JP2502928B2 - Image signal processor - 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 signal processing device for performing MTF correction on an image signal.

[0002]

2. Description of the Related Art When reproducing an image read by a scanner in a digital copying machine, a facsimile machine or the like, an MTF for sharpening a reproduced image by sharpening an edge portion of an image signal to correct "blur". The image signal is corrected.

This MTF correction is performed using an MTF filter as shown in FIG. 2 in the main scanning direction and an MTF filter as shown in FIG. 3 in the sub scanning direction. Further, by using the MTF filter as shown in FIG. 4, MTF correction in both the main and sub scanning directions is performed.

FIG. 5 is a waveform diagram for explaining the operation of the MTF filter of FIG. When an MTF filter is applied to an input image signal having a bad rise and fall characteristic (waveform is blunted) as shown in the upper part of the figure, a correction signal component as shown in the middle part of the figure is added to the input image signal. A corrected output image signal having a sharp rise and fall as shown in the lower part of the figure is obtained. The content of this MTF correction is expressed by the following equation.

[0005]

## EQU1 ## B '= (1 + 2a) * B-a * (A + C) = a * {2B- (A + C)} + B

Here, B is a target pixel, A and C are adjacent pixels before and after the target pixel, B'is a correction output, and α is an MTF correction coefficient.

FIG. 6 is a block diagram of a conventional image signal processing apparatus for performing MTF correction in the main scanning direction by the MTF filter as shown in FIG. The image signal processing device (indicated by a broken line) processes a multi-tone digital image signal input from the A / D converter 1.

In the image signal processing device, the latch circuit 2,
Reference numerals 3 and 4 temporarily hold image signal information of three pixels arranged in the main scanning direction. The shift register 5 doubles the information of the target pixel B held in the latch circuit 3 by left-shifting it by one bit, and the adder 6 doubles the adjacent pixels A and C held in the latch circuits 2 and 4. The information of is added. The subtractor 7 has a difference F = the output information D = 2B of the shift register 5 and the output information E = A + C of the adder 6.
2B- (A + C). The multiplier 8 obtains the product H of the MTF correction coefficient α set in the coefficient register 9 and the output information F of the subtractor 7. Adder 10
Holds the sum of B and H, I = B + H. This I
Corresponds to the corrected output B ′ represented by the above (Equation 1).

The clamp circuit 11 is provided to cope with overflow or underflow occurring in the adder 10.
When the value of exceeds FFH (= 255), the final corrected output image signal J is clamped to FFH, and when the value of I becomes negative, it is clamped to J = 0.

[0010]

However, according to such a configuration, when trying to obtain a sufficient MTF correction effect, black or white noise (notch) is likely to occur in the binarized image of the correction output. There was a point. The reason will be described below.

FIG. 7 shows the input image signal and its M in FIG.
It is a waveform diagram of a TF corrected output image signal, a solid line shows a waveform of an input image signal, and a broken line shows a waveform of an MTF corrected output image signal. However, when the MTF correction coefficient α = 1, that is, (-
This is the case where the MTF filters of 1, 3, -1) are used.

As shown in the figure, even in the solid black portion or the solid white portion of the image, the value of the input image signal is changed due to the quantization error of the scanner, the quantization error of the A / D converter, and the uneven density of the original. It fluctuates in the positive or negative direction. Both the positive-direction fluctuation and the negative-direction fluctuation are emphasized by the MTF correction as shown by the broken line.

The MTF correction output image signal is slice level L
When binarized by b (binarized to a normal density), even if such a variation is emphasized, there is almost no problem on the binarized image.

However, in the case of binarization at a high slice level La (in the case of deep binarization), the emphasized negative fluctuations (pixels P1, P2, etc.) appear as black noise on a white background. Further, when binarized at a low slice level Lc (binarized lightly), the emphasized fluctuation in the positive direction (such as the pixel P3) appears as white noise on a black background. Such noise significantly impairs the image quality of the binarized image.

The present invention has been made in view of the above-mentioned problems, and provides an image signal processing apparatus capable of suppressing the generation of noise due to minute fluctuations of an image signal and achieving a good MTF correction effect. The purpose is to

[0016]

In order to solve the above-mentioned problems, the present invention detects a positive or negative edge direction of a pixel to be MTF corrected , and an MTF correction output.
Input the signal corresponding to the slice level for binarization of the image signal
Input means and the signal from this input means.
Setting means for setting the MTF correction coefficient for each direction, and
The setting according to the edge direction detected by the detection means
Selecting means for selecting one of the MTF correction coefficients set by the means .

[0017]

According to the present invention, the strength of MTF correction can be adjusted independently for each edge direction by the above configuration. Therefore, for example, when the binarization slice level of the MTF corrected output image signal is high and the black noise in the white solid portion is conspicuous, the MTF correction coefficient corresponding to the negative direction is set to a small value to reduce the negative of the image signal. By weakening or not emphasizing the effect of directional variation and setting the MTF correction coefficient corresponding to the positive direction edge to a required size, noise in a white solid portion is suppressed and sufficient MTF is obtained.
A correction effect can be achieved. Similarly, when the slice level is low and the white noise in the black solid portion is conspicuous, the MTF correction coefficient in each edge direction is set to the opposite relationship to suppress the noise in the black solid portion and to obtain a sufficient MTF correction effect. I can give you.

[0018]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an image signal processing apparatus according to an embodiment of the present invention, and the broken line portion is the image signal processing apparatus. This image signal processing device performs MTF correction in the main scanning direction similarly to the device shown in FIG. 6, but parts corresponding to those in FIG. 6 are assigned the same reference numerals to avoid duplication of description.

