histogram with the density interval set to 5. The density on the highlight side exceeding a predetermined frequency (300 pixels in FIG. 4) is regarded as the density level of the background. In the case of FIG. 4, the density range is 1
Pixels 0 to 15 are the highlight side where 300 or more pixels appear, and the density 15 is regarded as the density level of the background. Then, the background density level 15 is stored in the background removal threshold register 12 as the background removal threshold. After the second scan, the background removal unit 5 performs the background removal processing using the contents of the background removal threshold register 12 set during the first scan as the background removal threshold.

In FIG . 2, the background determination unit 9 determines R,
When Max (R, G, B) of the pixel value for the G and B data is equal to or smaller than the background removal threshold, the image is determined to be the background, and the determination result is sent to the background density conversion unit 10. The background density conversion unit 10, the pixel value when the determination result is background (K, C, M, Y) or (C, M, Y), respectively, for example, (0,0,0,0),
(0,0,0), and is not converted when the determination result is non-ground. The background density conversion method is not limited to these methods and may be another method as long as the density is reduced. For example, a value (V) that adaptively changes according to the density value of a peripheral pixel.
(k, Vc, Vm, Vy) is subtracted from the pixel value (k, c, m, y) to obtain (k−Vk, c−Vc, m−Vm, y−V).
y). Here, the values of Vk, Vc, and Vy are, for example, average values of the pixel values of 3 × 3 pixels centered on the target pixel. In this way, the background density conversion
In section 10, reduce the density of the background to 0 or a value close to it
You.

Thus, at the time of the first scan,
At the same time that the K, C, M, Y version or the C, M, Y version of the recording signal is obtained, the background removal threshold generation unit 8 automatically sets the threshold used for the background detection after the second scan, and the second 3. The method according to claim 1, wherein the background detection and removal by the background removal unit after scanning uses a threshold set at the time of the first scan.
The invention described in claim 5, wherein a K-version signal is generated at the time of the first scan.

Next, the third aspect of the present invention will be described with reference to FIGS. As described above, the background removal threshold register 1 in the background removal threshold generation unit 8 in FIG.
A storage unit 2 stores a background removal threshold. In the first and second aspects of the present invention, the background removal threshold set in the background removal threshold register 12 at the time of the first scan is used after the second scan. The value of the removal threshold is set in advance in the background removal threshold register 12, and at the time of the first scan, the background removal unit 5
The background removal is performed using the threshold set in the background removal threshold register 12. The background removal threshold used in the first scan is arbitrarily determined. For example, if a value 20 is set in advance in the background removal threshold register 12 as a threshold value, the background removal is performed by the background removal unit 5 at the time of the first scan using the threshold value. The operation after the second scan is the same as that of the invention of claim 1.

Next, the invention according to claim 4 will be described with reference to FIG. FIG. 5 shows a background removal threshold generation unit by real-time processing . The achromatic color determination unit 13, the density level determination unit 14, the Max (R, G, B) storage unit 15, the background removal threshold memory 16, and the address register 17 Become. The background removal threshold used in the first scan is fixed at 20 in the embodiment of FIG. 3, but in the invention according to claim 4, the background removal threshold is determined in real time according to the pixel value. First, whether the target pixel (R, G, B) is an achromatic color or a highlight color is determined by the achromatic color determination unit 1.
3. The determination is made by the density level determination unit 14. The determination as to whether an image is an achromatic color is made by the following conditional expression: Max (R, G, B) −Min
(R, G, B) ≦ 10 That is, Max
When (R, G, B) -Min (R, G, B) is 10 or less, it is determined that the color is achromatic. To determine whether or not it is a highlight,
It is determined by the following conditional expression Max (R, G, B) ≦ 30. That is, when Max (R, G, B) is 30 or less,
Judge as highlight color. As a result of this determination, during the first scan, when achromatic and highlighted, Max
Background removal is performed using the value of (R, G, B) as a threshold, and the value of Max (R, G, B) is stored in the background removal threshold memory 16. The value of the address register 17 of the background removal threshold value memory 16 sets a value indicating the pixel position being processed. For the determination of the background after the second scan, the value held in the memory 16 is used as the background removal threshold.

