JPH07170395A - Area discriminating device and gray level transformation processor - Google Patents

Abstract

PURPOSE:To transform an N-value gray level image into an M-value gary level image (N<M) with an excellent accuracy for both of a contour area and a non-contour area. CONSTITUTION:A first transformation circuit 2 transforms K-value image data to be centralized in the vicinity of the central value of a signal range. An N-ary coding circuit 8 transforms the transformed K-value image into N-ary-coded value data (K>N). An image memory 9 stores the N-ary-coded value. An estimation circuit 93 transforms the N-ary-coded value data into an M-value image (N<M). A second transformation circuit 3 performs an inverse transformation for the M-value image based on the transformed amount by the first transformation circuit 2. Thus, the generated of density ripple is suppressed by even a small scanning opening. A control circuit 92 sets transformation tables to the first transformation circuit 2 and the second transformation circuit 3 by the setting signal from an operation panel 91 and performs the processing modes corresponding to various image and an image adjustment by outputting a control signal to the estimation circuit 93.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an area identification device for identifying a contour area and a non-contour area in an N-value gradation image obtained by converting a K-value gradation image into an N-value gradation image (K> N). And a gradation conversion processing device for converting an N-value gradation image into an M-value gradation image (N <M).

[0002]

2. Description of the Related Art In recent years, printing and display have been performed by multivalued data as an output device such as a printing device and a display device. Therefore, the amount of information is increasing when the image information used for these is handled electronically. Therefore, in a facsimile, a still image file, a copying machine, etc., an original image is converted into a binary signal and transmitted, stored and processed. However, if this binarized signal is reproduced as it is on the output device, the image quality deteriorates. In order to solve this deterioration in image quality, the original image is converted into a binary signal, transmitted, stored, and edited, and when reproduced by an output device, a multi-value gradation signal of the original image is estimated from the binary signal. There is a way.

Conventionally, there are a dither method and an error diffusion method as a digital halftone technique for expressing the gradation of an original image by a binarized signal (pseudo halftone image). Further, as a method for estimating the multi-value gradation signal of the original image from the binarized signal for the pseudo halftone image, there is a halftone estimation method.
(For example, Japanese Patent Laid-Open No. 61-251368) In this halftone estimation method, the halftone is estimated according to the ratio of the number of black and white pixels in the scanning aperture.

Further, the edges in the binary image are detected by one edge detection filter, the contour area / non-contour area / intermediate area is discriminated by the output of the detection filter, and the addition weighting coefficient in the scanning aperture is changed to smooth the edges. The characteristics were variable. (For example, JP-A-4-51378)

[0005]

However, in the above-mentioned conventional configuration, various document images handled by a facsimile, a still image file, and a copying apparatus have different image characteristics on a page-by-page basis, or within one page. There is a problem that the optimum multi-value gradation signal cannot be estimated when the image characteristics are different in different areas.

Further, the gradation of the K-value image is converted into a binary signal (K>
In the digital halftone technique expressed in 2), data is diffused at the time of binarization conversion and further a unique texture is generated. Therefore, if the contour area and the non-contour area are discriminated by only one edge detector, there is a problem that the data is diffused during the binarization process or the texture is affected by the dot arrangement pattern. Was there.

Further, when the scanning aperture is small, the estimated value fluctuates and the error is large. Further, since the facsimile, the still image file, and the copying apparatus handle various target images, there is a problem that desired identification processing and image quality adjustment cannot be performed according to the target image.

The present invention solves the above-mentioned problems of the prior art. Region identification is performed according to the target image without being affected by the diffusion and texture of the binarized data in the contour portion and the non-contour region in the image. A first object of the present invention is to provide an area identification device which performs an optimum gradation conversion in any area of an image, and to provide an excellent gradation conversion processing device that adjusts the image quality according to a target image. This is the second purpose.

[0009]

In order to achieve the first object, the area identification device of the present invention has the following first configuration,
It has a second configuration, a third configuration, and a fourth configuration. In the first configuration, a plurality of filters having the same edge detection characteristics and different positions of interest are compared with respective outputs of the plurality of filters with a predetermined value, and only the outputs having the predetermined value or more are added. And means for determining the output of the adding means at a predetermined determination level and outputting a predetermined determination value.

In the second configuration, in addition to the first configuration, a setting means for setting a plurality of processing modes or image quality adjustment values, and a control means for controlling the predetermined determination level according to the setting contents of the setting means. 1 control means. In the third configuration, a first detection unit that adds outputs of a plurality of first filters having the same edge detection characteristic and different focus positions, and a plurality of first detection units having different edge detection characteristics from the first filter. A second detection unit configured to add the outputs of the plurality of second filters having the same edge detection characteristics and different focus positions, and the output of the first detection unit as the first detection unit. A first determination unit that determines at a determination level and outputs a predetermined first determination value, and an output of the second detection unit at a second determination level, and outputs a predetermined second determination value. And a second determining means for performing the determination.

In the fourth configuration, in addition to the third configuration, setting means for setting a plurality of processing modes or image quality adjustment values, and the first determination value or the first determination value depending on the setting content of the setting means. The 2nd which controls the validity and invalidity of the judgment value of 2
Control means. In order to achieve the second object, in a gradation conversion processing device for converting an N-value gradation image into an M-value gradation image (N <M), the gradation conversion processing device of the present invention is Fifth configuration, sixth configuration, seventh configuration, eighth
The configuration, the ninth configuration, the tenth configuration, and the eleventh configuration.

In the fifth configuration, the pixel of interest and its peripheral pixels in the N-value gradation image are multiplied by a predetermined weighting coefficient and added, and the N-value gradation image is added to the M-value gradation image (N < M)
By changing the scanning aperture size or the weighting factor to be added to the first smoothing filter to be converted into the first smoothing filter, the smoothness is made larger than that of the first smoothing filter, and the N-value gradation image is converted to the M-value gradation image. A second smoothing filter for converting to (N <M), an edge detecting unit for detecting an edge amount in the N-value gradation image, a mixing ratio control unit for calculating a mixing ratio from the edge amount, The mixing processing unit that mixes the first smoothing filter and the second smoothing filter based on the mixing ratio, the identifying unit that identifies the non-contour region and the contour region in the N-value gradation image, and the identifying unit. In the non-contour area, the output of the mixing processing means is selected by the control signal of 1., and in the contour area, the output of the first smoothing filter is selected.

In the sixth configuration, the pixel of interest and its peripheral pixels in the N-value gradation image are multiplied by a predetermined weighting coefficient and added, and the N-value gradation image is converted into the M-value gradation image (N < M)
By changing the scanning aperture size or the weighting factor to be added to the first smoothing filter to be converted into the first smoothing filter, the smoothness is made larger than that of the first smoothing filter, and the N-value gradation image is converted to the M-value gradation image. A second smoothing filter for converting to (N <M), an edge detecting unit for detecting an edge amount in the N-value gradation image, a mixing ratio control unit for calculating a mixing ratio from the edge amount, A mixing processing unit that mixes the first smoothing filter and the second smoothing filter based on the mixing ratio, an emphasis unit that emphasizes the output of the first smoothing filter, and a non-contour in the N-value gradation image. The non-contour area selects the output of the mixing processing means in the non-contour area and the selection means selects the output of the emphasizing means in the contour area in response to a control signal from the identifying means.

In the seventh configuration, the pixel of interest and its peripheral pixels in the N-value gradation image are multiplied by a predetermined weighting factor and added, and the N-value gradation image is converted into the M-value gradation image (N < M)
By changing the scanning aperture size or the weighting factor to be added to the first smoothing filter to be converted into the first smoothing filter, the smoothness is made larger than that of the first smoothing filter, and the N-value gradation image is converted to the M-value gradation image. A second smoothing filter for converting to (N <M), an edge detecting unit for detecting an edge amount in the N-value gradation image, a mixing ratio control unit for calculating a mixing ratio from the edge amount, First processing means for mixing the first smoothing filter and the second smoothing filter based on the mixing ratio; and one row direction or one column in the scanning aperture including the pixel of interest of the first smoothing filter. The output of the first smoothing filter based on the sum of the predetermined weighting factors in the direction and the sum of the predetermined weighting factors in one row direction or one column direction located at the end of the scanning aperture that does not include the pixel of interest. Second process to change A step, an identifying means for identifying a contour area and a non-contour area in the N-value gradation image, and a control signal from the identifying means selects the first processing means for the non-contour area. Selecting means for selecting the second processing means.

In the eighth configuration, the fifth configuration, the sixth configuration, or the seventh configuration further includes setting means for setting a plurality of processing modes or image quality adjustment values, and setting contents of the setting means. And third control means for controlling the first control means or the second control means of the identification means. In the ninth configuration, a plurality of edge detecting means having different characteristics for detecting the edge amount in the input N-valued image data, and an extracting means for mixing outputs of the plurality of edge detecting means to generate an edge extraction amount. , A plurality of spatial filters having different weighting factors or different smoothing regions for smoothing the N-valued image, and a plurality of spaces normalized by the outputs of the plurality of spatial filters to a predetermined level and normalized based on the output from the extracting means. And a mixing processing unit that selects or mixes the outputs of the filters.

In the tenth configuration, the K-valued image data is converted so that the K-valued image data is concentrated in the vicinity of the median of the signal range of the K-valued image data, and the K-valued image data is N converted. N-valued processing means for converting the data into K-valued data (K> N), estimating means for estimating M-valued image data (N <M) from the N-valued data generated by the N-valued processing means, and And second conversion means for performing inverse conversion of the M-value image data (N <M) in conjunction with the data conversion of the first conversion means.

In the eleventh configuration, the tenth configuration further includes setting means for setting a plurality of processing modes or image quality adjustment values, and a plurality of types of conversion tables corresponding to the plurality of processing modes or image quality adjustment values. Then, one of the conversion tables is selected according to the setting contents of the setting means, and the first
Fourthly, resetting to the converting means and the second converting means respectively.
Control means.

[0018]

With this first configuration, the correlation between a plurality of filters can be used by using the combined value of a plurality of filters having the same edge detection characteristics and different focus positions as an edge component, so that the data diffusion, It is possible to accurately distinguish the contour area and the non-contour area in the N-value image by eliminating the influence of the texture.

Further, according to the second configuration, the first control means can control the predetermined judgment level according to the processing mode from the setting means or the set value of the image quality adjustment value.
Different identification areas can be obtained for different images. Further, according to the third configuration, the area can be determined from the determination values of the edge detection filters having different characteristics, so that the area identification can be performed while suppressing the influence of the image characteristics of the target image.

According to the fourth structure, the first control means controls the validity or invalidity of the first determination value or the second determination value according to the processing mode or the setting value of the image quality adjustment value from the setting means. As a result, it is possible to perform area identification according to the characteristics of the target image. Further, according to the fifth configuration, the output of the mixing processing means is selected in the non-contour area by the control signal from the identification means, and one of the smoothing filters mixed by the mixing processing means is selected in the contour area.
By suppressing the influence of erroneous identification, it is possible to perform the gradation conversion ensuring the resolution in the contour area in the N-value image and the gradation conversion ensuring the gradation in the non-contour area.

According to the sixth configuration, the output of the mixing processing means is selected in the non-contour area in accordance with the control signal from the identifying means, and in the contour area, one of the smoothing filters mixed by the mixing processing means is emphasized. Is selected, it is possible to suppress the influence of erroneous identification and perform gradation conversion in which the sharpness is ensured in the contour area in the N-valued image and gradation conversion is ensured in the non-contour area. .