The configuration different from that of FIG. 6 will be described below. Reference numeral 12 is a decoder as a means for setting the MTF correction coefficient for each edge direction. This decorator 12 is an MTF
Depending on the slice level SL for binarization of the corrected output image signal J, the forward MTF correction coefficient α1 as shown in the following table,
The negative direction MTF correction coefficient α2 is output.

[0021]

[Table 1]

Reference numeral 13 is a selector as a means for selecting the MTF correction coefficient according to the edge direction. This selector 13
When the sign output S of the subtractor 7 is positive, the positive MTF correction coefficient α1 is selected as the MTF correction coefficient α and input to the multiplier 8, and when the sign output S is negative, it is negative. The direction MTF correction coefficient α2 is selected as the MTF correction coefficient α and input to the multiplier 8.

As is clear from the explanation of the selector 13, in the present embodiment, the subtractor 7 for performing the operation of {2B- (A + C)} also serves as the edge direction detecting means, and its code output The edge direction detected by S is shown.

The MTF filter used for the MTF correction is the one shown in FIG. 2, but the MTF correction coefficient α is switched to α1 or α2 according to the edge direction as described above. Further, the MTF correction coefficient in each direction is switched and set according to the binarization slice level SL as shown in (Table 1).

The operation of the image signal processing apparatus configured as described above will be described below with reference to the waveform chart shown in FIG.

For example, the input image signal (solid line) shown in FIG.
Since the sign output S of the subtractor 7 becomes negative when P1 of P becomes a target pixel, α2 is selected as α by the selector 13 and MTF correction is performed using this α. here,
When the slice level SL is high as La shown in FIG. 7, since α1 = 1 and α2 = 0 are set as shown in Table 1, the P1 pixel is not corrected (emphasized) and the slice level L
Does not fall below a. The same applies to the pixel P2.
Therefore, pixels such as P1 and P2, which are noise components, are 2
It does not appear as black noise in the white solid part of the binarized image. On the other hand, for pixels on the positive edge, α1 = 1 is selected and emphasized in the same manner as in the related art, so that the MTF correction works well.

The slice level SL is L shown in FIG.
When it is low as c, α1 = 0, as shown in (Table 1),
α2 = 1 is set. In this case, when the pixel P3 becomes the pixel of interest, the sign output S of the subtractor 7 becomes positive and α1
= 0 is selected, so that the slice level Lc is not emphasized.
Does not exceed Therefore, this pixel P3 does not appear as white noise in the solid black portion of the binarized image. Since α2 = 1 is selected and emphasized for the pixel at the negative edge, the same MTF correction effect as in the conventional case can be obtained, and faintness of a thin line or the like can be corrected.

When the slice level L is a normal level as shown by Lb in FIG. 7 and noise in white solid and black solid portions is unlikely to occur, α1 = α2 = 1 is set, and therefore, as in the conventional case. MTF correction is performed in both the positive and negative directions.

In this embodiment, the MTF correction coefficients α1 and α2 for each edge direction are automatically set to three stages by the slice level SL so as to automatically correspond to three types of slice levels, but four stages are set. It is also possible to switch and set the above. Further, when the slice level is fixed, the fixed α1 and α2 corresponding thereto may be set. In this embodiment, the edge direction is detected by the subtractor 7 for MTF correction calculation, but a dedicated circuit may be provided to detect the edge direction.

It is also possible to realize functions such as MTF correction calculation by firmware or software.

Further, although the present embodiment performs MTF correction in the main scanning direction, the MTF filter as shown in FIG. 3 or 4 is used and the edge direction detection in the sub-scanning direction or each of the main and sub directions is performed. It is possible to perform the same MTF correction in the sub-scanning or both main and sub-scanning directions by performing the above procedure and selecting the MTF correction coefficient.

[0032]

As is apparent from the above description, according to the present invention, the MTF correction coefficient is selected according to the edge direction of the image signal in correspondence with the binarization slice level. Suppresses the generation of noise in the solid area, and M
By properly operating the TF correction, the quality of the reproduced image can be improved.

[Brief description of drawings]

FIG. 1 is a block diagram of an image signal processing apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing an MTF filter for the main scanning direction.

FIG. 3 is a diagram showing a sub-scanning direction MTF filter.

FIG. 4 is a diagram showing an MTF filter for both main and sub scanning directions.

FIG. 5 is a waveform diagram for explaining the operation of MTF correction in the main scanning direction.

FIG. 6 is a block diagram of a conventional image signal processing device.

FIG. 7 is a waveform diagram for explaining the relationship between the slice level and the MTF correction effect.

[Explanation of symbols]

2, 3, 4 Latch circuit 5 Shift register 6, 10 Adder 7 Subtractor 8 Multiplier 11 Clamp circuit 12 Decoder (edge direction MTF correction coefficient setting means) 13 Selector (MTF correction coefficient selecting means)

Claims (1)

(57) [Claims] 1. A detection means for detecting whether the edge direction of a pixel to be MTF corrected is positive or negative, and MTF correction output.
Input the signal corresponding to the slice level for binarization of the force image signal.
Depending on the input means to be applied and the signal from this input means.
Setting means for setting the MTF correction coefficient for each direction,
The above-mentioned setting is performed according to the edge direction detected by the detection means.
An image signal processing apparatus comprising: a selection unit that selects one of the MTF correction coefficients set by the setting unit.