Next, the invention according to claim 5 will be described with reference to FIGS. In order to improve the reproducibility of chromatic color highlights, in general, black is not applied to the highlight color. The density value of the signal is 0
It is. Therefore, background removal is not required for the K plate. As an embodiment of the invention according to claim 5, a method of generating a K-plate recording signal at the time of the first scan, and generating a K-plate recording signal with a density equal to or less than the background density as 0 in the undercolor removal processing or the printer γ conversion processing explain.

First, K under the background density is removed by removing the undercolor.
A method of generating a plate recording signal as 0 will be described. In the under color removal, when the signal value of the pixel value is (Y, M, C), a K signal is generated by the following calculation. K0 = Min
(Y, M, C), and further, K = f
(K0). f (K0) is a table conversion process for correcting the signal value K0 in the under color removal process. FIG. 6 shows an example of the table f. Here, the background density (here, an arbitrary value of 30) is determined so that the K signal of the highlight color is not generated. It has been converted to become. For the obtained K signal, the value of (Y, M, C) is given by the following equation: Y = y−K, M =
Conversion is performed by calculation of m−K, C = c−K. Next, a description will be given of a method of generating a K-plate recording signal having a background density or less as 0 by the printer γ conversion processing unit. The printer γ conversion processing unit converts the Kin signal into a Kout signal by table conversion Kout = g (Kin).
FIG. 7 shows an example of the table g. In table g, Kin
The signal is the background density (30 as an arbitrary value here)
In the following cases, the Kout signal is converted to 0. That is, when the Kin signal is equal to or lower than the background density (here, 30), Kout is output as 0, and no ink is applied to the background.

Next, the invention according to claim 6 will be described with reference to FIG. FIG. 8 shows the relationship between the histogram of all pixels and the background density level for Max (R, G, B) data at the time of the first scan, the second scan, and the third scan. Each background density level is slightly different due to variations in scanner characteristics. Therefore, for example, when the background removal threshold set at the time of the first scan is used as the background removal threshold after the second scan, the background may not be removed because variations in the characteristics of the scanner are not taken into account. In the example of FIG. 8, the background density level is 15 at the time of the first scan, but the characteristics of the scanner are changed at the time of the third scan, and the background density level is changed to 25. For this reason, if the background removal at the time of the third scan is performed using the background density level 15 set at the time of the first scan, the density level becomes 1
Pixels in the range 5-25 are not removed. According to the sixth aspect of the present invention, in order to set the background removal threshold with high accuracy, the background density level is automatically detected by the same method as that described in the first aspect of the invention during all scans, and the result is stored in a register. For example, a threshold value to be held and used after the second scan is determined by an arithmetic expression shown in Table 1 below.

[0020]

[Table 1]

In Table 1, TH1 indicates a background density level detected during the first scan, TH2 indicates a background density level detected during the second scan, and TH3 indicates a background density level detected during the third scan.

Here, at the time of the second scan, the first claim
Similar to the case described in the invention, the background removal is performed using the background density level TH1 detected during the first scan as the background removal threshold. For example, when TH1 = 15 , in the second scan, TH1 = 15 is used as the background removal threshold. Further, at the time of the third scan, for example, when the background density levels of the first scan and the second scan are TH1 = 15 and TH2 = 20, respectively, the background removal threshold applied at the time of the third scan does not use 15 at the time of the first scan. TH2 + (TH2-
25 of the calculation result of TH1) is used. Similarly, at the time of the force scan, the calculation result of TH3 + (TH3-TH2) is used. In this way, the background removal threshold can be set with high accuracy even when the scanner characteristics vary for each scan.

The present invention is not limited to the above embodiment, but is widely applied to an image processing apparatus, a color copying machine, an apparatus for improving image quality, etc., which removes the background at high speed without performing prescan. be able to.

[0024]

According to the first and second aspects of the present invention,
The background removal threshold can be set with high precision without performing prescan, and high-quality background removal can be performed on recording signals created after the second scan.

According to the third and fourth aspects of the present invention, the background can be simultaneously removed from a recording signal created at the time of the first scan without performing prescan.

According to the fifth aspect of the present invention, the accuracy of setting the background removal threshold is improved by creating a recording signal that does not require background removal at the time of the first scan, and background removal with higher image quality can be performed without performing pre-scanning. Can be.

