JP2008066789A - Image processor of image forming apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor in which a grayscale conversion unit and a screen processing unit can be configured independently as desired. <P>SOLUTION: The image processor which generates image reproduction data to be supplied to the printing engine of an image forming apparatus from input grayscale data representing gray scale of an image has a grayscale conversion unit 384 which converts L-bit input grayscale data into M-bit (L<M) first grayscale data on the basis of predetermined conversion characteristics, a diffusion processing unit 385 which converts the first grayscale data into N-bit (M>N) second grayscale data by adding a random shift amount and in which the average value of grayscale values of second grayscale data obtained by converting a plurality of first grayscale data having the same grayscale value is nearly equal to the grayscale value, and a screen processing unit 386 which converts the second grayscale data into the image reproduction data representing the gray scale of the image. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image processing apparatus built in an image forming apparatus such as a printer or a copying machine, and in particular, an image that performs gradation conversion, diffusion processing, and screen processing of input gradation data corresponding to predetermined characteristics. The present invention relates to a processing apparatus.

  An image forming apparatus such as a printer has an image processing apparatus and a print engine. The image processing apparatus performs image processing on input gradation data for each pixel obtained from the image data, generates image reproduction data, and prints the image. Supply image reproduction data to the engine. The image processing apparatus performs color conversion of RGB gradation data into CMYK gradation data, which is the color of the color material of the print engine, and makes the CMYK gradation data correspond to the device characteristics of the print engine. The conversion is performed, and the gradation-converted CMYK gradation data is screen-processed to generate image reproduction data.

  If the screen processing is an AM screen (area modulation) that expresses the shading of the image by the area of the halftone dots, the input gradation data is converted to data of the dot size of the pixels constituting the halftone dots, and the dots with the same particle size In the case of an FM screen (density modulation) that expresses the density of an image with a density of, input gradation data is converted into data on whether or not to form dots in a pixel by an error diffusion method or the like. The screen-processed data is generated as image reproduction data, and if it is an electrophotographic image, it is converted into a pulse width signal by a pulse width modulator and supplied to a print engine. In the case of an inkjet, it is supplied to the print head as dot formation data. Therefore, this screen processing is called halftone processing or binarization processing, and the image reproduction data is called screen processing data or binarization data.

  According to Patent Document 1, it is described that the accuracy of gamma characteristic improvement is improved by converting 8-bit input gradation data into 10-bit output gradation data in a gradation conversion unit in the image processing apparatus. ing.

Further, according to Patent Document 2, a gamma table referred to by a halftone processing unit in an image processing apparatus is changed to a diffusion gamma table, and even if image reproduction data with a small number of bits is represented by a virtual number that cannot be represented by the number of bits. It is described that an image having a gradation value can be reproduced. The diffusion gamma table diffuses output tone values of the same index pixel group arranged on a predetermined displacement vector in a large table so that the average value matches the virtual tone value. By doing so, a large number of grayscale images can be expressed with a small number of bits.
JP 2000-56525 A JP 2006-33402 A

  In an image processing apparatus, 8-bit input grayscale data can be converted into high-precision grayscale data corresponding to desired characteristics by performing grayscale conversion into, for example, 16-bit output grayscale data having more than 8 bits. it can.

  However, when the number of bits of gradation data after gradation conversion is increased, the screen processing unit to which the gradation data is input becomes large-scale and costs increase. For example, when the screen processing unit is an AM screen, the number of look-up tables to be referenced increases and the data capacity becomes enormous. Furthermore, the circuit scale of the pulse width modulator at the rear stage of the screen processing unit is increased in size, resulting in an increase in cost. Further, when the screen processing unit is an FM screen, the error diffusion arithmetic circuit becomes large.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an image processing apparatus in which a gradation conversion unit and a screen processing unit can be independently configured in a desired configuration.

  In order to achieve the above object, according to a first aspect of the present invention, in an image processing apparatus for generating image reproduction data to be supplied to a print engine of an image forming apparatus from input gradation data indicating the gradation of an image. , A gradation conversion unit for converting L-bit input gradation data to M-bit (L <M) first gradation data based on a predetermined conversion characteristic, and N-bit (M > N) The second gradation data is converted by adding a random shift amount and converted from the plurality of first gradation data having the same gradation value. A diffusion processing unit having an average value that substantially matches the same gradation value; and a screen processing unit that converts the second gradation data into image reproduction data that reproduces the shade of the image.

