JP2890456B2 - Image processing device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for adjusting the contrast of a color image in an image processing apparatus such as a color copying machine, a color printer, and a color facsimile.

[Conventional technology]

In a color copying machine, a full-color original is reproduced by reading a color original with a color separation signal, processing the image data, converting the color separation signal into a toner signal, and superimposing an image of each toner color. . In this case, since the reproducibility of density, contrast, color tone, and the like differs depending on the characteristics of the document reading unit and the image output unit, it is actually necessary to perform various corrections and adjustments on the image data. The outline will be described.

That is, in the color copying machine, when a color original is read by the color separation signals B (blue), G (green), and R (red), the color original is read by color masking to obtain a color material Y such as color toner, ink, or an ink donor film. (Yellow), M (magenta), C (cyan), and K (black) are converted into recording signals (for example, toner signals) and supplied to an image output unit. In color masking, for example, color separation signals B, G, and R of a color original are Is calculated and converted into recording signals Y, M, and C of the color material, and the color tone during color reproduction is controlled and the saturation is improved while maintaining the color balance.

However, the color signal is often affected by the spectrum of the illumination light source, the characteristics of the dichroic mirror, and the color characteristics of the photoelectric conversion element, color filter, lens, and the like. Also, in the image recording unit, the color material used for recording a color image has unnecessary absorption due to spectral characteristics. Has an element to be corrected. Furthermore, there are many factors that affect the tone reproducibility, such as the characteristics of paper and color ink. For this reason, it is extremely difficult to faithfully reproduce the color of the document by simply performing the color conversion processing.

Therefore, in the color image data processing, in order to faithfully reproduce the color of the original, END before the above color masking
Conversion (equivalent neutral density conversion), and thereafter, TRC (color tone correction control) and other various processes are performed.

Normally, B, G, and R signals and Y, M, and C signals are often expressed as END to facilitate processing. In this case, the color separation signal when gray is read is B =
G = R (equivalent neutral density). Conversely, when the signal of Y = M = C is color-corrected by TRC and sent to the recording unit, gray is reproduced.

That is, signals read by color separation of a color original are taken out as B, G, and R signals, but the color separation signals B, G, and R when a gray (achromatic) original is read are:
The values are not equal due to variations in various characteristics and conditions of the reading device. Therefore, the END conversion is for converting the color separation signals B, G, and R in such a case into gray densities having the same value. The color separation signals B, G, and R are converted into recording signals Y, M, and C of color materials such as toner and ink to be printed, and the signals Y, M, and C, and the black signal K from these signals are further converted. (Black or black) is generated and the corresponding amount of Y, M, C is removed and supplied to the recording unit to reproduce the color image. In this case, the color material signal Y , M, and C have equal values, gray is not faithfully reproduced due to the characteristics and environment of the color material in the recording unit. The TRC reproduces gray of a corresponding density when the color material signals Y, M, and C have the same value. On the other hand, the color masking is for converting the color separation signals B, G, and R into color material signals Y, M, and C. At this time, colors other than gray (hue,
It is important to reproduce the color saturation etc.). That is, EN
D-conversion, color masking, and TRC play a very important role in improving the reproducibility of the original color including gray.

Even if various conversion and correction processes are performed as described above, in general, image quality defects are more remarkable in a color copier than in a black and white copier. For color images, there are individual differences in the evaluation of the output copy, but in addition to this, changes in image quality due to changes in characteristics due to the passage of time and changes in machine conditions caused by the installation environment, etc. Changes in image quality, etc. As a result, for example, there are various colors in the output copy, such as turbidity and poor color tone, strong red, strong blue, tight overall image, good contrast but insufficient density as a whole, good density but poor contrast, etc. Defects in image quality appear. Therefore,
It is necessary to improve such image quality defects,
Therefore, in the color copying machine, the color adjustment function is particularly important.

Conventionally, in a color copying machine, color adjustment is performed by density adjustment to increase the correction amount from low density to high density, and so-called density adjustment commonly performed for Y, M, C, and K; , M, C, and K are individually adopted as color adjustment functions. That is, in the conventional color adjustment, the adjustment for increasing the density and the adjustment for setting the γ to increase the contrast, and the adjustment for lowering the density and the adjustment for decreasing the contrast by increasing the γ are performed simultaneously and simultaneously.

[Problems to be solved by the invention]

However, as described above, in the density adjustment in which the correction amount is increased from the low density to the high density, for example, in the case where the density is good but the contrast is too strong or the contrast is too weak and the contrast is poor, In order to improve the contrast, there is a problem that the density increases because a so-called conversion coefficient γ is set.
Conversely, even in the case where the contrast is good but the density is insufficient as a whole, since γ is set to increase the density, the contrast becomes worse even when the density becomes satisfactory. Or the entire image gives a tight impression.

As described above, with the conventional color adjustment function, it is possible to adjust the density and contrast independently, such as changing the density changes the contrast, and changing the contrast also changes the density. could not. Therefore, it has been difficult to obtain images satisfying all of these.

Further, for example, when gamma adjustment is performed by fixing the highlight portion and changing the shadow portion as shown in FIG. 12 (a), since the density is saturated at the portion A of the shadow portion, the image is not corrected at this portion. Information collapse occurs, and conversely, blur occurs in the B portion of the shadow portion. In such a case, jumping and fogging are less likely to occur in the highlight portion and a bad impression is not given, but in the shadow portion, the change in density is more conspicuous than the change in contrast.

Conversely, as shown in FIG. 12 (b), when γ adjustment is performed by fixing the shadow portion and changing the highlight portion, the output value is not 0 at the C portion of the highlight portion, so that the white input is made. On the other hand, fogging (turbidity) occurs, and jumping occurs because the output value remains 0 from white to a certain input value in the D portion. In such a case, crushing and fogging are unlikely to occur in the shadow portion, but jumps and fogging and the like occur in the highlight portion, and the defect is easily noticeable, so that an image with a poor impression is likely to occur.

The present invention has been made to solve the above problems, and an object of the present invention is to make it possible to independently adjust the density and the contrast. It is another object of the present invention to eliminate defects such as color turbidity, crushing, and roughness in highlights and shadows. Still another object of the present invention is to enable contrast adjustment using a small amount of data. Still another object of the present invention is to provide
The purpose is to allow the contrast to be adjusted without changing the overall density of the entire image.