According to the seventh configuration, the first processing means is selected in the non-contour area by the control signal from the identification means, and one of the smoothing filters mixed by the first processing means is used in the contour area. , The data converted output is selected so as to make the scanning aperture equivalently small, so that the effect of erroneous identification is suppressed, and the gradation conversion is performed in the contour area in the N-value image while ensuring the resolution, and in the non-contour area. It is possible to perform gradation conversion that ensures gradation.

Further, according to the eighth configuration, the third control means can control the first control means or the second control means according to the processing mode or the set value of the image quality adjustment value from the setting means. Differentiating the identification region according to the characteristics of the image to perform gradation conversion that secures the resolution or sharpness in the contour region in the N-value image, and performs gradation conversion that secures the gradation in the non-contour region. You can

Further, according to the ninth configuration, the output of a plurality of edge detecting means having different characteristics is mixed according to the image characteristics to generate an edge component amount, and a plurality of spatial filters are mixed according to the edge component amount. It is possible to extract a wide range of edge components with respect to the spatial frequency without changing the value processing state. Further, gradation conversion suitable for various images or image regions can be performed.

According to the tenth configuration, the K-valued image data converted by the first conversion means is converted into N-valued data (K> N), whereby the data diffusion range of the N-valued data is suppressed. Since the M-value image data (N <M) is estimated from the N-valued data, it is possible to suppress the fluctuation of the estimated value even if the scanning aperture is made small. Further, by using a small scanning aperture, the resolution can be secured and the circuit can be realized at low cost.

Further, according to the eleventh configuration, the fourth control means can change the conversion table according to the processing mode or the setting value of the image quality adjustment value from the setting means, and can perform the image quality adjustment corresponding to various images. As described above, according to the present invention, an area identification device capable of performing area identification according to a target image without being affected by the diffusion of binarized data and the texture in the contour portion and the non-contour area in the image. Can be realized. Further, it is possible to realize a gradation conversion processing device that performs optimum gradation conversion in any area in an image and further performs image quality adjustment according to the target image.

[0027]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An area identification device and a gradation conversion processing device according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a gradation conversion processing device including an area identification device A according to the first embodiment of the present invention.

In FIG. 1, the gradation conversion circuit 1 converts N-value gradation image data 100 into M-value gradation image data 200. (However, N <M.) The emphasis circuit 5 emphasizes the N-value gradation image 100 or the estimated value NA (signal 300) output from the gradation conversion circuit 1, and emphasizes the emphasis signal EF.
(Signal 500) is generated.

The data conversion circuit 7 receives the estimated value NA (signal 300) output from the gradation conversion circuit 1, executes data conversion processing in pixel units, and outputs a signal 700.
This data conversion process increases the sharpness of the estimated value NA (signal 300). The discriminating circuit 4 of the region discriminating apparatus A discriminates the discriminating signal 40 for discriminating the contour region and the non-contour region of the N-value gradation image.
0 (signal SE) is output.

The selection circuit 6 selects the input A to the input D according to the identification signal 400 and the control signal 61 (CONT1) and outputs the selection signal 600. With this choice,
Signal 200 (Xout), signal 300 (estimated value NA),
Either the signal 500 (EF) or the signal 700 is output as the selection signal 600. Regarding the gradation conversion processing device including the area identification device A configured as described above, the gradation conversion circuit 1, the identification circuit 4, the data conversion circuit 7, the enhancement circuit 5, and the selection circuit of the area identification device A are sequentially selected. The circuit 6 will be described.

First, FIG. 2 shows a block diagram of the gradation conversion circuit 1. The gradation conversion circuit 1 has N pixels per pixel.
The value gradation image is scanned, and the N value gradation image is converted to the M value gradation image (N
<M). The components of the gradation conversion circuit 1 will be described below. The first multi-value conversion circuit 11 includes a first addition circuit 111 and a first normalization circuit 112.
The first addition circuit 111 receives the N-value gradation image data 100 and performs addition processing on each of the pixel of interest and its peripheral pixels according to the set weighting coefficient. The first normalization circuit 112 converts the addition result signal SA obtained from the first addition circuit 111 into an estimated value NA based on the sum of the gradation number N and the weighting coefficient.

The second multi-value conversion circuit 12 comprises a second addition circuit 121 and a second normalization circuit 122. The second addition circuit 121 inputs the N-value gradation image data 100,
Addition processing is performed on each of the pixel of interest and the peripheral pixels according to the set weighting coefficient. Second normalization circuit 12
2 is an estimated value N based on the sum of the gradation number N and the weighting coefficient of the signal SB of the addition result obtained from the second addition circuit 121.
Convert to B.

The edge extraction circuit 19 detects an edge component in the N-value gradation image. The mixing circuit 13 uses the estimated value N obtained from the first normalization circuit 112 and the second normalization circuit 122.
A, the estimated value NB is mixed by the weighting coefficients Ka and Kb from the first weighting control circuit 134, and the mixed signal X
To calculate. The quantization circuit 14 requantizes the mixed signal X to convert it into an M-value gradation image. Here, the estimated value N
A. Even when the estimated value NB is converted from a real value of 0 to 1 and mixed conversion is performed so that 0 ≦ mixed signal X ≦ 1, the quantization circuit 14 still has [X · (M−1) +0.5. ], It is possible to requantize to the required number M of gradations. If the quantizing circuit 14 is composed of a rewritable memory such as a RAM, the number of gradations M required in the display device and the printer device can be met by requantization. Further, when the estimated value NA and the estimated value NB are converted from 0 to an integer value of (M−1) and mixed conversion is performed so that 0 ≦ mixed signal X ≦ (M−1), the quantization circuit 14 [X + 0.
5] is performed and the maximum value is set to M−1 and requantized to the required number of gradations M.

The operation of the gradation conversion circuit 1 configured as described above will be described below with reference to FIGS. 2, 3, 4, 5, and 24. In the first embodiment of the present invention, N,
M is not specified, but for example N = 2, M = 2
The following description is given as 56. FIG. 3A shows a filter diagram used in the second addition circuit 121, and FIG. 3B shows a filter diagram used in the first addition circuit 111.

Although the aperture size of the filter and the weighting coefficient of the filter are not specified, in the first embodiment of the present invention, the first adder circuit 111 has the 3 × shown in FIG. 3B.
A filter 1110 having a size of 3 and a weighting factor is used to generate the signal SA. Further, the second adder circuit 121 uses the filter 1210 having the size of 5 × 5 and the weighting coefficient shown in FIG. 3A to generate the signal SB. Let P (i, j) be the value of the pixel of interest and W be the weighting factor of the pixel of interest.
When (i, j), the signal SA = ΣΣW (i, j) ·
The calculation of P (i, j) is performed at all pixel positions in the scanning aperture area, and the calculation results are all added. The same applies to the signal SB. Therefore, the signal SA and the signal SB are calculated as in (Equation 1) and (Equation 2), respectively. Here, · is an operator indicating multiplication.

[0036]

[Equation 1]

[0037]

[Equation 2]

The first adder circuit 111 has a filter 111 having a size of 3 × 3 and a weighting coefficient shown in FIG. 3B.
When 0 is used to generate the signal SA, it can be realized by the circuit shown in FIG. The input N-value image data 100 is
The delay circuit 113 and the latch circuit 117 develop in the main scanning direction i and the sub scanning direction j. The image data of 9 pixels expanded in the 3 × 3 area is multiplied by the weighting coefficient from the weighting coefficient register 115 by the multiplier 114, and all the multiplication results are added by the adding circuit 116 to obtain the signal SA.
The addition result is M (= 17) value image data of "0" to "16". The multiplier 114 has a weight coefficient register 115.
If the weighting factor from 2 is 2n, it can be realized by a bit shift operation, and the circuit can be simplified. In the first adder circuit 111 and the second adder circuit 121, when the number of gradations of the input image data is n and the sum of weighting coefficients of the added regions is Sn, the maximum output value Mx is (Equation 3). Is calculated as follows.

[0039]

[Equation 3]

In the first adder circuit 111, if the sum of weighting factors is Sa and the maximum value of the output value SA is Ma, then M
It becomes a = (n-1) * Sa. In the first embodiment of the present invention, n = 2, and since the first addition circuit 111 performs addition processing based on the filter 1110 of FIG. 3B, Sa = 16, Mb from the weighting coefficient of each pixel. = 16.

Similarly, in the second adder circuit 121, if the sum of the weighting factors is Sb and the maximum output value SB is Mb, then Mb = (n-1) * Sb. In the first embodiment of the present invention, n = 2, and the second adder circuit 121 is shown in FIG.
Since the addition processing is performed based on the filter 1210 of (a), Sb = 25 and Mb = 25 from the weighting coefficient of each pixel.

Next, the first normalization circuit 112 and the second normalization circuit 122 will be described. The first normalization circuit 112,
The second normalization circuit 122 normalizes the input value SGx to the output value SCx based on (Equation 4), assuming that the input value is SGx and Mx obtained from (Equation 3) is the maximum value of the input values. . [] Is an integerization process that rounds down the fractional part.

[0043]

[Equation 4]

Equation (4) may be used as an arithmetic circuit, or CP
It is also possible to use a look-up table (hereinafter simply referred to as LUT) in which the correspondence from SGx to SCx is calculated in advance by U, DSP, etc. based on (Equation 4) and the calculated value is tabulated. By the above-described (Equation 3) and (Equation 4), the input value SGx as the addition result is M
Normalized to the value gradation image data. Specifically, the input value SA of the first normalization circuit 112 is normalized to the estimated value NA by (Equation 5), and the input value SB of the second normalization circuit 122 is changed to the estimated value NB by (Equation 6). Normalized. Here, the maximum value of the input value SA is Ma, and the input value S
The maximum value of B is Mb. In addition, [] is an integerization process that rounds down the fractional part.

[0045]

[Equation 5]

[0046]

[Equation 6]

In the first embodiment of the present invention, for example, Ma =
16, Mb = 25, and M = 256, output NA for input SA and output NB for input SB, respectively.
Are calculated and output as image data of 256 gradations to the mixing circuit 13. In the first adder circuit 111 and the second adder circuit 121, the value of the weighting coefficient of the filter used may be not only a positive numerical value but also a negative numerical value or zero value. Furthermore, not only an integer but also a real number may be used. For a pixel having a coefficient of zero, the input value is multiplied by "0", and the result is added as "0". If the addition result is negative, it is processed to "0", and the maximum value is processed to a numerical value based on (Equation 3).

The weighting factor and the added area size (scanning aperture) are a predetermined value and a predetermined size determined at the design stage, and are requested from the N-value gradation image data and the M-value gradation image data. Image quality, that is, the spatial filter characteristics required. Therefore, although different area sizes are used in the first embodiment of the present invention, they may have the same size. The degree of smoothness of the input image data (smoothness characteristic of the spatial filter) can be changed by changing the weighting factor.

Next, the mixing circuit 13 will be described with reference to FIGS. FIG. 5 shows an operation explanatory diagram of the mixing circuit 13. The weighted mixing coefficients Ka and Kb indicate the weighting characteristics of a and c shown in FIG. 5, and the edge extraction circuit 19
Is determined based on the signal Eout from The estimated value NA shown in FIG. 2 is multiplied by the coefficient Ka by the multiplier 131, and the estimated value NB is multiplied by the coefficient Kb by the multiplier 132. The results of the multiplications are added by the adder 133 and the mixed signal X is output.