According to the sixth aspect of the invention, even when the characteristics of the scanner fluctuate during scanning, a highly accurate background removal threshold can be set without performing pre-scanning, and the image quality can be improved.

[Brief description of the drawings]

FIG. 1 is a diagram showing functional blocks of an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a view showing an embodiment of a background removing unit according to the present invention.

FIG. 3 is a diagram showing an embodiment of a background removal threshold generation unit according to the present invention.

FIG. 4 is a diagram showing an example of a histogram at a density interval of 5;

FIG. 5 is a diagram illustrating a background removal threshold generation unit that performs local real-time processing according to pixel values.

FIG. 6 is a diagram showing a table f.

FIG. 7 is a diagram showing a table g.

FIG. 8 is a diagram showing a relationship between all pixels and a background density level.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 image input device 2 scanner γ conversion unit 3 color correction unit 4 under color removal unit 5 background removal unit 6 printer γ conversion unit 7 image output device 8 background removal threshold generation unit 9 background determination unit 10 background density conversion unit 11 histogram generation unit 12 background removal threshold register 13 achromatic determination unit 14 density level determination unit 15 Max (R, G, B) holding unit 16 background removal threshold memory

──────────────────────────────────────────────────の Continuing on the front page (51) Int.Cl. ⁷ identification symbol FI H04N 1/407 H04N 1/04 D (58) Investigated field (Int.Cl. ⁷ , DB name) H04N 1/46-1 / 64 H04N 1/40-1/409 G06T 1/00 510 G06T 5/00 100

Claims (6)

(57) [Claims] An image processing apparatus having a color correction processing means for reading an original with R, G, and B digital signals by an image reading apparatus and obtaining K, C, M, and Y recording signals in a frame-sequential manner. When scanning, the density histogram of the image
What detecting the background density level of the original, and background elimination threshold generating means for storing the detection result as the threshold value for background elimination, since the second scan, using the stored threshold value, detects the background of the image, Based on the detection result,
The density value of the pixel determined to be 0 or a peripheral pixel
Value that changes adaptively according to the density value of
The density of the background by subtracting from the density value of the pixel
The image processing apparatus characterized by comprising a background elimination unit that, the. 2. An image processing apparatus which has a color correction processing means for reading an original with R, G, B digital signals by an image reading apparatus and obtaining C, M, Y recording signals in a frame-sequential manner. And image density histogram
What detecting the background density level of the original, and background elimination threshold generating means for storing the detection result as the threshold value for background elimination, since the second scan, using the stored threshold value, detects the background of the image, Based on the detection result,
The density value of the pixel determined to be 0 or a peripheral pixel
Value that changes adaptively according to the density value of
The density of the background by subtracting from the density value of the pixel
The image processing apparatus characterized by comprising a background elimination unit that, the. 3. Prior to first scan comprises means for setting an arbitrary threshold value determined in advance, background removal means during first scan, using the set threshold, and detects the background of the image, the In the detection result
Based on whether the density value of the pixel determined to be the background is 0
Or a value that changes adaptively according to the density value of the surrounding pixels
By subtracting from the density value of the pixel determined to be the background,
3. The image processing apparatus according to claim 1 , wherein the density of the skin is reduced . 4. The background removal threshold generation means determines a background removal threshold in real time according to each pixel value during a first scan, performs background removal using the threshold, and stores the threshold. The image processing apparatus according to claim 1 or 2, wherein 5. The image processing apparatus according to claim 1, wherein a K-version recording signal is generated during the first scan. 6. The image reading an original by device R, G, read by the digital signal B, the image processing apparatus having a color correction processing means for obtaining a sequential recording signal plane, at each scan, the density histogram of the image taken by detecting the background density level of the original, by a calculation between said background density levels obtained when scanning the previous background density level obtained during scanning to determine the background removal threshold of the next scan,
A background removal threshold generation unit that stores the determined background removal threshold; and, after the second scan, the background of the image is detected using the stored background removal threshold for the scan , based on the detection result. The density of pixels determined to be background
Set the degree value to 0 or adapt according to the density value of surrounding pixels
Is the density value of the pixel determined to be the background color
An image processing apparatus comprising: a background removing unit that reduces the density of the background by reducing the background density .