[Means and actions for solving the problem]

Therefore, according to the present invention, as shown in FIG. 1, a plurality of color-separated image signals are inputted and converted into a plurality of recording signals of color toner by color conversion means 1 and 2, which are converted by the color conversion means. In an image processing apparatus provided with contrast adjusting means 3 for performing contrast adjustment on a plurality of recording signals, the contrast adjusting means 3 highlights the density value of the output recording signal with respect to the density value of the input recording signal. A conversion table set with a conversion characteristic that is non-linear in the halftone from the part to the shadow part, and performing a contrast adjustment by converting the density value of the recording signal using the conversion table. A density conversion means for converting the color-separated image signal when gray is read into an image signal having an equivalent neutral density having an equal density value; The image signal converted into time and converting the recording signal of the color toner. Also,
The conversion table fixes the density value of the print signal output with respect to the density value of the print signal input at the intermediate density value,
The conversion characteristic of the halftone from the intermediate density value to the highlight portion and the shadow portion is set according to the contrast adjustment amount.

Since the intermediate density value is fixed, there is no density change in the halftone, and the contrast can be adjusted. Similarly, since the highlight portion is fixed, jumping and fogging can be eliminated, and since the shadow portion is fixed, crushing and roughness can be eliminated. Further, since the overall density is kept constant, the density does not change even when the contrast is adjusted.

Further, a plurality of point information is provided as contrast adjustment data, and the contrast adjustment characteristic is set by a polygonal line connecting the plurality of points. The data of the point target position on both sides of the intermediate density value may be included as the data of the contrast adjustment, and the data of the intermediate target value and the data of the point target position may be complemented as the data of the contrast adjustment.

For example, even if the image data is represented by 256 gradations, the contrast can be adjusted with several data points, and the memory capacity of the contrast adjustment data can be reduced.

〔Example〕

 Hereinafter, embodiments will be described with reference to the drawings.

 First, a table of contents is shown prior to the description of the embodiment.

(I) Configuration overview of IPS (image processing system) (II) Hardware configuration of IPS (III) Contrast adjustment method (I) Configuration overview of IPS First, the configuration overview of IPS (image processing system) that processes image data explain.

FIG. 2 is a diagram showing an outline of the module configuration of the IPS.

In a color image processing apparatus, a primary color B of light is used by a CCD line sensor in an IIT (image input terminal).
(Blue), G (green), and R (red) are read into a color original, which is read as the primary colors Y (yellow), M (magenta), C (cyan), and K (black or black) of the toner. The image is then converted to a color image by IOT (image output terminal) exposure and development using a laser beam. In this case, a copy process (pitch) is performed once, in which each of Y, M, C, and K toner images is separated and Y is used as a process color. One cycle at a time, for a total of 4
A full-color image is reproduced by executing one copy cycle and superimposing the image by these halftone dots. Therefore, when converting the color separation signals (B, G, R signals) into toner signals (Y, M, C, K signals), how to adjust the color balance, the reading characteristics of IIT and the IOT The problem is how to reproduce the color in accordance with the output characteristics, how to adjust the balance of density and contrast, and how to adjust edge emphasis, blur, and moire.

The IPS receives B, G, and R color separation signals from the IIT and performs various data processing to improve color reproducibility, gradation reproducibility, and definition reproducibility. It converts the toner signal to on / off and outputs it to the IOT. As shown in FIG.
utral Density (equivalent neutral density conversion) module 301, color masking module 302, document size detection module 303, color conversion module 304, UCR (Under Color)
Removal; under color generation module & black generation module 305, spatial filter 306, TRC (Tone Reproduction Control) module 307, reduction / enlargement processing module 308, screen generator 309, IOT interface module
310, an area image control module 311 having an area generation circuit and a switch matrix, an edit control module having an area command memory 312, a color palette video switch circuit 313, a font buffer 314, and the like.

Then, for each of the B, G, and R color separation signals from IIT, 8-bit data (256 gradations) is input to the END conversion module 301, and converted into Y, M, C, and K toner signals. IOT interface module 310 selects the toner signal X of the image signal and binarizes it to generate ON / OFF data of the toner signal of the process color.
Output to IOT. Therefore, full color (4
Color), the prescan first detects the document size, detects the edit area, and detects other document information.
For example, each time a copy cycle in which the process color toner signal X is set to Y and then a copy cycle in which the process color toner signal X is set to M are sequentially executed,
The signal processing corresponding to the original scanning is performed twice.

In IIT, one pixel is read at a size of 16 dots / mm for each of B, G, and R using a CCD sensor, and the data is 24 bits (3 colors x 8 bits; 256 gradations)
Output. The CCD sensor has B, G, and R filters mounted on the upper surface and has a density of 16 dots / mm and 300 mm
Scan at 16 lines / mm at a process speed of 190.5mm / sec, so 15M / s for each color
It outputs read data at pixel speed. In the IIT, the analog data of the B, G, and R pixels are log-converted to convert the information of the reflectance into the information of the density, and further to the digital data.

 Next, each module will be described.

FIG. 3 is a diagram for explaining each module constituting the IPS.

(A) END Conversion Module The END conversion module 301 is a module for adjusting (converting) an optical reading signal of a color original obtained by IIT into a gray-balanced color signal. The amount of toner of a color image is equal in the case of gray, and gray is used as a reference. However, the values of the B, G, and R color separation signals input when a gray document is read from the IIT are not equal because the light source and the spectral characteristics of the color separation filters are not ideal. Therefore, END conversion is performed by using a conversion table (LUT; lookup table) as shown in FIG. Therefore, the conversion table has a characteristic that, when a gray document is read, it is converted into B, G, and R color separation signals at the same gradation corresponding to the level (black → white) and output. , Depends on the characteristics of the IIT. The conversion table is 16
Surfaces are prepared, of which 11 are tables for film projectors including negative films, and 3 are tables for ordinary copying, photography and generation copying.

(B) Color Masking Module The color masking module 302 converts the B, G, and R signals into signals corresponding to the amounts of Y, M, and C toners by performing a matrix operation, and adjusts the gray balance by END conversion. Is processed.

The conversion matrix used for color masking is 3 × 3, which calculates Y, M, and C from pure B, G, and R, respectively.
Is used, but not only B, G, and R,
Of course, various matrices may be used or other matrices may be used to take into account the components of BG, GR, RB, B ² , G ² , and R ² . The conversion matrix has two sets, one for normal color adjustment and one for intensity signal generation in the monocolor mode.