Estimated value NA (signal 300), estimated value NB,
When the edge amount Eout (signal 181) is input and the generated mixed signal is X, the mixing circuit 13 generates the mixed signal X based on (Equation 7). Here, MAX is the maximum value output by Eout (signal 181).

[0051]

[Equation 7]

Mixed signal X mixed based on (Equation 7)
Is subjected to integer processing by the quantization circuit 14 and finally converted into M-value gradation image data. For example, M = 256
In case of, the image data Xout (signal 2
00) is generated. Thus, the edge amount Eout
The (signal 181) causes the mixing circuit 13 to weight the plurality of input signals (estimated value NA, estimated value NB) and perform a mixing process to generate interpolated image data.

As the edge amount Eout increases, the mixing circuit 13 increases the mixing ratio of the estimated value NA to the estimated value NB. This is an operation to reduce the smoothness and tends to maintain the resolution. On the contrary, the edge amount Eout
As becomes smaller, the mixing circuit 13 increases the mixing ratio of the estimated value NB over the estimated value NA. This is an operation for increasing the smoothness, and the input image data is processed to be smoother. Therefore, the smoothness is improved in a region such as a photograph region where the edge change is small.

As shown in FIG. 5, the coefficient Ka,
Instead of using a and c having a linear change characteristic as the coefficient Kb, a coefficient showing a binary change characteristic by MAX / 2 like b and d may be used. This is a selection operation, and is optimal for processing characters and the like. If the first weighting control circuit 134 is composed of a LUT of RAM or ROM,
It is possible to realize coefficient manipulation of arbitrary change characteristics. However, Ka
Make sure that + Kb = 1.

As described above, the output of the plurality of different adder circuits (spatial filters) can be continuously changed by the edge amount Eout by the interpolation processing in the mixing circuit 13. As a result, the degree of smoothing processing can be locally and continuously changed according to the edge component of the input image data, and the input image data can be converted into a predetermined multilevel level. The multi-value conversion processing that adaptively preserves edges can achieve both the resolution of the character / line drawing area and the smoothness of the photographic area. In particular, in a region in which characters and photographs are mixed, since binary switching processing is not performed, image deterioration due to erroneous identification can be suppressed to a very small level.

In the first embodiment of the present invention, two inputs are mixed, but at least two or more inputs may be used, and the same effect can be obtained. Further, if the area size of addition (scanning aperture) or the types of addition circuits having different weighting factors are increased and mixed processing is performed, more continuous gradation image data can be generated. Furthermore, it is possible to easily perform simultaneous (parallel) weighting operations on three or more inputs, and it is possible to perform accurate interpolation processing. Further, although the mixing circuit 13 performs linear weighting, it may perform non-linear weighting.

Next, regarding the edge extraction circuit 19, FIG.
This will be described with reference to FIG. FIG. 4 is a filter diagram used in the edge extraction circuit 19, and FIG. 8 is a spatial frequency characteristic diagram of the filter used in the first embodiment. The edge extraction circuit 19 has
First-order differential filters of filters 1900 and 1901 shown in FIG. 4 are used. The filter 1901 detects edge components in the main scanning direction i, and the filter 1900 detects edge components in the sub scanning direction j. Therefore, by combining the absolute value of the addition value obtained by each coefficient of the filter 1900 and the absolute value of the addition value obtained by each coefficient of the filter 1901, the main scanning direction i in the N-value image
And the edge component in the sub-scanning direction j can be extracted.
As shown in FIG. 8, a filter 1900 and a filter 19
01 has a spatial frequency characteristic having a peak at f0. Therefore, the edge component centered on a specific frequency in the N-value image is extracted. The extracted edge component has the maximum value MAX.
Clipped to the mixing circuit 13 and the edge amount Eout is input to the mixing circuit 13.
It is output as (signal 181).

Here, the target image is various designs,
Since it is composed of photographs and characters, it is usually difficult to specify the spatial frequency of the edge component. Therefore, the effect can be expected by extracting the edge component centered on a specific frequency when the characteristics of the image can be grasped or limited, and the edge component of the desired region to be detected as the contour is extracted. Can be extracted.

However, in general, in a FAX, a document file, a word processor, a copying machine, etc., an arbitrary target image different in a plurality of pages is handled, and further, even in the image of one page,
Patterns, photographs and letters are mixed. In order to extract the edge component of the expected contour area in each pattern area, photograph area, and character area on one page, it is necessary to prepare a plurality of edge detectors having different spatial frequencies and handle them. Furthermore, since it is difficult to specify the spatial frequency of the edge component, it is necessary to adaptively use a plurality of edge detectors.

An embodiment in which the edge extraction circuit 19 is improved so as to adaptively correspond to edge components having different spatial frequencies will be described below. FIG. 6 is an improved block diagram of the edge extraction circuit 19. In the improved block diagram of FIG. 6, the edge extraction circuit 19 includes a first edge detection circuit 15,
The second edge detection circuit 16, the mixing circuit 17, and the clipping circuit 18 are included.

The first edge detection circuit 15 and the second edge detection circuit 16 detect the edge component at the pixel position of interest in the image. The mixing circuit 17 includes a first edge detection circuit 15 and a second edge detection circuit 15.
Edge component signal E obtained from the edge detection circuit 16
a and the signal Eb are mixed by the weighting coefficients Kc and Kd from the second weighting control circuit 174 to obtain the edge amount Eg.
Is output. Edge component Ea of the first edge detection circuit 15
Is multiplied by the coefficient Kc by the multiplication circuit 171, and the edge component Eb of the second edge detection circuit 16 is calculated by the multiplication circuit 1
It is multiplied by 72 with the coefficient Kd. The multiplied results are added by the adder 173, and the edge amount Eg is output. In addition, the mixing circuit 17 does not use an arithmetic circuit, a RAM,
It may be a ROM LUT.

The coefficients Kc and Kd are Kc = Ea / (Ea + E
b + Cn), Kd = 1-Kc, and the edge amount E
g is calculated by Eg = Kc × Ea + Kd × Eb. The constant Cn may be a small value that allows the required accuracy as compared with Ea and Eb. For example, C = 0.5.

The clipping circuit 18 quantizes the edge amount Eg input from the mixing circuit 17, clips the maximum value to MAX, and outputs the clipped edge amount Eout to the mixing circuit 13. 7A is a filter diagram used for the first edge detection circuit 15, FIG. 7B is a filter diagram used for the second edge detection circuit 16, and FIG. 23 is a circuit block diagram of the secondary differential filter 1500.

The edge component Ea is generated by using the secondary differential filter 1500 shown in FIG. 7A, and the absolute value of the result of addition with the weighting factor shown is the edge amount Ea. In addition, the edge component Eb is 1 shown in FIG.
Generated using the second derivative filters 1601, 1602,
The absolute values of the respective results added by the weighting factors of the illustrated filters are added to obtain the edge amount Eb.

FIG. 23 is a circuit block diagram of the second derivative filter 1500. The delay circuit 150 delays the N-valued image data 100 in the sub-scanning direction j in the line memory.
The data delayed in the sub-scanning direction j is delayed in the main scanning direction i by the latch, and the adder circuit 152, the adder circuit 153, and the adder circuit 15 are respectively depending on the coefficients of the filters used.
9, the addition circuit 160 adds. The multiplier 154 multiplies the output of the adder circuit 153 by four, and the multiplier 157 multiplies the output of the adder circuit 159 by -1. The adder 155 is the multiplier 1
The edge component is output by adding 57 and the output of the multiplier 154. The absolute value circuit 156 outputs the edge component Ea of the pixel of interest by calculating the absolute value of the adder 155. FIG. 23 shows a circuit for realizing a secondary differential filter, which can be adjusted to the coefficient of the secondary differential filter used by increasing or decreasing the delay latch in the main scanning direction i and the delay line memory in the sub scanning direction j. it can.

The spatial frequency characteristics of the edge component Ea and the edge component Eb are different f1 and f2 as shown in FIG.
The spatial frequency characteristic has a peak at, and the peak frequencies have a relationship of f1> f2. Further, the relationship of f1>f0> f2 is established for the peak frequencies of the filters 1900 and 1901 shown in FIG. The mixing circuit 1 receives the outputs of the different edge components Ea and Eb.
7 controls to adaptively increase the component ratio of the output of the edge component which becomes larger. As a result, edge components in a wider frequency range can be extracted as compared with the case of detecting mainly at a specific frequency of f0, and edge components of contour regions with different conditions such as photo contours, character contours, and graphic contours can be extracted. It can be detected adaptively. In FIG. 7A, a second-order differential filter is used to generate the edge component Ea.
In (b), the first-order differential filter is used to generate the edge component Eb, but any detector that detects the set spatial frequency may be used. Also, the same effect can be obtained by mixing the outputs of three or more edge detectors.

By thus mixing the edge components having different characteristics, it is possible to eliminate the switching of binary processing and to adaptively extract a wide range of edge components in the image in terms of spatial frequency. If this edge amount Eout is used in the mixing circuit 13, the scanning aperture mixing process can be performed in accordance with the local image edge characteristic, and the complete mixing process including the edge component can be performed. That is, the letters,
Even if images with different characteristics such as figures and photos are mixed,
The estimation process completely eliminates the binary switching process,
In addition, it is possible to automatically adapt to the characteristics of the local image edge. Therefore, the optimum edge component extraction for each area and the estimation processing according to the edge component can be executed, so that the influence of local misidentification of the image can be suppressed to a small extent and the image deterioration can be suppressed to a very small extent. Furthermore, the operation of specifying the image to be handled becomes unnecessary. Since the operation for specifying the image requires a processing time, the speed of the processing can be increased by using the edge extraction circuit 19 shown in FIG. Further, the optimum processing can be executed even for an image whose image cannot be specified, such as a received image such as a fax.

Next, the operation of the discriminating circuit 4 of the area discriminating apparatus A will be described with reference to FIGS. 8, 9, 10, 11 and 12. FIG. 9 is a block diagram of the identification circuit 4, FIG.
0 (a) is a layout diagram of the detection positions of the edge detection filter 41 and the edge detection filter 42, FIG. 10 (b) is a filter diagram of the edge detection filter 41, and FIG. 10 (c) is a filter diagram of the edge detection filter 42. It is a thing. Although N and M are not specified in the first embodiment of the present invention, the following description will be made assuming that N = 2 and M = 256, for example.

In FIG. 9, the edge detection filter 41 is composed of the first detector A (41a) to the ninth detector A (41i).
Each of the detectors and an adder 410 that adds the outputs of the detectors (41a to 41i), and generates an addition signal ESA. As shown in FIG. 10A, nine detectors (4
1a to 41i), when the target pixel position processed by the gradation conversion circuit 1 is the position E, the peripheral amounts A to I including the position E are detected as the respective detection positions, and the edge amount is detected. The same filters are used for the detectors (41a-41i). For example, to extract the edge component, 1 shown in FIG.
Using the secondary differential filters 4100 and 4101, the absolute values of the respective results added by the weighting factors of the illustrated filters are added to obtain the edge amount.