  According to the first aspect of the present invention, a diffusion processing unit is provided between the gradation conversion unit and the screen processing unit, and the first gradation data that is the output of the gradation conversion unit is converted into a small number of bits. 2 is converted into gradation data. Therefore, the gradation conversion unit can have desired conversion characteristics and conversion accuracy without increasing the size of the screen processing unit. Moreover, since the number of bits is reduced by the diffusion processing unit, the accuracy of the first gradation data is converted to the second gradation data without loss, and the conversion characteristics of the gradation conversion unit can be utilized.

  In the first aspect, according to a preferred embodiment, the range of the random shift amount of the diffusion processing unit can be dynamically changed.

  In the first aspect, according to a preferred embodiment, the predetermined conversion characteristic of the gradation conversion unit is an output of the print engine corresponding to the input gradation data corresponding to the output density characteristic of the print engine. The conversion characteristics are such that the density characteristics have a linear relationship.

  In the first aspect, according to a preferred embodiment, input gradation data whose predetermined conversion characteristic of the gradation conversion unit is lower than a predetermined reference gradation value is set to a gradation value of zero, or a predetermined reference The conversion characteristic has one or both of maximizing the gradation value of input gradation data higher than the gradation value.

  In the first aspect, according to a preferred embodiment, the predetermined conversion characteristic of the gradation conversion unit is modified in accordance with an aging change of an output density characteristic of the print engine.

  In the first aspect, according to a preferred embodiment, the screen processing unit generates the image reproduction data by performing area modulation or density modulation on a gradation value of the second gradation data.

In order to achieve the above object, according to a second aspect of the present invention, in an image processing apparatus for generating image reproduction data to be supplied to a print engine of an image forming apparatus from input gradation data indicating the intensity of an image. ,
A gradation conversion unit for converting L-bit input gradation data into M-bit first gradation data based on predetermined conversion characteristics;
The first gradation data is converted by adding a random shift amount to N-bit (M> N) second gradation data and converted from a plurality of first gradation data having the same gradation value. A diffusion processing unit in which an average value of gradation values of the second gradation data substantially matches the same gradation value;
A screen processing unit for converting the second gradation data into image reproduction data for reproducing the gradation of the image,
The range of the random shift amount of the diffusion processing unit can be dynamically changed.

  According to the second aspect, the gradation conversion unit can perform gradation conversion of the input gradation data with an arbitrary conversion characteristic independently of the screen processing unit, and the degree of freedom in design while reducing the cost. Can be increased.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the technical scope of the present invention is not limited to these embodiments, but extends to the matters described in the claims and equivalents thereof.

  FIG. 1 is a configuration diagram of an image forming apparatus and an image processing apparatus built therein according to the present embodiment. The image forming apparatus in FIG. 1 is a printer apparatus 14 as an example, and a controller 16 corresponds to the image processing apparatus.

  A printer driver 11 installed in the host computer 10 supplies print data to the printer device 14. The interface 30 of the controller 16 receives the print data 12, and the CPU 32 interprets the print data by a processing program in the program memory 34, generates RGB gradation data developed on the pixels, and stores it in the memory 36. The image processing unit 38 processes RGB gradation data corresponding to this pixel to generate image reproduction data. The configuration of the image processing unit 38 will be described later. The image reproduction data is modulated by the pulse width modulation unit 40 into the pulse width data of the laser drive signal of the print engine 20 and transferred to the print engine 20. In the print engine 20, the mechanical control unit 22 controls the engine unit 24, and in the engine unit 24, a latent image is formed by driving a laser of an exposure unit (not shown) based on the transferred pulse width data. CMYK toner is attached to the latent image, developed, transferred to a recording sheet, and fixed. As a result, an image is formed on the recording paper.

  FIG. 1 is an example of electrophotography using an AM screen. In the case of using an FM screen, image reproduction data generated by error diffusion in an image processing unit is transferred to a print engine, and dots are formed in pixels based on the image reproduction data. It is formed.