As described above, when processing the IIT video signal by the IPS, first of all, the gray balance is adjusted. If you do this after color masking,
Since it is necessary to perform gray balance adjustment using a gray document in consideration of the characteristics of color masking, the conversion table becomes more complicated.

(C) Original Size Detecting Module In some cases, not only originals of standard size but also originals of any shape such as cutouts may be copied. In this case, in order to select a sheet of an appropriate size corresponding to the document size, the document size needs to be detected. Also, when the copy paper is larger than the document size, erasing the outside of the document can improve the quality of the copy. For this reason, the document size detection module 303 performs document size detection during prescanning and platen color erasing (frame erasing) processing during document scanning. For this purpose, the platen color is set to a color that can be easily distinguished from the original, for example, black, and the upper limit value / lower limit value of platen color identification is set in the threshold register 3031 as shown in FIG. 3B. At the time of pre-scanning, a signal X (using the output of the spatial filter 306 described later) X converted (γ-converted) into information close to the reflectance of the document and an upper limit value / lower limit value set in the threshold register 3031 are compared.
The comparison is made in 3032, the edge detection circuit 3034 detects the edge of the document, and the maximum and minimum values of the coordinates x and y are sorted by the maximum / minimum sorter 3.
035.

For example, when the document is inclined or not rectangular as shown in FIG. 3D, the maximum and minimum values (x ₁ , x ₂ , y ₁ , y ₂ ) are detected and stored. You. When scanning the original, the comparator 3033 uses the Y, M,
C is compared with the upper limit value / lower limit value set in the threshold register 3031, and a platen color erasing circuit 3036 erases the read signal outside the edge, that is, the platen, and performs a frame erasing process.

(D) Color Conversion Module The color conversion module 304 converts a specified color in a specific area. As shown in FIG. 3C, a window comparator 3042, a threshold register 3041, a color palette 3043 etc., and when performing color conversion, each of Y, M,
The upper limit value / lower limit value of C is set in the threshold register 3041, and the values of Y, M, and C of the conversion color are set in the color palette 3043. Then, although the area signal is input from the area image control module, the NAND gate 3044 is controlled. If the area signal is not in the color conversion area, the Y, M, and C of the original are sent out from the selector 3045 as they are, and When the Y, M, and C signals of the original enter between the upper and lower limits of Y, M, and C set in the threshold register 3041, the selector 3045 is switched by the output of the window comparator 3042 and set in the color palette 3043. The converted colors Y, M, and C are transmitted.

The designated color is recognized by directly averaging 25 pixels of B, G, and R around the coordinates designated at the time of pre-scan by directly pointing the original with the digitizer.
By this averaging operation, for example, even a 150-line original can be recognized with an accuracy within a color difference of 5 or less. When reading the B, G, and R density data, the designated coordinates are converted into addresses from the IIT shading correction RAM and read out. At the time of address conversion, it is necessary to readjust the registration adjustment as in document size detection. In prescan, IIT operates in sample scan mode. The B, G, and R density data read from the shading correction RAM are averaged after being subjected to shading correction by software, and further averaged.
After performing END correction and color masking, it is set in the window comparator 3042.

Registered colors can be registered in the color palette 3043 from 16.7 million colors up to 8 colors at the same time. Standard colors are Y, M, C,
G, B, R, and their intermediate colors, and 14 colors of K and W are prepared.

(E) UCR & black generation module If Y, M, and C are equivalent, it becomes gray.
Theoretically, the same color can be reproduced by replacing equivalent amounts of Y, M, and C with black. However, in reality, replacing black with black causes turbidity of the color and deteriorates the reproducibility of a vivid color.
Therefore, the UCR & black generation module 305 generates an appropriate amount of K so as not to cause such color turbidity, and performs a process of reducing Y, M, and C by an equal amount (under color removal) according to the amount.
Specifically, the maximum and minimum values of Y, M, and C are detected,
K is generated below the minimum value from the conversion table according to the difference, and a constant undercolor removal is performed for Y, M, and C according to the amount.

In the UCR & black generation, as shown in FIG. 3 (e), for example, when the color becomes close to gray, the difference between the maximum value and the minimum value becomes small. Is generated. If the difference between the maximum value and the minimum value is large, the amount of removal is made smaller than the minimum values of Y, M, and C, and K
By reducing the generation amount of black, the incorporation of black ink and the decrease in saturation of low-brightness and high-saturation colors are prevented.

In FIG. 3 (f) showing a specific circuit configuration example, the maximum value / minimum value detection circuit 3051 detects the maximum value and the minimum value of Y, M, and C, and the arithmetic circuit 3053 calculates the difference. And
K is generated by the conversion table 3054 and the arithmetic circuit 3055.
The conversion table 3054 adjusts the value of K. If the difference between the maximum value and the minimum value is small, the output value of the conversion table 3054 becomes zero. If the difference between the maximum value and the minimum value is large, the output value of the conversion table 3054 is not zero.
Is output as the value of A conversion table 3056 is a table for obtaining values to be removed from Y, M, and C corresponding to K, and an arithmetic circuit 3059 passes through the conversion table 3056 to calculate Y, M, and C.
To K corresponding to K. And And Gate 305
7, 3058 gate the K signal and the signals after the undercolor removal of Y, M, C in accordance with each signal of the mono color mode and the 4 full color mode.
Numeral 50 is for selecting one of Y, M, C and K according to the process color signal. Thus, in practice, Y,
Since the colors are reproduced by the halftone dots of M and C, the elimination of Y, M, and C and the generation ratio of K are set using a curve, a table, or the like generated empirically.

(F) Spatial filter module In the apparatus applied to the present invention, the IIT
Since the original is read while scanning the CCD with, using the information as it is will result in blurred information, and since the original is reproduced with halftone dots, the dot cycle of the printed matter and 16 dots
Moire occurs between the sampling period of / mm. Moire also occurs between the dot cycle generated by itself and the dot cycle of the document. The spatial filter module 306 has a function of recovering such blur and a function of removing moiré. Then, a low-pass filter is used for removing moire, and a high-pass filter is used for edge enhancement.