FIG. 22 shows the nth edge detection filter 41.
It is a circuit block diagram which realizes one of the detectors A (41a-41i). Here, n is 1-9. Delay circuit 4
A line memory 18 delays the N-value image data 100 in the sub-scanning direction j. The data delayed in the sub-scanning direction j is delayed in the main scanning direction i by the latch, and the adder circuits 412a to 412 are respectively depending on the coefficients of the filter used.
It is added at d. The adder 413 calculates the difference value in the sub-scanning direction j, and the absolute value circuit 415 calculates the absolute value in the adder 413 to determine the edge component (filter 4) in the sub-scanning direction j.
100 edge components) are output. In addition, the adder 414
Calculates the difference value in the main scanning direction i, and the absolute value circuit 416 calculates the absolute value of the adder 414 to output the edge component in the main scanning direction i (edge component of the filter 4101). The adder 417 outputs the edge component of the pixel of interest by adding the edge component in the main scanning direction i and the edge component in the sub scanning direction j. FIG. 22 shows a circuit for realizing a first-order differential filter. By increasing or decreasing the delay latch in the main scanning direction i and the delay line memory in the sub-scanning direction j, it is possible to match the coefficient of the first-order differential filter used. it can.

Similarly, the edge detection filter 42 adds nine detectors of the first detector B (42a) to the ninth detector B (42i) and the adder 420 for adding the outputs of the detectors (42a to 42i). And is configured to generate an addition signal ESB. As shown in FIG. 10A, nine detectors (42a-
42i), when the target pixel position processed by the gradation conversion circuit 1 is the position E, the edge amount is detected with the peripheral positions A to I including the position E as respective detection positions. Detector 42a-
The same filter is used for 42i. For example, for the edge component extraction, the first-order differential filter 4 shown in FIG.
200 and 4201, the absolute values of the respective results added by the weighting factors of the illustrated filters are added to obtain the edge amount.

As shown in FIG. 8, the edge detection characteristic shows a spatial frequency characteristic having different peaks. The peak frequencies of the filters 4100 and 4101 are set to FS.
If a and the peak frequencies of the filters 4200 and 4201 are FSb, the relationship of FSa> FSb is established. Therefore, the edge detection filter 41 is suitable for detecting the contour of a pattern having a high frequency such as characters and figures, and the edge detection filter 42 is suitable for detecting the contour of a low frequency such as a photograph which is not affected by the texture and halftone dot frequencies. Further, by using the added value of the plurality of detectors as the edge amount, a large output is obtained when the detected values of the plurality of detectors are large and the correlation is high. Therefore, edge detection can be performed without being affected by the diffusion of data caused by the binarization process or the unique texture.

In FIG. 9, the first comparison circuit 43 compares the signal ESA with the comparison level CPA (signal 64), and outputs ES.
When A> CPA, “H” is output, and when ESA ≦ CPA, “L” is output. Similarly, the second comparison circuit 44 compares the signal ESB with the comparison level CPB (signal 65), and outputs E
When SB> CPB, “H” is output, and when ESB ≦ CPB, “L” is output. Here, "H" is determined to be a contour area, and "L" is a determination level determined to be a non-contour area.

The AND circuit 45 receives the control signal 62 (signal C
When ONT2) is "H", the output of the first comparison circuit 43 is valid, when it is "L", it is invalid and the output is always "L". Similarly, the AND circuit 46 controls the control signal 63.
When the (signal CONT3) is “H”, the second comparison circuit 44
Output is valid, invalid when "L", and output is always "L".

The OR circuit 47 outputs the identification signal 400 of "H" when either of the outputs of the AND circuit 45 and the AND circuit 46 is "H", and outputs "L" when both are "L". . Here, "H" is determined to be a contour area, and "L" is a level determined to be a non-contour area. Thus, the first
The comparison circuit 43 and the second comparison circuit 44 have a comparison level CP.
The detection signals of the signals ESA and ESB obtained by adding the edge signals detected at a plurality of positions are determined using A and CPB as parameters. As shown in FIG. 8, the comparison level CP
The area of the identification a is detected by A, and the comparison level CPB
The area of identification b is detected by. Therefore, it is possible to control the frequency region identified by the parameters of the comparison levels CPA and CPB.

The control signal 62 controls validity / invalidity of the determination result of the first comparison circuit 43, and the control signal 63 controls validity / invalidity of the determination result of the second comparison circuit 44. result,
It is possible to detect the contour area of the target image with different optimal conditions such as the contour of the photographic image, the contour of the character image, the contour of the graphic image, the contour of the halftone image, the contour of the mixed character / photo image, and control the detection characteristics. .

Further, the identification circuit 4 shown in FIG.
As shown in FIG. 1, nine detectors (41a to 41i)
Output and the value of the register 411 are compared by nine comparators (41
j to 41r) and the values in the register 411 or more may be added by the adder 410. Similarly, the outputs of the nine detectors (42a to 42i) and the value of the register 421 are compared with each other by the nine comparators (42j to 42r), and the values of the register 421 and above are added by the adder 42.
You may improve so that it may add by 0.

The effect improved by this improvement will be described below with reference to FIG. 12 (a), FIG. 12 (b), and FIG.
This will be described with reference to (c) and FIG. 12 (d). Figure 12 (a)
3 is a pattern diagram of the binarized data 100. Area 2000 indicates a character / graphic area, and area 20
An area 2200 inside 00 is an example in which the data is diffused by the binarization process and the image data of the character / graphic is moved. This data movement does not occur with electronic data drawn by a computer. However, when an original is read by an input from a scanner or the like, it often occurs when the original density is low or when the original is faint. Further, the area 2100 in the non-contour area is an isolated point that often occurs due to the binarization processing when the gray value of the K-value image data is small. This is because, assuming that the K-value image data has a constant gray value, in digital halftone techniques such as the error diffusion method and the dither method, the binarization processing is performed so that the gray values are uniformly stored in a certain wide area. Because.

FIG. 12B shows the edge detection filter 41.
Of these, the fifth detector A (41
FIG. 6 is a diagram illustrating an output diagram when an edge of an image is detected by using one e). The fifth detector A (41e) detects the edge of the image in the main scanning direction and the sub-scanning direction by using the filter 4100 and the filter 4101, and detects the edge at the target pixel position E by combining the respective outputs. To do. The implementation circuit is the same as in FIG. The area is identified using the output data of this edge detection. In the digital halftone technique such as the error diffusion method and the dither method, even if the data before binarization is a constant value, the data is diffused by the binarization process, so the edge output Eg does not become "0". Therefore, considering the influence (misidentification) of the isolated point area 200 in the non-contour area and setting the identification level to be “3” or more, the area 3000 surrounded by the dotted line in FIG. 12B.
Can be detected as a contour area. As shown in FIG. 12B, although the isolated point area 2100 is not erroneously detected, the character / graphic area 2000 is not detected and the outer area is detected. Further, the detection area 3000 is the area 2 where the data is moved by the binarization processing shown in FIG.
Under the influence of 200, the region 3000 is identified in an isolated state. When the determination level is “2” or higher, the identification area is expanded, but the isolated state of the area 3000 cannot be solved. Further, there are more false detection areas around the isolated point area 2100.

FIG. 12C shows the edge detection filter 41.
FIG. 6 is a diagram showing an output diagram of the edge output ESA of FIG. The edge detection filter 41 includes the first detector A (41a) to the ninth detector.
Nine detectors of detector A (41i) and detectors (41a-
41i) is configured by an adder 410 for adding the outputs,
Since the addition signal ESA is generated, each detector (41a to 4a-4
When the correlation of 1i) is large, a larger edge component is output, and when the correlation is small, the output of the edge component does not increase. Therefore, in the non-contour region, the influence of isolated points can be suppressed, and in the contour region, the region can be identified without being affected by the diffusion of data. Here, if the value of the determination level CPA of the first comparison circuit 43 shown in FIG. 9 is set to “8”, the area 3000 surrounded by the dotted line in FIG. 12C can be detected as the contour area. Although the isolated point area 100 is erroneously detected, a wide area including the periphery of the character / graphic area 2000 is detected as one connected state. In addition, the detection area 3000 is not affected by the area 2200 in which the data is moved by the binarization processing illustrated in FIG. Here, the value of the judgment level CPA is "15".
By setting to, it is possible to extract and identify only the character / graphic area 2000. In this way, the decision level CP
By changing the setting of the value of A, it is possible to change the extraction area and change the identification area.

FIG. 12D shows an output diagram of the edge output ESA of the improved edge detection filter 41 shown in FIG. By this improvement, since the edge component is targeted for addition at a predetermined level or higher, the influence of the isolated point can be eliminated, and the false detection of the isolated point in FIG. 12C is improved. Here, the comparison value (REG1) is set to "1", and
The edge outputs Eg of the nine edge detectors (41a to 41i) shown in FIG.
, And when Eg> 1, the adder 410 outputs the edge output E
Add g. Furthermore, the first comparison circuit 4 shown in FIG.
When the value of the judgment level CPA of 3 is set to "8",
A region 3000 surrounded by a dotted line 2 (d) can be detected as a contour region. There is no erroneous detection of the isolated point area 2100, and a wide area including the periphery of the character / graphic area 2000 is detected as one connected state. In addition, the detection area 3000
Is not affected by the area 2200 where the data is moved by the binarization processing shown in FIG.
Therefore, as compared with FIG. 12C, it can be seen that the detection accuracy of the edge detection filter 41 is improved. The same processing can also improve the detection accuracy of the edge detection filter 42. Here, by setting the value of the determination level CPA to "12", only the character / graphic area 2000 can be extracted and identified. In this way, by changing the setting of the value of the determination level CPA, the area to be extracted is changed,
The identification area can be changed.

From the above, by adding the edge amounts detected by a plurality of edge detection filters from the peripheral positions including the pixel position of interest and comparing them with a predetermined level, the character / graphic area, which is the contour area, can be accurately measured. Can be detected. Furthermore, by adding a process of comparing the edge amount detected by the edge detection filter with a predetermined value and adding a value equal to or more than a predetermined value, the influence of data diffusion caused by the binarization process can be eliminated, and Can eliminate erroneous detection due to texture in the binarized image. As a result, the amount of erroneous detection in the non-contour region can be reduced and the edge amount of the contour region can be added, so that the edge detection accuracy can be further improved.

The identification circuit 4 shown in FIGS.
Identifies a contour area and a non-contour area in an N-value gradation image, and therefore uses different gradation conversion processing (processing for converting an N-value image into an M-value image [N <M]) in each identification area. be able to. Therefore, by using the optimum gradation conversion processing for each of the contour area and the non-contour area, it is possible to realize the gradation conversion processing that is compatible with the resolution and the uniformity for the entire N-value gradation image.

In the vicinity of the contour area, the n × m scanning aperture of the gradation conversion process is the character / graphic area 20 of the contour area.
00 and the estimated value NA of the intermediate state from the binarized data including the non-contour area. Since this intermediate region is a region where resolution is prioritized in terms of image quality, it is necessary to identify it as a contour region. To determine the position where the n × m scanning aperture includes the character / graphic area 2000 as the contour area, the character / graphic area 2
It is necessary to detect a wide area including the periphery of 000 in a single connected state. For this reason, the identification circuit 4 illustrated in FIG. Since it is detected in the connected state, it is very compatible with the gradation conversion processing.

Next, regarding the data conversion circuit 7, FIG.
This will be described with reference to FIGS. 14 and 15. 13 is a block diagram of the data conversion circuit 7, and FIG. 14A is a data conversion circuit 7.
FIG. 14B is a filter diagram for explaining coefficient setting of FIG.
Is a data conversion curve diagram, and FIG. 15 is an operation explanatory diagram of the data conversion circuit 7. Here, N = 2, M = 17
Further, the normalization coefficient of the first normalization circuit 112 shown in FIG. 2 will be described as 1 (SA = NA).