  FIG. 2 is a configuration diagram of a general image processing unit. The image processing unit 38 includes a color conversion unit 382 that converts RGB gradation data into CMYK gradation data, a gradation conversion unit 384 that performs gradation conversion with predetermined characteristics, and gradation-converted gradation data. And a screen processing unit 386 for converting into image reproduction data.

  The color conversion unit 382 refers to the conversion table LUT1 from the RGB color space to the CMYK color space, and converts the 8-bit RGB gradation data into 8-bit CMYK gradation data C0, M0, Y0, K0. Convert color. The gradation conversion unit 384 refers to, for example, a gradation conversion table LUT2 having a conversion characteristic γ0 corresponding to the characteristic of the engine unit 24, and converts each 8-bit CMYK gradation data C0, M0, Y0, K0 to It is converted to 8-bit CMYK gradation data C1, M1, Y1, K1. Then, the screen processing unit 386 refers to the screen table LUT3 having the conversion tables γ1, γ2, and γ3 corresponding to the screen, and sets the 8-bit CMYK gradation data C1, M1, Y1, and K1 to one bit or a number. Bit image reproduction data C2, M2, Y2, and K2 are converted.

  FIG. 3 is a diagram showing a specific example of the gradation conversion table referred to by the gradation conversion unit 384. This gradation conversion table LUT2 has 8-bit output gradation data corresponding to 8-bit input gradation data. The conversion characteristic has an S-shape such that the output density of the image corresponding to the input gradation value is linear corresponding to the density characteristic of the engine unit, for example.

  As shown in FIG. 3, when the conversion characteristic of the gradation conversion table LUT2 has a non-linear characteristic such as an S-shape, when the 8-bit input gradation data is converted into the output gradation data having the same number of bits, the input gradation Even if the data has 255 gradations, gradation distortion occurs in the output gradation data somewhere. That is, the output gradation data includes an area where there is no gradation change and an area where the gradation change is intense and causes a tone jump. For example, when the input gradation data is “0, 1, 2, 3, 4”, the output gradation data is only “0” and “1”. Similarly, the input gradation data is “252” to “255”. Then, the output gradation data is only “254” and “255”. In the example of FIG. 3, the number of gradations taken by the output gradation data is 200, which is significantly reduced from the number of gradations 256 taken by the input gradation data.

  As described above, since the number of gradations that the output gradation data can take is smaller than 256 of the input gradation data, the gradation resolution of the output gradation data is lowered by the gradation conversion, the conversion accuracy is lowered, and the image quality is lowered. Invite.

  Therefore, it can be considered that the number of bits of the output gradation data converted by the gradation conversion unit 384 is larger than the number of bits 8 of the input gradation data. However, as a result, the number of bits of the converted gradation data C1, M1, Y1, and K1 generated by the gradation conversion unit 384 becomes large, and the scale of the screen table LUT3 to be referred to by the screen processing unit 386 to which it is input. Will increase. Alternatively, when the screen processing unit 386 is configured by an error diffusion arithmetic circuit or the like, the circuit scale increases.

  FIG. 4 is a configuration diagram of the image processing unit in the present embodiment. As in FIG. 2, the image processing unit 38 includes a color conversion unit 382 and its color conversion table LUT1, a gradation conversion unit 384 and its gradation conversion table LUT2, a screen processing unit 386 and its screen table LUT3. Have. However, the gradation conversion unit 384 converts the 8-bit input gradation data C0, M0, Y0, and K0 into the first gradation data C1, M1, and the first gradation data C1, M1, which have a higher number of bits (for example, 16 bits). Convert to Y1, K1. As a result, the gradation-converted first gradation data C1, M1, Y1, and K1 have substantially the same number of gradations as the input gradation data C0, M0, Y0, and K0. Therefore, even if gradation conversion is performed, gradation resolution does not decrease, and deterioration in image quality due to gradation conversion is avoided.

  The image processing unit 38 in FIG. 4 includes a diffusion processing unit 385 between the gradation conversion unit 384 and the screen processing unit 386. The diffusion processing unit 385 converts the 16-bit first gradation data C1, M1, Y1, K1 into the second gradation data C1d, M1d, Y1d having a smaller number of bits (for example, 8 bits or 4 bits). , K1d. The diffusion processing unit 385 converts the first gradation data C1, M1, Y1, K1 into a small number of bits and adds a random shift amount to the second gradation data C1d, M1d, Y1d, K1d. Is generated. In the diffusion processing unit 385, the average value of the gradation values of the second gradation data converted from the plurality of first gradation data having the same gradation value substantially coincides with the same gradation value. As a result, a random shift amount is generated. Specific diffusion processing will be described in detail later.