In the spatial filter module 306, as shown in FIG. 3 (g), one color of the input signals of Y, M, C, Min and Max-Min is extracted by the selector 3003, and information close to the reflectance is obtained by using the conversion table 3004. Convert to This is because this information makes it easier to pick up an edge, and for example, Y is selected as one color. Also, the threshold register
3001, a 4-bit binarization circuit 3002, and a decoder 3005, for each pixel, Y, M, C, Min and Max-Min to Y,
The colors are separated into eight colors of M, C, K, B, G, R, and W (white). The decoder 3005 recognizes the hue in accordance with the binarized information and outputs whether or not the process color is a necessary color as 1-bit information.

The output of FIG. 3 (g) is input to the circuit of FIG. 3 (h). Here, the FIFO 3061 and the 5 × 7 digital filter 30
63, information of halftone removal is generated by the modulation table 3066, and the FIFO 3062 and the 5 × 7 digital filter 3064;
The modulation table 3067 and the delay circuit 3065 generate edge enhancement information from the output information in FIG. The modulation tables 3066 and 3067 are selected in accordance with a copy mode such as exclusive use for photographs and characters, and mixed mode.

In edge enhancement, for example, when trying to reproduce a green character as shown in FIG.
And M is not emphasized as indicated by the solid line. This switching is performed by the AND gate 3068. In order to perform this processing, if the color is emphasized as indicated by a dotted line, turbidity due to the color mixture of M occurs at the edge as shown in FIG.
The delay circuit 3065 is used to switch such emphasis by the AND gate 3068 for each process color.
62 and the 5 × 7 digital filter 3064 are synchronized. When a bright green character is reproduced by normal processing, magenta is mixed with the green character and turbidity occurs. Therefore, when green is recognized as described above, Y and C are output as usual, but M is suppressed so that edge enhancement is not performed.

(G) TRC conversion module IOT, Y, according to ON / OFF signal from IPS
Although four copy cycles (in the case of four full-color copies) are executed by each process color of M, C, and K to enable reproduction of a full-color original, actually, the color theoretically obtained by signal processing is used. To play faithfully, IO
Subtle adjustments that take into account the characteristics of T are required. The TRC conversion module 307 is for improving such reproducibility, and the combination of Y, M, and C densities
As shown in FIG. 3 (j), the RAM has an address conversion table for inputting 8-bit image data as an address, and performs density adjustment, contrast adjustment, negative / positive inversion, color balance adjustment, character mode, and watermark according to the area signal. It has editing functions such as composition. For the upper three bits of the RAM address, bits 0 to 3 of the area signal are used.
Further, the above functions can be used in combination in the out-of-area mode. This RAM is composed of, for example, 2 kbytes (256 bytes × 8 planes) and has a conversion table of 8 planes, and is stored for a maximum of 8 planes during the IIT carriage return for each cycle of Y, M, and C. Is selected in accordance with an area designation or a copy mode. Of course, if the RAM capacity is increased, it is not necessary to load each cycle.

(H) Reduction / enlargement processing module The reduction / enlargement processing module 308 performs the enlargement / reduction processing through the reduction / enlargement processing circuit 3082 in the process of temporarily holding and sending the data X to the line buffer 3083 as shown in FIG. The resampling generator & address controller 3081 generates a sampling pitch signal and a read / write address of the line buffer 3083. The line buffer 3083 is a ping-pong buffer composed of two lines so that one line can be read and the next line data can be written to the other line at the same time. In the enlargement / reduction processing, digital processing is performed by the enlargement / reduction processing module 308 in the main scanning direction, but the scanning speed of the IIT is changed in the sub-scanning direction. Scan speed is 50% to 400% by changing from 2x speed to 1 / 4x speed
Can be scaled up and down. In digital processing, the line buffer 30
When data is read / written, the data can be reduced by thinning and complementing the data, and can be enlarged by supplementing the data. When the complementary data is in the middle, as shown in FIG. 1A, the complementary data is generated by performing a weighting process according to the distance from the data on both sides. For example, when the data Xi 'are on both sides of the data X _i, the distance d _1, d ₂ between the X _{i + 1} and these data and the sampling _{point, (X i × d 2)} + (X i + 1 × d ₁ ) where d ₁ + d ₂ = 1 is calculated.

In the case of the reduction processing, the data is written to the line buffer 3083 while complementing the data, and at the same time, the reduced data of the previous line is read from the buffer and transmitted. In the case of the enlargement processing, the data is temporarily written as it is, and at the same time, the data of the previous line is read and complementarily enlarged and transmitted. When the complementary enlargement is performed at the time of writing, the clock at the time of writing must be increased according to the enlargement ratio. However, as described above, writing / reading can be performed with the same clock. Also, by using this configuration, it is possible to perform shift image processing in the main scanning direction by reading out in the middle or by reading out at a delayed timing, and it is possible to perform processing repeatedly by reading out repeatedly, and to read out from the opposite side. Mirror image processing can also be performed.

(I) Screen Generator The screen generator 309 converts a process color gradation toner signal into an on / off binarized toner signal and outputs the same, and compares the threshold value matrix with a gradation-expressed data value. Binarization processing and error diffusion processing are performed. In the IOT, this binarized toner signal is input, and is approximately 80 μm in height and 6 in width to correspond to 16 dots / mm.
A halftone image is reproduced by turning on / off an oval laser beam of 0 μmφ.

First, a gradation expression method will be described. As shown in FIG. 3 (n), for example, a 4 × 4 halftone cell s
Will be described. First, in the screen generator, a threshold value matrix m is set corresponding to such a halftone cell s, and the threshold value matrix m is compared with a data value expressed in gradation. In this comparison processing, for example, assuming that the data value is “5”, a signal for turning on the laser beam is generated at a portion of “5” or less in the threshold value matrix m.

100 4 x 4 halftone cells at 16 dots / mm
Although the spi is a halftone dot of 16 gradations, the image is coarse and the reproducibility of a color image is poor. Therefore, in the present invention, as a method of increasing the gradation, the 16-dot / mm pixel is vertically divided into four (in the main scanning direction), and the on / off frequency of the laser beam in pixel units is shown in FIG. 1/4 as shown
, Ie, four times higher, to express a tone four times higher. Accordingly, a threshold value matrix m 'as shown in FIG. Further, it is also effective to adopt a sub-matrix method to increase the number of lines.

In the above example, the same threshold matrix m having the vicinity of the center of each halftone cell as the sole growth nucleus is used. However, the sub-matrix method is configured by a set of a plurality of unit matrices, and is shown in FIG. In this way, the growth nucleus of the matrix is set at two or more (multiple) locations. If such a screen pattern design method is adopted, for example, a bright place is 141 spi, 64 gradations, and as it gets darker, a dark place is made by 200 spi, 128 gradations,
The number of lines and the gradation can be freely changed according to a bright place. Such a pattern can be designed by visually judging the smoothness of gradation, fine line property, graininess, and the like.