In FIG. 13, the input estimated value NA
(Signal 300) is a signal generated by the first multi-value conversion circuit 11 shown in FIG. The filter 1110 illustrated in FIG. 3B is used for the first addition circuit 111 of the first multi-value conversion circuit 11. The coefficient setting circuit 71 of the data conversion circuit 7 sets the data conversion coefficient in the coefficient setting register 72. The data conversion coefficient is calculated based on the weighting coefficient of the filter 1110 used in the first addition circuit 111 of the first multi-value conversion circuit 11. The calculation method will be described later.

The estimated value NA (signal 300) output from the first multi-value conversion circuit 11 is applied to the register 7 by the subtractor 77.
The value “4” of 2a is subtracted, and the subtraction result is multiplied by the value “2” of the register 72c by the multiplier 78 to obtain the signal 780. The comparator circuit 73 sets the output signal 730 to “H” level when the input condition is A> B, and sets it to the “L” level when A ≦ B. The selection circuit 75 selects the value "16" of the register 72d when the output signal 730 of the comparison circuit 73 is at "H" level, and the output signal 78 when it is at "L" level.
0 is selected and the output signal 750 is output. Comparison circuit 74
Sets the output signal 740 to the “L” level when the input condition is A <B, and sets it to the “H” level when A ≧ B. Removal circuit 76
Outputs the output signal "0" when the signal 740 is "L" level.
When it is at “H” level, the output signal 750 is output as the output signal 700.

Here, a method of calculating the values of the registers 72a to 72d set in the coefficient setting register 72 will be described with reference to FIG. In FIG. 14A, the filter 1110 will be described as an example. If the weighting factors at each pixel position are W11 to W33, W11, W13,
W31, W33 are "1", W12, W21, W23, W
32 is "2" and W22 is "4". Although the weighting factor is not specified, the effect is greater as the weighting factor at the pixel position of interest is larger than the surrounding pixel positions. Furthermore, if the coefficients are arranged so as to be symmetrical in the main scanning direction i and the sub-scanning direction j, it becomes independent of the direction, which is preferable in terms of image quality.

When the value of the register 72a is reg (a), the value of the register 72b is reg (b), the value of the register 72c is reg (c), and the value of the register 72d is reg (d), each register value is It is calculated as in 8).

[0090]

[Equation 8]

In reg (a), the weighting factors at the upper ends of the row lines are added, but the weighting factors on any one of the upper ends of the row lines, the lower ends of the row lines, the left ends of the column lines, and the right ends of the column lines may be added. The sum of weighting factors is set in reg (d). The calculation of (Equation 8) is performed by previously removing from the estimated value NA an estimated estimated value obtained from pixel positions located in the end row in the row direction or the end column in the column direction of the scanning aperture. Since the data conversion is performed in the direction of reducing the equivalent, the sharpness is improved. Even if the scanning aperture size becomes large, a similar effect can be obtained because the estimated amount obtained from the pixels located in the end row in the row direction or the end column in the column direction of the scanning aperture is deleted from the estimated value in advance. .

By the above data conversion, the input estimated value NA (signal 300) is converted by the data conversion curve shown in FIG. 14 (b) and output as the output signal 700. FIG. 15 shows an estimated value NA (signal 300) and an estimated value NA (signal 300) when the binary image data is moved in the main scanning direction i by the filter 1110 of FIG. 14A.
It is a drawing comparing the signals 700 processed by the data conversion circuit 7, and it can be seen that the sharpness (resolution) is improved.

By using this converted signal 700 as the input D of the selection circuit 6 shown in FIG. 1, the sharpness (resolution) of the contour area identified by the identification circuit 4 can be improved. Further, even if the erroneous identification is present in the non-contour area, a sharp density change does not occur, so that the image quality deterioration can be suppressed to a small level. The register 72 of the data conversion circuit 7
When the respective values (72a to 72d) set to are fixed, the data conversion circuit 7 is replaced with a ROM, and a data conversion table corresponding to the data conversion coefficient is stored therein to perform the same data conversion. good. Further, each value (72a to 72) set in the register 72 of the data conversion circuit 7 is set.
When d) is variable, the data conversion circuit 7 is replaced with a RAM, a coefficient corresponding to the weighting coefficient of the filter 1110 is calculated, a conversion table corresponding to the conversion curve is created again, and the created conversion table is stored in the RAM. It may be set again.

Next, the emphasizing circuit 5 will be described with reference to FIGS. FIG. 16 is a block diagram of the emphasis circuit 5. In FIG. 16, the delay circuit 50 is composed of a line memory of 3 lines and delays the estimated value NA (signal 300) in the sub-scanning direction j. The delayed data is delayed by the delay circuit 51 in the main scanning direction i. The delay circuit 51 is composed of a latch. The multiplication circuit 52 multiplies the value P (i, j) at the pixel position of interest to be emphasized by a factor of 5. Adder circuit 5
4 is a pixel value P (i-
1, j-1), P (i + 1, j-1), P (i-1, j)
+1) and P (i + 1, j + 1) are added. Multiplication circuit 5
6 multiplies the output of the adding circuit 54 by -1. Adder circuit 53
Adds the outputs of the multiplication circuit 52 and the multiplication circuit 56. The absolute value circuit 55 calculates the absolute value of the output of the adding circuit 53,
The enhancement signal EF (signal 500) is output.

This emphasizing circuit 5 uses the estimated value NA (signal 30
The image 0) is emphasized at each pixel position of interest, that is, in pixel units. Therefore, when the emphasis position of interest is P (i, j), the emphasis signal EF is calculated by (Equation 9). However, ABS [] is an operation that takes an absolute value.

[0096]

[Equation 9]

Also, a similar enhancement signal EF (signal 500)
Can also be generated without delaying the estimated value NA (signal 300) in a line memory of 3 lines. in this case,
A position (i, j) including the target pixel position (i, j) to be emphasized
j), position (i-1, j-1), position (i + 1, j-
1), position (i-1, j + 1), position (i + 1, j +)
The estimated value NA is calculated using the first multi-value conversion circuit 11 of FIG. 2 with the five points 1) as estimated positions, and the estimated values NA (i, j), NA at each calculated position are calculated. (I-1,
j-1), NA (i + 1, j-1), NA (i-1, j)
+1) and NA (i + 1, j + 1) are substituted as pixel values corresponding to the positions of (Equation 9) to obtain the N-value gradation image 1
The enhancement signal EF (signal 500) can be obtained directly from 00. If the enhancement signal EF is obtained directly from the N-value gradation image 100, the capacity of the line memory of the three lines of the delay circuit 50 can be significantly reduced. Since N = 2 and M = 256 in the first embodiment of the present invention, the delay of the signal 300 is 8
Bit / pixel is performed, whereas N-value gradation image 100
Is 1 bit / pixel. If the number of delay lines is the same, the line memory capacity is 1/8. N-value gradation image 1
In the method of directly obtaining the enhancement signal EF from 00, since the five points are estimated as the positions of interest, the scale of the first multi-value conversion circuit 11 is five times as large. However, 7/8 of the reduced memory capacity
In terms of the gate conversion number corresponding to 5 and the gate conversion number of the scale of the first multi-value conversion circuit 11 which is 5 times, it is the gate conversion number that the enhancement signal EF (signal 500) is obtained directly from the N-value gradation image 100. Is small. This is advantageous in making an LSI, and has an effect of speeding up.

By using the enhancement circuit 5 described above, it is possible to obtain an enhanced image of the estimated value NA with a high sharpness. By using this enhancement signal 500 (signal EF) for the contour area in the N-value gradation image, it is possible to suppress deterioration of the sharpness of the edge. Further, by emphasizing only the contour area, an M-value gradation image can be obtained without deteriorating the smoothness of the non-contour area. As a result, the edge is saved in the contour area, and the M-value gradation image (N <N
M) can be obtained.

Next, the selection circuit 6 will be described with reference to FIG. The selection circuit 6 controls the control signal 61 (signal CONT
1) and the identification signal 400 (signal SE), input A
(Signal 200), input B (signal 300), input C (signal 200), and input D (signal 500 or signal 700) are selected, and a selection signal 600 is output.

The conditions to be selected are as shown in (Table 1). Here, the selection signal 600 is OUT. Further, the “H” level of the signal 400 (signal SE) represents the result of determination as a contour area and the “L” level represents a non-contour area.

[0101]

[Table 1]

From Table 1, the control signal 61 (signal CON
T1) has a signal level of 3 steps. Therefore, the selection circuit 6 makes the following selection by the control signal 61. When the signal level is "0", input A is selected for the non-contour area and input B is selected for the contour area. When the signal level is "1",
Input C is selected for the non-contour area and input D is selected for the contour area. When the signal level is "2", the input B is selected for the non-contour area and the input D is selected for the contour area.

Therefore, in the "photograph mode", since the gradation is emphasized, the processing is performed using the input A and the input B, and the outline area is not emphasized. In "text / photo mode",
Since the resolution and the gradation are emphasized, the processing is performed using the input C and the input D to emphasize the contour area. Further, in the "character mode", since the resolution is emphasized, the processing is performed using the input B and the input D to emphasize the contour area.

As described above, the "processing mode" shown in (Table 1) is set by the setting operation from the operation panel 91 described later, and the control signal 92 (signal C
By setting ONT1), it is possible to execute a conversion process from an N-value gradation image that is optimal for various target images to an M-value gradation image (N <M). Further, the output signal 500 of the emphasis circuit 5 that performs the emphasis process may be replaced with the output signal 700 of the data conversion circuit 7. The data conversion circuit 7 equalizes the openings by previously removing, from the estimated value NA, an estimated estimated value obtained from pixel positions located in the end row in the row direction or the end column in the column direction of the scanning opening. The sharpness is improved because the data conversion is performed in the direction in which it is made smaller.

As described above, according to the first embodiment, it is possible to obtain an M-value gradation image (N <M) with less edge deterioration in the contour area and good smoothness in the non-contour area. A gradation conversion processing apparatus according to the second embodiment of the present invention will be described below with reference to the drawings. FIG. 17 is a block diagram of a gradation conversion processing device according to the second embodiment of the present invention.

In FIG. 17, the first conversion circuit 2 is a RAM.
And a control signal 9 from the control circuit 92.
The K-value image data is converted by the conversion table set via 20 (signal TAB1). N-value conversion circuit 8
Converts the K-value image data from the first conversion circuit 2 into N-value gradation image data (K> N).

The image memory 9 is composed of, for example, a semiconductor memory or a hard disk device and the like.
The value gradation image data is stored. The estimation circuit 93 is the gradation conversion processing device of the first embodiment of the present invention shown in FIG. 1, and the N-value gradation image data 100 from the image memory 9 is converted into M
It is converted into value gradation image data 600 (N <M).

The second conversion circuit 3 is composed of a LUT such as a RAM, and the control circuit 92 outputs a control signal 921 (signal TAB).
The M-value gradation image data is converted by the conversion table set via 2). The control circuit 92 outputs a control signal to the estimation circuit 93, the first conversion circuit 2, and the second conversion circuit 3 according to the setting signal from the operation panel 91.

The operation panel 91 is composed of, for example, a plurality of keys, and sets the resolution level of the image, the "photograph mode",
The setting signal is output according to the contents set in the control circuit 92 in accordance with the setting of the input switches such as the "character / photo mode", the "character mode", the "character / graphic mode" and the "halftone mode". The operation of the gradation conversion processing device configured as described above will be described below.