  By performing such diffusion processing, the second number of bits with a smaller number of bits can be obtained without reducing the amount of gradation information of the first gradation data C1, M1, Y1, K1 generated by the gradation conversion unit 384. The gradation data C1d, M1d, Y1d, K1d can be generated. That is, although the second gradation data has a small number of bits, it is possible to maintain the density specification of the first gradation data.

  Then, the second gradation data C1d, M1d, Y1d, K1d having a small number of bits is subjected to halftone processing or binarization processing by the screen processing unit 386, and converted into image reproduction data C3, M3, Y3, K3. The In the case of an AM screen, the screen processing unit 386 corresponds to the pixel group PX consisting of four pixels shown in FIG. 4 and uses the gamma conversion tables γ1, γ2, and γ3 to generate second gradation data C1d, M1d, Y1d and K1d are converted into image reproduction data C3, M3, Y3, and K3 having the same or less number of bits. In the case of an FM screen, the screen processing unit 386 converts it into 1-bit image reproduction data indicating whether or not to form a dot on a pixel by error diffusion calculation or the like.

  FIG. 5 is a diagram showing a specific example of the gradation conversion table LUT2 in the present embodiment. As shown in the drawing, in the gradation conversion table LUT2, 16-bit 65536 gradation output data is associated with 8-bit 256 gradation input data. Compared with the gradation conversion table of FIG. 3, the output “0, 39, 102” is assigned to the input “0, 1, 2”, and the same output gradation is not assigned as shown in FIG. , The amount of gradation information is not reduced. Similarly, the output “65416, 65504, 65535” is assigned to the inputs 253, 254, 255 ”.

  Therefore, by increasing the number of bits of the output gradation data of the gradation conversion unit 384, the gradation conversion of the input gradation data C0, M0, Y0, K0 can be performed with the desired conversion characteristics with higher accuracy.

  FIG. 6 is a diagram showing another example of the gradation conversion table in the present embodiment. In the present embodiment, the gradation conversion unit 384 converts the output gradation data having a larger number of bits than the input gradation data, so that desired conversion characteristics can be given to the converted output gradation data with high accuracy. . For example, the gradation conversion table LUT2-1 in FIG. 6 has an S-shaped gamma characteristic γ0. The gradation conversion table LUT2-2 has a background removal characteristic γ10 for converting low-density input gradation data to zero-density output gradation data. The gradation conversion table LUT2-3 has a black emphasis characteristic γ12 for converting high density input gradation data to maximum density output gradation data. Further, the gradation conversion table LUT2-4 has a characteristic γ14 for converting low-density input gradation data to density 0 and high-density input gradation data to maximum density.

  In addition to the above, when the characteristic of the output density value with respect to the input gradation data of the print engine changes due to aging, the print engine corrects the characteristic of the gradation conversion table LUT2 so as to follow the changed device characteristic. It is also possible to keep the intensity of the reproduced image constant. Changing the gradation conversion table LUT2 so as to follow the device characteristics of the engine is a well-known technical matter, as described in Patent Document 1, for example.

  As described above, in the gradation conversion unit 384, by increasing the number of bits of the output gradation data after conversion, gradation conversion can be performed with high accuracy based on arbitrary characteristics.

  Next, the diffusion process of the diffusion processing unit in the present embodiment will be described. The diffusion processing unit 385 converts the first gradation data C1, M1, Y1, and K1 converted by the gradation conversion unit 384 into a second bit having a smaller number of bits without losing the gradation information amount of the gradation data. Gradation data C1d, M1d, Y1d, and K1d.

  FIG. 7 is a configuration diagram of the diffusion processing unit 385 in the present embodiment. The diffusion processing unit converts each of the first gradation data C1, M1, Y1, and K1 into a smaller number of bits and a random number generation unit 50 that generates a random number in a range corresponding to the shift amount S. And diffusion calculators 52 to 55 that generate second gradation data C1d, M1d, Y1d, and K1d by adding a random shift amount corresponding to the random number generated by.