When a halftone image is reproduced by the dot matrix as described above, the number of gradations and the resolution have an opposite relationship. In other words, there is a relationship that the resolution deteriorates as the number of gradations increases, and the number of gradations decreases as the resolution increases. Further, when the matrix of the threshold data is reduced, a quantization error occurs in an image to be actually output. Error diffusion processing
As shown in FIG. 29 (q), the density conversion circuit 3093 and the subtraction circuit 3094 detect a quantization error between the on / off binary signal generated by the screen generator 3092 and the input gradation signal, and a correction circuit. 3095, to improve the reproducibility of the gradation when viewed macroscopically by feedback using the adder circuit 3091.For example, an error in which the corresponding position of the previous line and the pixels on both sides thereof are convolved through a digital filter. Diffusion processing is being performed.

In the screen generator, as described above, the threshold data and the feedback coefficient of the error diffusion process are switched for each document or area according to the type of image such as a halftone image or a character image, and the reproducibility of a high gradation and high definition image is improved. .

(J) Area image control module The area image control module 311 has a configuration in which seven rectangular areas and their priorities can be set in the area generation circuit, and area control information is set in the switch matrix corresponding to each area. Is done. The control information includes color conversion, color mode such as mono color or full color, modulation select information such as photographs and characters, TRC select information, screen generator select information, etc., the color masking module 302, the color conversion module 304, and the like. , UCR 305, spatial filter 306, and TRC module 307. The switch matrix can be set by software.

(K) Edit Control Module The edit control module reads a document such as a pie chart instead of a rectangle, and enables a coloring process in which a specified area having an unlimited shape is filled with a specified color. As shown in m), the AGDC (advan
cd Graphic Digital Controller) 3121, font buffer 3126, logo ROM 3128, and DMAC (DMA Controller) 3129 are connected. Then, the encoded 4-bit area command is written from the CPU to the plane memory 3122 through the AGDC 3121, and the font is written to the font buffer 3126. The plane memory 3122 is composed of four sheets. For example, in the case of "0000", each point of the original can be set with four bits of planes 0 to 3 such that the command is 0 and the original original is output. . The decoder 3123 decodes the 4-bit information into the command 0 to the command 15, and the switch matrix sets whether the command 0 to the command 15 is a command to execute a fill pattern, a fill logic, or a logo. 3124. Font address controller 3125
The address of the font buffer 3126 is generated in accordance with a pattern such as a halftone shade or a hatched shade by a 2-bit fill pattern signal.

The switch circuit 3127 selects the document data X, the font buffer 3126, the color pallet, and the like based on the fill logic signal of the switch matrix 3124 and the content of the document data X. The fill logic is information for filling only a background (background portion of a document) with a color mesh, color conversion of a specific portion (foreground), masking, trimming, filling, and the like.

In the IPS of the present invention, as described above, the IIT original reading signal is first subjected to END conversion and then color masked,
Processing with full-color data is more efficient than document color, frame erasure, and color conversion, and then undercolor removal and black generation to narrow down to process colors. However, processes such as spatial filters, color modulation, TRC, and scaling are performed by processing process color data.
Reduces the amount of processing compared to processing with full-color data, reduces the number of conversion tables used to 1/3, and increases the number of types to increase the flexibility of adjustment, color reproducibility, and gradation reproduction. Improves the reproducibility of sex and definition.

(II) Hardware Configuration of IPS FIG. 4 is a diagram showing an example of a hardware configuration of the IPS.

In the IPS of the present invention, two substrates (IPS-A, IPS-B)
The first substrate (IPS-A) has a portion that achieves basic functions as a color image forming apparatus such as color reproducibility, gradation reproducibility, definition reproducibility, and the like. In this way, the part that achieves the application and specialized functions is placed on the second substrate (IPS-
B). The former configuration is shown in FIG.
(C), and the latter configuration is (d) in FIG. In particular, if the basic function can be sufficiently achieved by the first substrate, the second substrate
It is possible to flexibly respond to application and specialized functions simply by changing the design of the board. Therefore, in order to further enhance the function of the color image forming apparatus, it is possible to cope only by changing the design of the other substrate.

As shown in FIG. 4, the CPU bus (address bus ADRSBUS, data bus DATABUS, control bus CTRLBUS) is connected to the IPS board, and IIT video data B, G,
R, a video clock IIT • VCLK, a line synchronization (main scanning direction, horizontal synchronization) signal IIT • LS, and a page synchronization (sub scanning direction, vertical synchronization) signal IIT • PS are connected as synchronization signals.

Since the video data is pipeline-processed after the END conversion unit, data delay occurs in each processing stage in clock units required for processing. Therefore,
The line synchronization generation & fail check circuit 328 generates and distributes a horizontal synchronization signal in response to such a delay of each process, and performs a fail check of the video clock and the line synchronization signal. Therefore, line synchronization occurs &
The fail check circuit 328 has a video clock IIT / VC
LK is connected to the line synchronization signal IIT / LS, and the CPU bus (ADRSBUS, DATABU
S, CTRLBUS) and a chip select signal CS.

The video data B, G, and R of the IIT are input to the ROM 321 of the END conversion unit. The END conversion table is, for example, a CPU using RAM.
However, since the apparatus is in use and there is almost no need to rewrite during image data processing, two ROMs of 2 kbytes each for B, G, and R are used. LUT (Look Up Table) method using ROM. It has a conversion table of 16 planes, and can be switched by a 4-bit selection signal END Sel.

The END-converted output of the ROM 321 is connected to a color masking unit including three arithmetic LSIs 322 having two 3 × 1 matrices for each color. Each path of the CPU is connected to the arithmetic LSI 322, and a matrix coefficient can be set from the CPU. The setup signal SU and the chip select signal CS are connected to switch from the processing of the image signal to the bus of the CPU for rewriting by the CPU and the like, and the 1-bit switching signal MONO is connected to the selection switching of the matrix. The power down signal PD is input, and the internal video clock is stopped when the IIT is not scanning, that is, when image processing is not being performed.