First, an embodiment of converting a K-value image into an N-value image (K> N) in the N-value conversion circuit 8 will be described with reference to FIG. A conversion method for performing the N-value conversion processing includes a dither method and an error diffusion method. In the second embodiment of the present invention, for example, the K-value image data is converted into the N-value image data by the error diffusion method shown in FIG. Convert to. In FIG. 21, the K-value image data fxy is added by the adder 81 to the integrated error value Err around the pixel of interest, and the output f′xy is output. This f'xy is quantized into an N value by being compared with a plurality of preset threshold values in the N-value conversion processing circuit 82, and the output gxy is output. The adder 83 is exy =
The calculation of f′xy−gxy is performed, and the quantization error exo is distributed around the pixel of interest with a certain weighting coefficient, and the error memory 86
Memorized in. The multiplier 84 multiplies the error value e distributed around the pixel of interest by the weighting coefficient of the weight matrix Wij and outputs the integrated error value Err. Signal 801 is E
rr = ΣΣe (x + i, y + j) × Wij. For example, if N = 2, gxy is 2 of “0” and “255”.
The value data is output, and the normalization circuit 87 outputs "0", "25".
Binary data of "5" is normalized to binary data of "0" and "1". As a result, the K-value image data is normalized to the N-value image data.

When the halftone image generated by the error diffusion method shown in FIG. 21 is used as the input image of the estimation circuit 93,
It is possible to perform estimation and restoration in which deterioration of the edge portion is small, and resolution and gradation are compatible without pseudo contours. Although the error diffusion method is used here as the N-value conversion process for changing the K-value gradation to the N-value gradation (K> N), the N-value conversion circuit 8 is specified in the second embodiment of the present invention. However, the process may be any process that saves the average data value in the image. Therefore, the processing may be performed by a known N-value conversion (K> N) means, for example,
The minimum average error method, the dither method, the multivalued dither method, or the like may be used.

Next, the first conversion circuit 2 and the second conversion circuit 3 will be described with reference to FIGS. 17, 18, 19, and 20. FIG. 18A shows the first conversion circuit 2 and the second conversion circuit 3.
18B is a relationship diagram between the scanning aperture position and the black pixel position, FIG. 19 is an explanatory diagram of data conversion, and FIG.
Reference numeral 0 is an explanatory diagram of the conversion table. In the second embodiment of the present invention, N, M and K are not specified, but N =
2, M = 256 and K = 256. Further, in order to facilitate the description, the estimated value NA of the first multi-value conversion circuit 11 in the gradation conversion circuit 1 shown in FIG. 2 will be described as an example. In addition, the filter 11 used in the first adder circuit 111
The estimation operation will be described assuming that the 10 scanning apertures are 2 × 2 and the addition weighting factors in the apertures are all 1. Therefore, the first addition circuit 111 sets the pixel value at the pixel position of interest to P
Assuming that (i, j) and the weighting coefficient are W (i, j), the signal SA is SA = ΣΣW (i, j) · P (i, j). Therefore, (Equation 3) is SA = P (i , J) + P (i + 1,
j) + P (i, j + 1) + P (i + 1, j + 1). Here, · is an operator indicating multiplication. Also, Sa
= 4 and Ma = 4. Accordingly, the normalization of the first normalization circuit 112 follows Ma (4) and M (2) according to (Equation 5).
It is calculated by substituting 56, and the estimated value NA is output.
Since (M-1) / Ma is about 64, one black pixel counted in the area of the scanning aperture 2 × 2 has 64 estimated data values.

First, the case where the conversion curve 3 shown in FIG. 18A is set in the first conversion circuit 2 and the second conversion circuit 3 will be described. The conversion curve 3 has a relationship in which the input signal Xin and the output signal Yout are Yout = Xin. this is,
It is a conversion table in which an input signal is directly used as an output signal. The estimation circuit 93 estimates an M-value gradation image (N <M) from the N-value gradation image. If K = M, the N-value conversion circuit 8
The K-value image input to will be estimated. The N-valued circuit 8 converts the grayscale data of the pixel of interest into binary data of “0” and “255” and diffuses the generated error to peripheral pixels. Binary data of "0" and "255" are "0" and "1"
And is stored in the image memory 9. By this binary quantization processing, 8 bits / pixel is converted to 1 bit / pixel, and the data capacity is compressed to 1/8. This compressed image has a bitmap structure and has a fixed compression rate. Therefore, as compared with the compression by the variable length coding, the compression time is constant, which is suitable for real time processing. Furthermore, since the image data in the image memory 9 has a bit map structure, it is not necessary to decompress and then edit the compressed image as in the case of a compressed image by variable-length coding, and therefore superimposition editing can be performed at high speed. . An output device that expresses "0" of this binary data as white dots and "1" as black dots, or
If "1" of binary data is output to a display device that expresses white dots and "0" as black dots, a digital halftone image due to the density of dots can be obtained.

From the binary data of “0” and “1” stored in the image memory 9, the M-value gradation image is converted into the first multi-value conversion circuit 1.
Estimate at 1. For example, the K-value image data having a constant value "16" is normalized by the N-value conversion circuit 8 to be "0" and "1".
18B is converted into a white pixel and “1” is represented by a black pixel, “0” and “1” having a distance of Dx in the main scanning direction and a distance of Dy in the sub-scanning direction as shown in FIG. 18B. Is converted into binary data. The first multi-value conversion circuit 11 estimates the M-value gradation image by adding the data values in the scanning aperture and performing a normalization process. When the scanning aperture is 2 × 2 and the weighting coefficients for addition are all 1, M = 256 and Ma = 4, and the black dot “1” is converted into a data value of approximately 64 by (Equation 5).

When the K value data is small, that is, in the highlight area of the image, the scanning aperture size is smaller than the diffusion range of black pixels (“1” of binary data), and the estimated value NA varies depending on the scanning position. . At the scanning position A illustrated in FIG. 18B, the estimated value is 64, the estimated value is 0 at the position B, and the estimated value is 64 at the position C. For K value image data with a constant value of "16",
The estimated value is not constant and fluctuates. Even in the shadow area, the scanning aperture size becomes smaller than the diffusion range of white pixels (“0” of binary data), and the estimated value NA also fluctuates.

In order to improve the fluctuation of the estimated value NA in the highlight area and the shadow area, the conversion curve 1 shown in FIG. 18A is used as the first conversion circuit 2 and the conversion curve 2 is used as the second conversion circuit 3. Set. A conceptual explanation of the conversion process is shown in FIG.
This will be described with reference to (a), FIG. 18 (b) and FIG. 18 (c). Here, the input data has a constant value of 16, that is, FIG.
The conversion processing of K-value image data at point H in (a) will be described. In addition, in FIG. 18 (b) and FIG. 18 (c),
Binary data "0" is shown as a white pixel and "1" as a black pixel.

As shown in FIG. 18A, the input data "16" is quadrupled by the conversion curve 1 of the first conversion circuit 2 by the conversion curve 1, and the output data of "64" (point B). ) Is converted to. Also, the K-value image data “64” that has been quadrupled is “0” by the N-value conversion circuit 8.
Converted to "1". Conceptually, since the data is quadrupled, Dx / in the main scanning direction as shown in FIG.
2, "0" and "1" having a distance of Dy / 2 in the sub-scanning direction
Is converted into binary data. At the scanning position A shown in FIG. 18C, the estimated value NA is 64, and at the position B, 6
4 and 64 at position C. However, since the K-value image data is quadrupled by the conversion curve 1, it is quadrupled (point E) by using the conversion curve 2, 16 at the position A, and 1 at the position B.
6 and 16 at position C. As a result, compared to FIG. 18A, it is possible to suppress the fluctuation of the estimated value NA and obtain an accurate estimated value.

Point G in FIG. 18A is the median of the signal range of the K-value image data, and is set at the position of 128 when the signal range of 0 to 255 is set. In the first conversion circuit 2, conversion is performed so that the K-value image data is larger than the straight line AG in the highlight area below the point G (for example, a curve passing through the point ABG), and in the shadow area above the point G from the straight line GD. Convert to smaller values (eg, point GC
Curve through D). As a result, the diffusion range of black pixels (binary data “1”) in the highlight region or the diffusion range of white pixels (binary data “0”) in the shadow region is reduced. In the second conversion circuit 3, the first conversion circuit 2
Inverse conversion processing of the conversion processing performed in (1) is performed. That is, in the second conversion circuit 3, the curve AB in the highlight area below the point G
The conversion value by G is inversely converted by the curve passing through the AEG point, and the conversion value by the curve GCD in the shadow region at the point G or higher is inversely converted by the curve passing through the GFD point.

As described above, when the K-value image data is converted by the first conversion circuit 2 so as to be concentrated near the median of the signal range and the estimated value NA of the M-value data is inversely converted by the second conversion circuit 3, The fluctuation of the estimated value NA is small in the highlight area or the shadow area. The outline of the conversion process has been described above, and the details will be described with reference to FIGS.
(B), FIG. 19 (c), FIG. 19 (d), FIG. 19 (e),
This will be described with reference to FIG.

FIG. 19A shows an example of K-value image data. In the description, the maximum value of the K-value image data is 255 and the minimum value is 0. In the embodiment, the constant value "16", which is approximately 1/16 of the maximum value 255 of the K-value image data, is the target data to be processed. FIG. 19 (b)
Is converted into 2 by the N-value conversion circuit 8 from the K-value image data of FIG.
It shows an example of the binarized data 100 when the binarization process. As shown in FIG. 19B, binarized data in which one “1” exists in the region A (4 × 4) is obtained in an average state.

FIG. 19C shows the output of the estimated value NA when the binarized data 100 of FIG. 19B is estimated by the first multi-value conversion circuit 11. The average value in the area A is "16", and the K-value image data is estimated with the average value in the unit of area A, but the variation of the estimated value NA is 0 at each internal pixel position. It is as large as 64.

On the other hand, FIG. 19D shows K of FIG. 19A.
The value image data “16” is converted four times by the conversion curve 1 (conversion 1), and “64” which is approximately ¼ of the maximum value 255 of the K value image data is binarized by the N-value conversion circuit 8. 1 shows an example of binarized data 100. As shown in FIG. 19D, in the average state, "1" is set in the area A.
Is converted into binary data in which there are four.

FIG. 19 (e) shows the output of the estimated value NA when the binarized data 100 of FIG. 19 (d) is estimated by the first multi-value conversion circuit 11. Next, FIG.
(F) shows the estimated value NA of FIG.
This is the output that is inversely converted to 1/4 in (Conversion 2). Here, the reason for multiplying by 1/4 is the K input to the N-value conversion circuit 8.
This is because the value image data is quadrupled by the conversion curve 1 (conversion 1), and thus it is necessary to perform the inverse conversion to restore the state before conversion. In FIG. 19F, within the area A, the estimated value NA does not change at each pixel position, and the K-value image data “16” is accurately estimated.

Here, the effect of the conversion will be described. Compared with the estimated value NA of FIG.
FIG. 19F in which conversion is performed by the conversion circuit 2 and the second conversion circuit 3
As for the estimated value NA, the fluctuation of the estimated value is obviously suppressed, and an accurate estimated value is obtained. This conversion process is particularly effective in a highlight area or a shadow area where binarized data is likely to diffuse.