  FIG. 8 is a diagram showing the principle of diffusion processing of the diffusion processing unit in the present embodiment. The horizontal axis indicates 0 to 255 gradations of the input gradation data C0, M0, Y0, K0. This input gradation data is converted by the gradation conversion unit 384 using the gradation conversion table LUT2, and is converted into first gradation data C1, M1, Y1, K1 of black rectangular points. For example, when the input gradation data is “93”, the first gradation data is converted to “13011” as shown in FIG.

  On the other hand, the 16-bit first gradation data is converted into 8-bit second gradation data by the diffusion process. Since the first gradation data “13011” of the black rectangular point is located between the second gradation data “50” and “51” on the vertical axis, if the number of bits is simply reduced, The tone data becomes “50” or “51”, and the tone information included in the first tone data “13011” is lost.

  Therefore, the first gradation data of the black rectangular points is diffused to the gradation data of, for example, four black circle points d1 to d4 by diffusion processing. That is, the first gradation data is converted into second gradation data having a smaller number of bits, and a random shift amount is added to convert the first gradation data into second gradation data d1 to d4. Then, the average value of the plurality of second gradation data (black circle points d1 to d4) corresponding to the first gradation data (black rectangular points) having the same gradation value is the scale of the first gradation data. It should be almost the same as the key value. That is, an image having an area having the same density is reproduced by the pixels of the second gradation data d1 to d4, and the gradation information of the first gradation data is reproduced as a whole.

  Increasing the shift amount S increases the number of diffused black circle points, increasing the granularity of the reproduced image and making it more noisy, but the gradation information is maintained with high accuracy. On the contrary, when the shift amount S is reduced, the number of black dots that are diffused is reduced, and the granularity of the reproduced image is reduced, but the gradation information is lost.

  Next, calculation processing of the diffusion calculator will be described. Assume that the number of bits of the first gradation data after conversion of the gradation conversion table LUT2 is M bits, and the number of bits of the second gradation data subjected to diffusion processing is N bits (M> N). The high-precision gradation conversion result INT_X of the first gradation data converted by the gradation conversion table LUT2 is converted into a real number precision value DBL_X corresponding to the number of gradations of the second gradation data after the diffusion processing. Then, it is converted by the following conversion formula.

That is, the first gradation data INT_X is divided by the maximum value 2 ^M −1 of the gradation conversion table and multiplied by the maximum value 2 ^N −1 of the second gradation data.

  Next, a uniform random number Rs is obtained in the range 0 to S of the shift amount S, and an integer precision value DIFF_X is obtained from the real precision value DBL_X by the following equation (2).

Here, the function INT is truncated, and the function RAND is a random number generation function in the range of 0 to S. By adding the random number Rs to the real precision value DBL_X and subtracting S / 2, the upper and lower random numbers (Rs−S / 2) are added to the real precision value DBL_X, and 0.5 is added. The rounding process is performed by performing the rounding process INT. That is, the real number precision value DBL_X is diffused in a range of ± S / 2.

  Since Rs is a random number, even if the first gradation data having the same gradation value is converted into the plurality of diffused second gradation data, the average value of the plurality of second gradation data is the first value. It becomes almost equal to the value of the real number accuracy of the gradation value of 1.

  An explanation will be given using the gradation conversion table of FIG. 5 as an example. In this example, the first gradation value is 16 bits, and the second gradation value after diffusion is 8 bits. The first tone value obtained by tone conversion of the input tone value 93 of the tone conversion table LUT2 is “13011”. Therefore, according to the above equation (1), it is as follows.

That is, the real number precision value DBL_X is 50.625, and is a value between “50” and “51” in 8-bit conversion.

  Next, assuming that the random amount is generated by setting the shift amount S to “3” and the random number Rs = 2.476387 is obtained, the second gradation data diffused by the above equation (2) is as follows: Become.

That is, for the random number Rs = 2.476387, the second gradation data is “52” (black circle d1 in FIG. 8).

  Similarly, when the random number Rs is generated for the 50 first gradation data having the same gradation value and the second gradation data DIF_X subjected to the diffusion process is obtained, the second floor data is obtained. The average value of the tone data was 50.68. This average value is substantially the same as the real number precision value DBL_X “50.625” obtained by converting the first gradation data “13011” into 8 bits.