The signal converted from B, G, R to Y, M, C by the operation LSI 322 is a second substrate (IPS-B) shown in FIG.
After color conversion processing through the color variable LSI353,
Input to SI323. The color conversion LSI 353 has four color conversion circuits consisting of a threshold register for setting non-conversion colors, a color palette for setting conversion colors, and a comparator. The DOD LSI 323 has a document edge detection circuit and a frame eraser. It has a circuit, etc.

The output of the DOD LSI 323 subjected to the frame erasure processing is sent to the UCR LSI 324. This LSI includes a UCR circuit, a black generation circuit, and a required color generation circuit, and includes a process color X, a required color Hue, and an edge Edg corresponding to a toner color in a copy cycle.
Output each signal of e. Therefore, a 2-bit process color designation signal COLR and a color mode signal (4COLR, MONO) are also input to this LSI.

The line memory 325 stores each signal of the process color X, the required color Hue, and the edge Edge output from the UCR LSI 324 by 5 ×
7 includes a FIFO for storing data for four lines to be input to the digital filter 326 and a FIFO for matching the delay. Here, the process color X and the edge Edge are accumulated for four lines, and a total of five lines are sent to the digital filter 326, and the required color Hue is delayed by the FIFO and synchronized with the output of the digital filter 326. To send to.

The digital filter 326 has two sets of 5 × 7 filters (low-pass LP and high-pass HP) each including three 2 × 7 filter LSIs. On the other hand, the processing for the process color X is performed. Is being processed. In the MIX LSI 327, these outputs are subjected to halftone dot removal and edge emphasis processing using a conversion table and mixed with the process color X. Here, Edge EDGE and Sharp Shar are used as signals for switching the conversion table.
p has been entered.

The TRC342 is a 2k-byte conversion table that holds eight conversion tables.
Consists of RAM. The conversion table is configured to rewrite the conversion table using a carriage return period before each scan, and a 3-bit switching signal TRC
Switched by Sel. Then, the processing output from this is sent from the transceiver to the LSI 345 for scaling processing. The enlargement / reduction processing unit configures a ping-pong buffer (line buffer) using two 8-kbyte RAMs 344. The LSI 345 generates a resampling pitch and generates an address for the line buffer.

The output of the scaling processing section returns to the EDF LSI 346 through the area memory section of the second substrate shown in FIG. LSI for EDF3
46 has an FIFC that holds information on the previous line, and performs error diffusion processing using the information on the previous line. Then, the signal X after the error diffusion processing is output to the IOT interface via the SG LSI 347 constituting the screen generator.

In the IOT interface shown in FIG. 4 (c), the signal from the SG LSI 347, which is input as a 1-bit on / off signal, is collected into 8 bits by the LSI 349 and transmitted to the IOT in parallel.

In the second substrate shown in FIG. 4, the data actually flowing is 16 dots / mm.
It is reduced to / 4 and binarized and stored in the area memory. The expansion decode LSI 359 has a fill pattern RAM 360, expands to 16 dots / mm when reading area information from the area memory and generating a command, generating a logo address,
Performs color palette and fill pattern generation processing. The DRAM 356 stores coded 4-bit area information composed of four planes. AGDC355 is a dedicated controller for controlling area commands.

(III) Contrast Adjustment Method (A) Outline of Color Adjustment System First, an outline of an overall color adjustment system that performs color correction, conversion, and adjustment on a color separation signal obtained by reading a document will be described.

In IPS, as described above, B output from IIT,
First, for the G and R color separation signals,
In addition to correcting the spectrum of IT lighting light sources and the characteristics of dichroic mirrors, as well as the variation in color characteristics of photoelectric conversion elements, color filters, lenses, etc., positive to negative inversion when using a film projector and film transmittance by film manufacturers Are corrected and converted into gray-balanced color separation signals ENDB, ENDG, ENDR. Then, in the color masking circuit, the gray balance color separation signal ENDB output from the END conversion circuit,
Based on the ratio of ENDG, ENDR, calculations are performed using a matrix to which a gray balance method such as 3 × 3, 3 × 6, 3 × 9, etc. is applied, and in full color, three color gray-balanced toner signals ENDY, ENDM, ENDC pixels Data is generated, and a luminance signal is generated at the time of mono-color. Then, in the UCR circuit, K is generated from the ratio of the toner signals ENDY, ENDM, ENDC at the time of four full colors, and the values of the toner signals ENDY, ENDM, ENDC are subtracted according to the generated K value (under color removal). ). In the generation of K, the maximum value and minimum value detection circuit detects the maximum value and the minimum value of the toner signals ENDY, ENDM, ENDC, obtains the maximum value and the minimum value by a subtractor, and obtains the chroma according to the difference. The value converted by the function is subtracted from the minimum value by a subtractor. Further, the value obtained by converting K by the UCR function is subtracted from the values of the toner signals ENDY, ENDM, ENDC by a subtractor. In the case of three colors, full color and mono color, the generation of K and the removal of the under color of ENDY, ENDM, ENDC are not performed, so that the UCR circuit is bypassed. Then, the TRC circuit determines the value of the pixel data to be output for the input pixel data based on the TRC curve, and changes the TRC curve to perform density adjustment, contrast adjustment, color balance adjustment, inversion, and the like. Is going.

With respect to the color separation signal input from the IIT as described above, the IPS first performs END conversion and gray balance, and then performs color masking to generate a toner signal. In addition, in this color masking, the gray balance method is applied, and further, the black signal generation and the same amount of under color removal are performed on the toner signal by the UCR, and the toner signals ENDY ′, ENDM ′, ENDC ′ are always equal to gray. Adjusted to be a value. Then, each toner signal ENDY ', ENDM', END
When C 'is equal, the toner signals Y, M, and C are adjusted by the TRC so that gray is output from the IOT, the gray balance is adjusted in the image data processing system, and gray is reproduced in the output system. Is increasing the character.

The END conversion circuit and the TRC circuit are both LUT-type conversion tables each including a memory storing conversion values, and the TRC circuit is rewritten by the CPU. The color masking circuit performs a matrix operation. Basically, the multiplication value is stored in a conversion table of the LUT method, and the operation value is calculated only by adding and subtracting the read memory value without multiplication. It is configured to be obtained. LU like this
When the conversion table of the T method is adopted, it is possible to cope with the conversion of the non-linear characteristic, and there is an advantage that not only the theoretical conversion value but also the conversion value can be set empirically. Moreover,
The value of the memory is rewritten by the CPU similarly to the TRC circuit. Therefore, the CPU has basic parameters, and calculates the data to be written to the memory of the TRC circuit and the color masking circuit at the time of copy start or IIT carriage return (back scan) according to the execution conditions such as the copy mode. Perform write processing. The execution conditions include, for example, whether the copy original is a photo original, a text original, a print original, a mixed original of these, whether the output copy is monocolor, and the like. Is also changed by And these select signals are END
The conversion circuit is supplied from the CPU, and the TRC circuit and the color masking circuit are supplied from the edit control unit.