As described above, the K-value image data is converted by the first conversion circuit 2 so as to be concentrated near the median of the signal range, and the converted K-value image data is converted into N-valued data (K> N).
By converting to N-value,
Alternatively, the data diffusion range of white dots in the shadow area can be suppressed. When the estimated value NA of the M-valued data (N <M) obtained from this N-valued data (K> N) is inversely transformed by the second conversion circuit 3, the estimated value changes in the highlight area or the shadow area. Suppressed and accurate estimated value NA
Can be obtained. This corresponds to controlling the diffusion range of the data that occurs due to the binarization processing of the N-valued circuit 8, and the white dot “0” or the black dot “1” is set in the area where the distance from the target pixel position is small. By suppressing the diffusion range of the data, it is possible to obtain the estimated value NA in which the local fluctuation is suppressed.

In the conversion operation of the first conversion circuit 2 and the second conversion circuit 3, the data diffusion range of the white dot “0” or the black dot “1” is set in the highlighted area or the shadow area in the pixel of interest. Since the distance from the position is suppressed to be small, it is possible to suppress the fluctuation of the estimated value even if the scanning aperture is set small. By reducing the scanning aperture, the circuit scale can be reduced and the resolution can be increased. If the estimated value NA (M-value image data) with a small variation is output to the output device or the display device, a high-quality image with little density unevenness can be obtained.

Although the estimation circuit 93 is the gradation conversion processing device shown in FIG. 1, the gradation conversion circuit 1, the first multi-value conversion circuit 11 or the multi-value conversion circuit 1 shown in FIG. 2 is used.
It may be 2. Any device that converts an N-value gradation image into an M-value gradation image (N <M) by estimating the multi-valued level from the pixel value within a predetermined scanning aperture is effective. Further, the fluctuation amount of the estimated value NA is determined by the conversion amount of the first conversion circuit 2 and the scanning aperture size of the filter 1110 used in the first addition circuit 111. When the aperture size is constant, the larger the conversion amount is, the more effective it is. However, the straight line BC in FIG.
Of the edge component is suppressed, and the detection amount of the edge component is suppressed. This has a great influence on a device that controls the aperture size estimated by the detected edge component. For example, in the case of characters / figures, it is preferable that the variation of the straight line BC is large.

Therefore, in order to correspond to the image to be processed,
An operation key for setting the "processing mode" is provided on the operation panel 91, and a setting signal is sent to the control circuit 92 by setting the operation key.
Output to. The control circuit 92 has a built-in conversion memory and stores a plurality of conversion tables. The conversion table signal 92 is sent according to the "processing mode" set from the operation panel 91.
The conversion table of the first conversion circuit 2 and the second conversion circuit 3 is reset via 0 and the signal 921. As illustrated in FIG. 20, the conversion table for image quality adjustment is stored in, for example, the conversion memory. Characters that emphasize edge components
When the “figure mode” is set, the table of the conversion curve 3 is read from the head address of the LUT 3, and the first conversion circuit 2 is sent via the signal 920 to the second conversion circuit 2 via the signal 921.
The conversion table is reset in the conversion circuit 3.

Similarly, "photograph mode" in which smoothness is important
Is set, the conversion curve 1 starts from the start address of LUT1.
Table is read out, the table of the conversion curve 2 is read out from the head address of the LUT 2 via the signal 920, and the conversion table is read out to the second conversion circuit 3 via the signal 921. Reset. As shown in FIG. 20, by storing a plurality of types of conversion tables in the conversion memory, it is possible to deal with various images and adjust the image quality.

Further, the estimation circuit 93 can deal with various images by controlling the control signals 61 to 65 from the control circuit 92, and the image quality can be adjusted. The operation will be described below with reference to FIG. 9, (Table 1) and (Table 2).

[0131]

[Table 2]

By the operation shown in (Table 1), the control signal 61
The (signal CONT1) selects each estimated output such as the presence / absence of emphasis processing (or the presence / absence of data conversion) according to the “processing mode”. The control signal 62 (signal CONT2) and the control signal 63 (signal CONT3) control the identification signal 400 shown in FIG. Control signal 62 (signal C
The ONT 2) controls the AND circuit 45 shown in FIG. 9 to control the validity and invalidity of the result determined by the first comparison circuit 43. The control signal 63 (signal CONT3) controls the AND circuit 63 shown in FIG.
Control the validity and invalidity of the result determined in 4. It is valid at "H" level and invalid at "L" level. Comparison level C
The PA, comparison level CPB, manipulates the discriminating level. The larger the comparison level value (CPA, CPB), the larger the edge detection level is required, and the identification area is limited.
Conversely, the smaller the comparison level value (CPA, CPB),
Since the identification is performed with a small edge detection level, the identification area is widened. The optimum control value for the "processing mode" is set using the above control signals.

In the "photograph mode", since the smoothness is important, the identification process is not performed. Therefore, CONT2 is set to "L" and CONT3 is set to "L". In this setting, the comparison level values CPA and CPB are invalid. In the "character / photo mode", since both the resolution of the character area and the smoothness of the picture area are important, the identification processing is performed. However, in consideration of erroneous identification due to a halftone image in which the spatial frequency is distributed over a wide range, the edge detection filter 4 having a high identification frequency
The identification signal using 1 is invalidated, and the identification signal using the edge detection filter 42 whose identification frequency is lower than the halftone dot frequency and whose misidentification due to the halftone dot frequency is small is enabled. Therefore, CONT2 is set to "L" and CONT3 is set to "H". In this setting, the value of the comparison level CPA is invalid and the value of the comparison level CPB is valid.

In the "character mode", since the resolution of the character area is emphasized, the identification processing is performed. Further, in order to obtain the identification signal in which the spatial frequency is detected in a wide range, the result obtained by using both the edge detection filter 41 and the edge detection filter 42 is validated. Therefore, CONT2 is "H", CONT
Set 3 to "H". With this setting, the comparison level CP
A, the value of the comparison level CPB functions effectively. further,
In order to widen the edge detection area, the value of the comparison level CPB is changed from B1 to B2 (B1> B2).

Further, by providing a plurality of image quality setting keys on the operation panel 91 and performing the setting operation, the control circuit 92 sets the comparison level values CPA and CPB in a plurality of setting steps,
Image quality is adjusted by controlling the identification area. As described above, according to the second embodiment, by controlling the control signals 61 to 65, the estimation circuit 93 can deal with various images and can adjust the image quality. Therefore, an N-value gradation image is
When converting to a value gradation image (N <M), optimum conversion processing corresponding to various images is performed, and desired image quality adjustment is also performed at the time of conversion. A toned image (N <M) can be generated.

As described above, according to the first and second embodiments of the present invention, by providing the identifying means for detecting the contour area by the addition of the plurality of edge detections, the K-value image is N The contour region and the non-contour region in the image can be accurately discriminated without being affected by the data variation due to the diffusion of the data when the value image (K> N) is converted. Further, since the contour area in the N-valued image has a means for executing the estimation processing with emphasis on sharpness and the non-contour area with emphasis on gradation, it is possible to perform the estimation processing in any of the contour area and the non-contour area in the N-valued image. It is possible to estimate a high-quality M-value image (N <M) even in the area of.

Further, there is provided means for performing conversion for concentrating the K-value image data in the vicinity of the median of the signal range, and by converting the converted K-value image data into N-valued data (K> N), The diffusion range of N-valued data can be suppressed.
By estimating this N-valued data by the estimating means, there is an effect that the variation of the estimated value can be suppressed even if the scanning aperture is made small. The processing with a small scanning aperture can ensure the resolution and can be realized at low cost.

Further, since there is a means for mixing the outputs of a plurality of edge detecting means having different characteristics in accordance with the image characteristics to generate an edge component quantity and mixing a plurality of spatial filters according to the edge component quantity, it is a binary value. It is possible to extract a wide range of edge components with respect to the spatial frequency without changing the processing state. Therefore, gradation conversion suitable for various images or image regions can be performed.

Further, the N-value gradation image is converted into the M-value gradation image (N <
When the image is converted into M), it has a control means for controlling so as to perform an optimum conversion process corresponding to various images, and has a means for executing a desired image quality adjustment at the time of conversion. It is possible to generate a quality M-value gradation image (N <M).
Further, the present invention can be applied to a copying machine, a printer, a display device, a FAX,
By applying it to a document file or the like, it is possible to reduce the image data capacity, realize storage and data transfer, and obtain a high-quality estimated image at the output and display stages. further,
If the present invention is applied not only to a monochrome image but also to a color output device, a color display device, and a color FAX, the image data capacity can be further reduced while maintaining the image quality. The color can be realized by performing the same processing as black and white on each color of the RGB image or the YMC image.

Further, the area identification device and the gradation conversion processing device in the first and second embodiments of the present invention are all CPUs,
Alternatively, it may be realized by arithmetic processing (software processing) using a DSP. Further, when the gradation conversion processing device of the present invention is used as compression / expansion processing of image data, fixed-length coding and compression / expansion processing maintaining a bitmap structure can be realized. Fixed-length encoding is optimal for a device that performs real-time processing because it performs compression / decompression in a fixed time. Furthermore, the compression that maintains the bitmap structure enables overwriting and editing at any position in the image recording device, and does not require recompressing and editing the compressed image once unlike the variable-length coding method. It is possible to speed up the editing process.

[0141]

As described above, according to the present invention, when the K-value image is converted into the N-value image (K> N) by having the identifying means for detecting the contour area by the addition of the plurality of edge detections, There is an effect that the contour area and the non-contour area in the image can be accurately distinguished without being affected by the data variation due to the diffusion of the data.

Further, the contour area in the N-valued image emphasizes sharpness, and the non-contoured area has means for executing the estimation processing with emphasis on gradation. There is an effect that a high-quality M-value image (N <M) can be estimated in any of the areas. Further, it has means for performing conversion for concentrating the K-value image data in the vicinity of the median of the signal range, and by converting the converted K-value image data into N-valued data (K> N), N
The diffusion range of the digitized data can be suppressed. By estimating this N-valued data by the estimating means, there is an effect that the variation of the estimated value can be suppressed even if the scanning aperture is made small. The processing by the small scanning aperture has the effects that the resolution is secured and the processing can be realized at low cost.

Further, since the output of a plurality of edge detecting means having different characteristics is mixed according to the image characteristics to generate an edge component amount, and a means for mixing a plurality of spatial filters according to the edge component amount is provided, it is a binary value. It is possible to extract a wide range of edge components with respect to the spatial frequency without changing the processing state. Therefore, there is an effect that gradation conversion suitable for various images or image regions can be performed.

Further, the N-value gradation image is converted into the M-value gradation image (N <
When the image is converted into M), it has a control means for controlling so as to perform an optimum conversion process corresponding to various images, and has a means for executing a desired image quality adjustment at the time of conversion. There is an effect that a quality M-value gradation image (N <M) can be generated.

[Brief description of drawings]

FIG. 1 is an area identification device A according to a first embodiment of the present invention.
It is a block diagram of a gradation conversion processing device including.

FIG. 2 is a block diagram of the gradation conversion circuit 1 in FIG. 1 according to the first embodiment of the present invention.

FIG. 3A is a filter diagram (1210) used in the second adder circuit 121; (B) Filter diagram used in the first adder circuit 111 (111
0).

FIG. 4 is a filter diagram used in an edge extraction circuit 19 (19
00, 1901).

FIG. 5 is an operation explanatory diagram of the mixing circuit 13 used in the first embodiment.

6 is an improved block diagram of the edge extraction circuit 19. FIG.

7A is a filter diagram (1500) used in the first edge detection circuit 15. FIG. (B) Filter diagram used in the second edge detection circuit 16 (1
600, 1601).