  FIG. 9 is a diagram showing an experimental result of the diffusion processing in the present embodiment. FIG. 9A shows the result when the input gradation data is “93” when the second gradation data after diffusion is 8 bits and the shift amount S is 3 as described above. . The horizontal axis represents the number of data points 0 to 50, and the vertical axis represents the output gradation value subjected to the diffusion processing. A white circle indicates the diffusion value DBL_X with real number precision, and a black circle indicates the diffusion value DIFF_X with integer precision. The average value of the real number precision (dashed line) and the average value of the integer precision (dashed line) are almost the same. Since the shift amount S is 3, it is diffused to the four gradation values “49, 50, 51, 52” of the random number Rs (= 0-3).

  In the diffusion processing unit of the present embodiment, the number of bits of the second gradation data subjected to the diffusion processing can be set to 4 bits. However, when the number of bits is small, the number of gradation values that can be taken by the second gradation data is also small, and the range of influence of the diffusion effect on the entire gradation value is widened. Therefore, it is appropriate to reduce the shift amount S, for example, 1. The above example shows the diffusion calculation when the number of bits of gradation data after diffusion is 4 and the shift amount S is 1.

  First, the real number precision diffusion value DBL_X is as follows, and 2.978 is obtained.

  Next, assuming that a random number is generated with a shift amount S = 1 and a random number Rs = 0.825462 is obtained, the integer-precision diffusion value DIFF_X is as follows, and “3” is obtained.

  Similarly, FIG. 9B shows the result of calculating the diffusion value by generating random numbers for 50 points. According to this experiment, the average value of the integer precision diffusion values (black circles) is 2.96, which is substantially the same as the average value of the real number precision diffusion values (white circles). Since the shift amount S = 1, the random number Rs is generated in the range of 0 to 1, and the second spread data is spread to either “2” or “3”. However, as the number of data increases, it is clear that the average value of integer-precision spread values approaches the real-precision spread value as much as possible.

  Returning to FIG. 4, the image processing unit 38 converts the second gradation data C1d, M1d, Y1d, K1d having a small number of bits subjected to diffusion processing into image reproduction data C2, having a predetermined number of bits by the screen processing unit 386. Converted to M2, Y2, and K2. As shown in FIG. 4, the screen table LUT3 is composed of a reference table having three characteristics γ1, γ2, and γ3. For the four cross-shaped pixels PX, the middle pixel has a characteristic γ1 conversion table. However, the remaining four pixels are associated with the conversion table of the characteristic γ2 or γ3, and the second gradation data C1d, M1d, Y1d, K1d of the pixel are converted into the image reproduction data C2, M2, Y2, K2. The Since the number of bits of the second gradation data is small, the data amount of the conversion table is small.

  Alternatively, the second gradation data C1d, M1d, Y1d, K1d of the pixel is converted into image reproduction data C2, M2, Y2, K2 by error diffusion calculation. Also in this case, since the number of bits of the second gradation data is small, the operation circuit scale of the error diffusion calculation is also reduced.

  In either case, since the number of bits of the second gradation data is reduced, the scale of the screen processing unit 386 including the screen table LUT3 can be reduced, and the cost can be reduced.

  In the present embodiment, it is desirable that the shift amount S in the diffusion process can be dynamically changed. As described above, by increasing the shift amount S, it is possible to maintain the gray level of the image reproduced by the second gradation data after the diffusion processing with high accuracy. However, on the other hand, when the shift amount S is increased, the image becomes noticeable and noisy. This graininess depends on the number of bits of the second gradation data after the diffusion processing, and also depends on a change in the intensity of the image. Accordingly, after confirming the image to be reproduced with the shift amount S or by predicting the image to be reproduced, if the change can be dynamically set, the optimum image can be reproduced. For example, the user can set the shift amount to a desired value via the external interface. In this case, it is desirable that the user is shown a range of shift amounts that can be set depending on the number of bits of the second gradation data after the diffusion processing.