Next, the contrast adjustment performed by the TRC circuit in the above circuit will be described.

(B) Contrast adjustment FIG. 5 is a diagram showing an example of a conversion table used in the contrast adjustment method of the image processing apparatus according to the present invention, and FIG. 6 is a diagram showing an example of setting the conversion table.

The conversion table for contrast adjustment of the present invention is combined with the conversion table (LUT) of the TRC circuit that performs color tone adjustment in the color adjustment system. The conversion table for contrast adjustment includes a highlight portion A, as shown in FIG.
The intermediate density portion B and the shadow portion C are respectively fixed, and the distance between these fixed points is changed. By doing so, first, jumping and fogging as described above hardly occur in the highlight portion, and similarly, crushing and blurring hardly occur in the shadow portion. That is, since a gradation change of OUT is continuously obtained with respect to IN, a defect caused by discontinuity of gradation is less likely to occur. As described above, defects in highlight portions and shadow portions that are generally conspicuous can be prevented, so that an image with a good impression can be obtained. In addition, since the intermediate density portion B of the conversion characteristic is fixed and the inclination is changed, the overall change in density is suppressed, and the contrast can be adjusted without giving the impression that the density has changed. That is, when the density increases (or decreases) at a certain gradation, there is a gradation in which the density decreases (or increases), so that the change in the total density of the entire image decreases. Therefore, the change of the relative density difference in the image becomes more noticeable than the density fluctuation of the entire image. Therefore, it is possible to obtain a good contrast-adjusted image in which a change in contrast of the halftone is large.

FIG. 5 (a) shows the contrast adjustment characteristics based on a smooth curve, while FIG. 5 (b) shows that the capacity of the stored contrast adjustment data is reduced. In this example, the output value is set in a plurality of stages corresponding to the input data D _IL and D _IH so that the contrast adjustment characteristic by the polygonal line connecting the highlight portion A, the intermediate density portion B and the shadow portion C is provided. It was done.

The table shown in FIG. 5 (b) corresponds to, for example, FIG. 6 (a)
As shown in the _figure , a _L , b _L ,...,
e _L, a _H with respect to the input data _{D IH, b H, ......,} e H is set, the intermediate density portion and the point symmetry as shown in FIG. (b)
_{a L -a H, b L -b} H, adjustment table is set by a set of .... Therefore, the polygonal line shown in FIG. 5B is selected by this set, and the contrast is adjusted. Also,
As the data of the contrast adjustment table, it is also possible to set only one point x as shown in FIG. That is, when one point x is selected, ΔIN and ΔOUT are obtained between x and the intermediate density part B, and these values are added to the intermediate density part B. And a point x ′ that is symmetrical with the point can be obtained. Then, a conversion table for the contrast adjustment of the polygonal line shown in FIG. 5B can be generated from x 'thus obtained and the selected point x.

(C) Generation algorithm of TRC conversion table FIG. 7 is a diagram showing a basic generation algorithm of the TRC conversion table, FIG. 8 is a diagram for explaining a process of generating the TRC conversion table, and FIG. It is a figure showing an example.

The TRC circuit adjusts density, contrast, and color balance using an LUT as a conversion table.
Has a standard TRC curve and coordinate data for polygonal line approximation, and a TRC curve is generated using these data and written into the RAM of the LUT. When the image data has, for example, 256 gradations, the data of the polygonal line approximation are 0, x, and x as shown in FIG.
_It consists of data corresponding to the X coordinate of four points consisting of ₁ , x ₂ , and 255. This data is read from the memory, and the blood of the Y coordinate with respect to the X coordinate of 0 to 255 is obtained. Then, the value of the Y coordinate is converted by a standard TRC curve (Graph 2), and the converted value is written to the RAM of the LUT.

According to the data of the 0, x _1, x _2, broken line approximation four points consisting of 255 as described above, the bright area, the dark area, it is possible to set the conversion characteristic is divided into their middle region, a small number of data points Thus, a desired TRC conversion table can be efficiently generated. Further, since the TRC conversion table is generated by the butting method using the data of the approximation of the broken line and the standard TRC curve, the table can be generated by simple processing. The number of data points may be four or more. For example, if the number of points to be set is set to about 10, a more accurate approximate curve can be obtained.

As shown in FIG. 8, broken line approximation data includes density adjustment, contrast adjustment, color balance adjustment, inversion, and monocolor data. For example, the case of obtaining data to be written to the address X of the LUT will be described. First, OUT corresponding to the X value of IN in the virtual TRC for density adjustment shown in FIG.
Of 'seeking, then the Y _1' value Y ₁ obtaining a value Y ₂ 'of OUT in virtual TRC for contrast adjustment of the figure obtained from the data of the polygonal line approximation for contrast adjustment based on (b). Hereinafter, similar processing is performed on the virtual TRCs of FIGS. (C), (d) and (e) to obtain Y ₅ ′, and this Y ₅ ′ is set to IN of the standard TRC shown in FIG. Is read. And this Y
The value corresponds to the value of X in the LUT, and is written into the LUT by the CPU using the value of X as an address. The above-mentioned polygonal line approximation data has a plurality of data corresponding to, for example, toner signals Y, M, C, and K, and a plurality of data corresponding to photographs, characters, printing, and mixed image types, respectively. A total of eight TRC curves, one for normal and seven for adjustment for each area, are generated and written to the LUT. Therefore, in a full-color copy, if the TRC curve of the toner signal is written to the LUT upon carriage return, writing of eight TRC curves is sufficient, but if such writing is not performed every carriage return, the copy is performed. At the start, 32 TRC curves need to be written. FIG. 9 shows an example of a TRC conversion curve expressed in 256 gradations.