FIG. 8 is a spatial frequency characteristic diagram of the filter used in the first embodiment.

9 is a block diagram of an identification circuit 4. FIG.

FIG. 10A is a layout diagram of detection positions of the edge detection filter 41 and the edge detection filter 42. (B) Filter diagram used for the edge detection filter 41 (4
100, 4101). (C) Filter diagram used for the edge detection filter 42 (4
200, 4201).

11 is an improved block diagram of an edge detection filter 41 and an edge detection filter 42. FIG.

12 (a) is a pattern diagram of the binarized data 100. FIG. (B) The fifth detector A (41
It is an edge output figure of e). (C) An output diagram of the edge output ESA of the edge detection filter 41. (D) Edge output E of the improved edge detection filter 41
It is an output figure of SA.

FIG. 13 is a block diagram of a data conversion circuit 7.

14A is a filter diagram (1000) for explaining coefficient setting of the data conversion circuit 7. FIG. (B) is a data conversion curve diagram.

FIG. 15 is an operation explanatory diagram of the data conversion circuit 7.

16 is a block diagram of an emphasis circuit 5. FIG.

FIG. 17 is a block diagram of a gradation conversion processing device according to a second embodiment of the present invention.

FIG. 18A is an explanatory diagram of a conversion operation of the first conversion circuit 2 and the second conversion circuit 3. (B) It is a relationship diagram between the scanning aperture position and the pixel position before conversion. FIG. 6C is a relationship diagram between the scanning aperture position and the pixel position after conversion.

FIG. 19 (a) is a data diagram of K-value image data. FIG. 19B is a binarized data diagram when the K-value image data of FIG. 19A is binarized. FIG. 20C is an output data diagram of the estimated value NA when the binarized data 100 of FIG. 19B is estimated. FIG. 20D is a binarized data diagram when the K-value image data of FIG. 19A is converted by the conversion curve 1 and binarized. 19E is an output data diagram of the estimated value NA when the binarized data 100 of FIG. 19D is estimated. (F) An output data diagram obtained by inversely converting the estimated value NA of FIG.

FIG. 20 is an explanatory diagram of a conversion table.

FIG. 21 is a block diagram of an N-value conversion circuit 8.

22 is a circuit block diagram of the n-th detector A of the edge detection filter 41. FIG.

FIG. 23 is a circuit block diagram of a filter 1500.

FIG. 24 is a circuit block diagram of a first adder circuit 111.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 gradation conversion circuit 2 1st conversion circuit 3 2nd conversion circuit 4 identification circuit 5 emphasizing circuit 6 selection circuit 7 data conversion circuit 8 N-value conversion circuit 9 image memory 11 first multi-value conversion circuit 12 second multi-value conversion Conversion circuit 13 Mixing circuit 14 Quantization circuit 15 First edge detection circuit 16 Second edge detection circuit 17 Mixing circuit 19 Edge extraction circuit 41 Edge detection filter 42 Edge detection filter 91 Operation panel 92 Control circuit 93 Estimating circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04N 1/40 101 E

Claims (23)

[Claims] 1. An adding means for comparing a plurality of filters having the same edge detection characteristic and different focus positions with respective outputs of the plurality of filters with a predetermined value, and adding only outputs having the predetermined value or more. And a determination unit that determines the output of the addition unit at a predetermined determination level and outputs a predetermined determination value, and a region that discriminates a contour region and a non-contour region in an N-value image from the predetermined determination value. Identification device. 2. The area identification device further comprises setting means for setting a plurality of processing modes or image quality adjustment values, and first control means for controlling the predetermined determination level according to the setting contents of the setting means. The area identification device according to claim 1, further comprising: 3. A first detection unit for adding outputs of a plurality of first filters having the same edge detection characteristic and different focus positions, and a plurality of second detection units having different edge detection characteristics from the first filter. Second detection means for adding the outputs of the plurality of second filters having the same edge detection characteristics and different focus positions, and the output of the first detection means for the first determination. A first determination unit that makes a determination based on a level and outputs a predetermined first determination value, and an output from the second detection unit is determined at a second determination level, and outputs a predetermined second determination value. A region discriminating apparatus, comprising a second discriminating means, for discriminating a contour region and a non-contour region in an N-value image based on the predetermined first judgment value and the predetermined second judgment value. 4. The output of the first detecting means is the result of comparing the outputs of the plurality of first filters with the first level and adding only the outputs of the first level and above. Detecting means compares each output of the plurality of second filters with a second level, and outputs only the output of the second level or higher as the output of the second detecting means. The area identification device according to claim 3. 5. The area identification device further includes setting means for setting a plurality of processing modes or image quality adjustment values, and the first determination value or the second determination value according to the setting content of the setting means. The area identification device according to claim 3, further comprising: second control means for controlling the validity and invalidity of the. 6. An N-value gradation image is changed to an M-value gradation image (N <M).
In the gradation conversion processing device for converting to N, the target pixel and its peripheral pixels in the N-value gradation image are multiplied by a predetermined weighting coefficient and added, and the N-value gradation image is converted into the M-value gradation image (N <M) The first smoothing filter and the scanning aperture size or the weighting coefficient to be added are made different to increase the smoothness as compared with the first smoothing filter, and the N-value gradation image is converted into the M value. A second smoothing filter for converting into a gradation image (N <M); edge detecting means for detecting an edge amount in the N-value gradation image; and mixture ratio control means for calculating a mixture ratio from the edge amount. Mixing processing means for mixing the first smoothing filter and the second smoothing filter based on the mixing ratio; identification means for identifying a non-contour area and a contour area in the N-value gradation image; By the control signal from the identification means A gradation conversion processing device comprising: a selection means for selecting an output of the mixing processing means in the non-contour area and a selection means for selecting an output of the first smoothing filter in the contour area. 7. The gradation conversion processing apparatus according to claim 6, wherein the identification means is the area identification apparatus according to claim 1 or 3. 8. An N-value gradation image is converted to an M-value gradation image (N <M).
In the gradation conversion processing device for converting to N, the target pixel and its peripheral pixels in the N-value gradation image are multiplied by a predetermined weighting coefficient and added, and the N-value gradation image is converted into the M-value gradation image (N <M) The first smoothing filter and the scanning aperture size or the weighting coefficient to be added are made different to increase the smoothness as compared with the first smoothing filter, and the N-value gradation image is converted into the M value. A second smoothing filter for converting into a gradation image (N <M); edge detecting means for detecting an edge amount in the N-value gradation image; and mixture ratio control means for calculating a mixture ratio from the edge amount. A mixing processing unit that mixes the first smoothing filter and the second smoothing filter based on the mixing ratio; an emphasizing unit that emphasizes the output of the first smoothing filter; Distinguish between non-contoured and contoured regions of Identifying means, and a selecting means for selecting the output of the mixing processing means in the non-contour area and the output of the emphasizing means in the contour area according to the control signal from the identifying means. Gradation conversion processing device. 9. The gradation conversion processing apparatus according to claim 8, wherein the identification means is the area identification apparatus according to claim 1 or 3. 10. An N-value gradation image is converted into an M-value gradation image (N <
In the gradation conversion processing device for converting into M), the target pixel in the N-value gradation image and its peripheral pixels are multiplied by a predetermined weighting coefficient and added, and the N-value gradation image is converted into the M-value gradation image. The first smoothing filter for converting to (N <M) and the scanning aperture size or the weighting coefficient to be added are made to have a higher smoothness than that of the first smoothing filter, and the N-value gradation image is A second smoothing filter for converting into an M-value gradation image (N <M), an edge detecting means for detecting an edge amount in the N-value gradation image, and a mixture ratio control for calculating a mixture ratio from the edge amount. Means, first processing means for mixing the first smoothing filter and the second smoothing filter based on the mixing ratio, and one row in a scanning aperture including a pixel of interest of the first smoothing filter. Direction or one column direction The change of the output of the first smoothing filter based on the sum of a row direction or one predetermined weighting factor in the column direction on the edge of the scanning opening without the sum and the pixel of interest seen factor 2
Processing means, identifying means for identifying a contour area and a non-contour area in the N-value gradation image, and a control signal from the identifying means selects the first processing means for the non-contour area, Then, a gradation conversion processing device comprising: a selection unit that selects the second processing unit. 11. The gradation conversion processing apparatus according to claim 10, wherein the identification means is the area identification apparatus according to claim 1 or 3. 12. The gradation conversion processing device further includes a setting means for setting a plurality of processing modes or image quality adjustment values, and a first control means or a second control means of the identification means according to the setting contents of the setting means. 12. The gradation conversion processing device according to claim 7, further comprising a third control means for controlling the control means, wherein the identification processing is different depending on the setting content. 13. A plurality of edge detecting means having different characteristics for detecting an edge amount in input N-valued image data, and an extracting means for mixing outputs of the plurality of edge detecting means to generate an edge extraction amount. A plurality of spatial filters having different weighting factors or different smoothing areas for smoothing the N-valued image, and a plurality of spatial filters that normalize outputs of the plurality of spatial filters to a predetermined level and normalize based on the output from the extracting means. And a mixing processing unit that selects or mixes the outputs of the above, and converts the N-valued image into an M-valued image (N <M). 14. The edge extracting unit calculates the mixing ratio from outputs of a plurality of edge detecting units having different detection characteristics, and mixes outputs of the plurality of edge detecting units based on the mixing ratio. 14. The gradation conversion processing device according to claim 13, wherein the gradation conversion processing device generates an amount. 15. The gradation conversion processing apparatus according to claim 13, wherein the spatial filter adds the pixel of interest of the N-value image and its peripheral pixels based on a weighting coefficient or a smooth region. 16. The tone conversion processing apparatus according to claim 13, wherein the normalization is performed by converting to a predetermined level based on a sum of weighting coefficients of the spatial filter and the number of gradations N of the N-valued image. . 17. A first conversion means for converting the K-valued image data so that the K-valued image data is concentrated in the vicinity of the median of the signal range of the K-valued image data, and the K-valued image data converted into N-valued data ( K>
N-value conversion processing means for converting into N), estimation means for estimating M-value image data (N <M) from the N-valued data generated by the N-value conversion processing means, and the M-value image data (N <M), and second conversion means for performing inverse conversion by interlocking with data conversion of the first conversion means, a gradation conversion processing device. 18. The gradation conversion processing apparatus according to claim 17, wherein the first conversion means minimizes the conversion amount at the median value of the signal range of the K-value image data. 19. The first conversion means maximizes the conversion amount in an area smaller than the median of the signal range of the K-value image data or in an area larger than the median of the signal range of the K-value image data. 18. The gradation conversion processing device according to claim 17, which is characterized in that. 20. The gradation conversion processing device according to claim 17, wherein said estimating means is the gradation conversion processing device according to claim 6, 8 or 10, or 13. 21. The gradation conversion processing device further comprises setting means for setting a plurality of processing modes or image quality adjustment values, and a plurality of types of conversion tables corresponding to the plurality of processing modes or image quality adjustment values. 4. A fourth control means for selecting one of the conversion tables according to the setting content of the setting means and resetting the conversion table to the first conversion means and the second conversion means, respectively. Item 17. The gradation conversion processing device according to item 17. 22. The area identifying apparatus according to claim 1, wherein the N-valued image is a binary image. 23. The image processing method according to claim 6, wherein the N-valued image is a binary image.
3. The gradation conversion processing device according to claim 17.