  As shown in FIG. 4, in this embodiment, a diffusion processing unit 385 is provided independently between the gradation conversion unit 384 and the screen processing unit 386. Therefore, the gradation conversion unit 384 can convert the input gradation data C0, M0, Y0, K0 into the first gradation data C1, M1, Y1, K1 based on an arbitrary conversion characteristic. In this case, the conversion accuracy can be improved by increasing the number of bits of the first gradation data. On the other hand, since the number of bits of gradation data can be reduced without losing high-precision gradation information by the diffusion processing unit 385, the configuration of the screen processing unit 386 in the subsequent stage can be simplified and the cost can be reduced. Can be achieved.

  In the above embodiment, the gradation conversion unit 384 converts the input gradation data C0, M0, Y0, K0 into the first gradation data C1, M1, Y1, K1 having a larger number of bits. However, even when converted to the first gradation data having the same number of bits or a small number of bits, the diffusion processing unit 385 in which an arbitrary shift amount S can be set allows the second floor having a smaller number of bits under optimum diffusion conditions. Since the data can be converted into tone data C1d, M1d, Y1d, and K1d, the cost of the screen processing unit at the subsequent stage can be reduced.

1 is a configuration diagram of an image forming apparatus and an image processing apparatus built therein according to the present embodiment. It is a block diagram of a general image processing unit. It is a figure which shows the specific example of the gradation conversion table which a gradation conversion unit refers. It is a block diagram of the image processing unit in this Embodiment. It is a figure which shows the specific example of the gradation conversion table LUT2 in this Embodiment. It is a figure which shows the other example of the gradation conversion table in this Embodiment. It is a block diagram of the spreading | diffusion processing unit 385 in this Embodiment. It is a figure which shows the principle of the spreading | diffusion process of the spreading | diffusion processing unit in this Embodiment. It is a figure which shows the experimental result of the spreading | diffusion process in this Embodiment.

Explanation of symbols

382: Color conversion unit 384: Gradation conversion unit 385: Diffusion processing unit 386: Screen processing unit S: Shift amount

Claims (7)

In an image processing apparatus for generating image reproduction data to be supplied to a print engine of an image forming apparatus from input gradation data indicating the intensity of an image,
A gradation conversion unit for converting L-bit input gradation data to M-bit (L <M) first gradation data based on predetermined conversion characteristics;
The first gradation data is converted by adding a random shift amount to N-bit (M> N) second gradation data and converted from a plurality of first gradation data having the same gradation value. A diffusion processing unit in which an average value of gradation values of the second gradation data substantially matches the same gradation value;
An image processing apparatus comprising: a screen processing unit that converts the second gradation data into image reproduction data that reproduces the gradation of an image. In claim 1,
An image processing apparatus, wherein a range of a random shift amount of the diffusion processing unit can be dynamically changed. In claim 1,
The predetermined conversion characteristic of the gradation conversion unit is a conversion characteristic such that the output density characteristic of the print engine corresponding to the input gradation data has a linear relationship corresponding to the output density characteristic of the print engine. An image processing apparatus comprising: In claim 1,
Input gradation data whose predetermined conversion characteristic of the gradation conversion unit is lower than a predetermined reference gradation value is set to a gradation value of zero, or input gradation data higher than a predetermined reference gradation value is converted to a gradation value. An image processing apparatus having a conversion characteristic having either one or both to maximize. In claim 1,
The image processing apparatus according to claim 1, wherein the predetermined conversion characteristic of the gradation conversion unit is modified in accordance with a secular change in the output density characteristic of the print engine. In claim 1,
The image processing apparatus, wherein the screen processing unit generates the image reproduction data by performing area modulation or density modulation on a gradation value of the second gradation data. In an image processing apparatus for generating image reproduction data to be supplied to a print engine of an image forming apparatus from input gradation data indicating the intensity of an image,
A gradation conversion unit for converting L-bit input gradation data into M-bit first gradation data based on predetermined conversion characteristics;
The first gradation data is converted by adding a random shift amount to N-bit (M> N) second gradation data and converted from a plurality of first gradation data having the same gradation value. A diffusion processing unit in which an average value of gradation values of the second gradation data substantially matches the same gradation value;
A screen processing unit for converting the second gradation data into image reproduction data for reproducing the gradation of the image,
The range of the random shift amount of the diffusion processing unit can be dynamically changed.

Images