(D) Configuration Example of TRC Circuit FIG. 10 is a diagram showing a specific configuration example of a TRC circuit configured by an LSI. In FIG. 10, a latch circuit 722 latches a 5-bit area signal, and a latch circuit 723 latches 8-bit image data. Also, the buffer
724 and 725 hold the CPU address signal, respectively.
The decoder (PAL) 721 decodes a control signal. The RAM 726 is a memory for storing the TRC curve, and its output is sent out through the latch circuit 727, and writing and reading from the CPU are performed through the bidirectional data bus controller 728. Therefore, in the normal mode for processing image data, the TRC curve stored in the RAM 726 is switched according to the area signal, and the image data latched by the latch circuit 723 is converted and transmitted from the latch circuit 727. Further, in the access mode of the CPU, selection of a TRC curve and designation of an address are performed from the buffers 724 and 725, and reading / reading from the decoder 721 is performed.
Writing is controlled, and the bidirectional data bus controller 72
Through RAM 8, the contents of RAM 726 are read / written from the CPU.

In addition, instead of storing the standard TRC 256 bytes / color in the ROM, only the coefficients of the polynomial are stored in memory by approximating them with a higher-order polynomial, and the TRC curve is generated by expanding the data by the CPU when actually using it. May be configured.
In this way, the memory capacity can be reduced, and the standard TRC itself can be easily changed.

It should be noted that the present invention is not limited to the above embodiment, and various modifications are possible. For example, in the above embodiment, the contrast adjustment table is stored in the LUT of the TRC circuit for color tone adjustment for density adjustment, color balance adjustment, inversion,
Although they are synthesized together with the mono color, etc., it goes without saying that they may be configured independently of the TRC circuit. In addition, the contrast adjustment table can be easily selected by using the data of four points. However, the number of data points may be increased. May be provided.

FIG. 11 is a diagram showing another embodiment of the contrast adjusting method of the image processing apparatus according to the present invention. As a modification of the present invention, it may be modified as shown in FIGS. . In the example shown in FIG. 11A, since only the intermediate density portion is fixed and γ is changed, some jumping, fogging, crushing, blurring, etc. occur in both the highlight portion and the shadow portion. However, an extreme defect unlike the conventional example described above does not occur. Moreover, since the overall density of the image does not change, the impression that the density has changed is not received, and the change in contrast is noticeable. In the example shown in FIG. 2B, as shown in FIG. 2A, adjustment is performed by changing γ and freely shifting without particularly fixing the intermediate density portion. By adopting such a table, the density and the contrast can be simultaneously adjusted even if a certain amount of jumping, fogging, crushing, blurring or the like occurs in a highlight portion or a shadow portion by using the table of the embodiment in which the density and the contrast are independently adjusted. The one to be adjusted may be mixed.

〔The invention's effect〕

As is apparent from the above description, according to the present invention, the highlight portion, the shadow portion, and the intermediate density portion are fixed, and the contrast is adjusted in the middle of these points. It is possible to prevent the shadow portion from being crushed or blurred. Therefore, a good impression image can be output. Further, since the intermediate density portion is fixed so that the sum of the densities of the entire image is not changed, the contrast can be adjusted without giving an impression that the density has changed. Further, since the conversion table is set by a polygonal line composed of several points, the number of data points in the conversion table can be reduced.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining an embodiment of the image processing apparatus according to the present invention, FIG. 2 is a diagram showing an outline of the module configuration of the IPS, and FIG. 3 is a diagram for explaining each module constituting the IPS. FIG. 4 is a diagram showing an example of the hardware configuration of the IPS,
FIG. 5 is a diagram showing an example of a conversion table used in the contrast adjustment method of the image processing apparatus according to the present invention, FIG. 6 is a diagram showing a setting example of the conversion table, and FIG. Diagram showing generation algorithm, FIG. 8 shows TR
FIG. 9 illustrates an example of a TRC curve, and FIG. 10 illustrates a TR configured by an LSI.
FIG. 11 is a diagram showing a specific configuration example of a C circuit, FIG. 11 is a diagram for explaining another embodiment of the present invention, and FIG. 12 is a diagram for explaining a conventional color adjustment function. 1... Neutral density conversion means 2... Recording signal conversion means 3
...... Contrast adjusting means.

Claims (8)

(57) [Claims] 1. A color conversion means for inputting a plurality of color-separated image signals and converting them into a plurality of recording signals of color toner, and performing contrast adjustment on the plurality of recording signals converted by the color conversion means. In an image processing apparatus provided with a contrast adjusting unit, the contrast adjusting unit converts the density value of the output recording signal with respect to the density value of the input recording signal so as to be non-linear in a halftone from a highlight portion to a shadow portion. An image processing apparatus comprising: a conversion table set by characteristics; and performing contrast adjustment by converting a density value of a recording signal using the conversion table. 2. The image processing apparatus according to claim 1, wherein the color conversion unit includes a density conversion unit configured to convert the color-separated image signal when gray is read into an image signal having an equivalent neutral density having an equal density value.
2. The image processing apparatus according to claim 1, wherein the image signal converted into the equivalent neutral density is converted into a recording signal of the color toner. 3. The conversion table according to claim 1, wherein the density value of the output recording signal is fixed with respect to the density value of the recording signal input at the intermediate density value, and the intermediate value from the intermediate density value to a highlight portion and a shadow portion is fixed. 2. The image processing apparatus according to claim 1, wherein a tone conversion characteristic is set according to a contrast adjustment amount. 4. The conversion table according to claim 1, wherein the density value of the output recording signal is fixed with respect to the density value of the recording signal input in the highlight portion and the shadow portion, and the highlight portion and the shadow portion are converted from the intermediate density value. 4. The conversion characteristic of the halftone up to the above is set according to the contrast adjustment amount.
The image processing apparatus according to any one of the preceding claims. 5. The conversion table according to claim 1, wherein the density value of the output recording signal is fixed with respect to the density value of the input recording signal at a specific point, and the density before and after the density is increased or decreased according to the contrast adjustment amount. The image processing apparatus according to claim 1, wherein the conversion characteristic is set by using: 6. The apparatus according to claim 1, wherein said contrast adjustment means has a plurality of point information as data for contrast adjustment, and sets contrast adjustment characteristics in said conversion table by a polygonal line connecting said plurality of points. The image processing apparatus according to any one of the preceding claims. 7. An image processing apparatus according to claim 6, wherein said contrast adjustment data includes data of point target positions on both sides of an intermediate density value. 8. An image processing apparatus according to claim 1, wherein said contrast adjustment means combines said contrast adjustment data with a color tone conversion table for converting a color tone of an image output means.