JP2007020117A - Image forming apparatus suppressing texture - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image forming apparatus including a threshold matrix capable of suppressing generation of texture. <P>SOLUTION: In an image forming apparatus including an image processing unit 14 and a printing unit 20, the image processing unit 14 includes: an original threshold matrix SCR2 disposing threshold values to dispose increasing dots in a distributed manner in accordance with increase of dots of a dot group regarding the relevant dot group on a displacement vector corresponding to the number of screen lines; a correction unit 15 for correcting the original threshold matrix based on density characteristics in a relationship between image reproducing data in the printing unit and an output image density and for generating a normalized threshold matrix SCR4 resulting from normalizing the corrected original threshold matrix in accordance with the number of gradations of input gradation data; and a screen processing unit 16 for generating image reproducing data from the input gradation data based on the normalized matrix. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image forming apparatus having an image processing unit that generates image reproduction data indicating the presence / absence of dots from multi-tone input gradation data, and in particular, an image in which the texture of an image generated by the image reproduction data is suppressed. The present invention relates to a forming apparatus.

  An image forming apparatus using electrophotographic technology such as a printer or a copier incorporates an image processing unit that generates image reproduction data (binary data) indicating the presence or absence of dots from input multi-gradation data. Such an image processing unit converts input multi-gradation data into image reproduction data by screen processing. The screen processing is generally performed by a dither method, referring to a threshold matrix in which thresholds are arranged on a two-dimensional matrix, comparing input gradation values with corresponding thresholds in the threshold matrix, and depending on the comparison result. Then, image reproduction data with or without dots is generated. The image reproduction data is subjected to pulse width conversion and supplied to the print engine, and an image is formed by electrophotographic technology.

  Since the positions of the dots of the image to be formed differ according to the size of the threshold matrix and the arrangement of the thresholds, the threshold matrix is called a screen. Therefore, image formation characteristics of the image forming apparatus, such as gradation and resolution, depend on the design of the threshold matrix that is the screen.

  In general, a threshold matrix is designed based on the desired number of screen lines. For example, dot formation positions that increase for each gradation according to a predetermined number of screen lines are specified. The dot formation position is specified by a displacement vector, and the number of gradations that can be generated is determined by the outer product of the displacement vector. When the displacement vector is (4, 4) (−4, 4), the outer product is 4√2 × 4√2 = 32 and the number of gradations is “32”.

Generally, such a threshold matrix is designed for a dot-concentrated AM screen. That is, 32 dots are generated while being concentrated in response to an increase in the input gradation value. Such an AM screen is described in Patent Document 1.
JP 2003-152999 A (published May 23, 2003) Japanese Patent Application No. 2005-1000042 (filed on March 31, 2005)

  However, if the resolution is improved by narrowing the interval between the screen lines, the outer product is reduced and the number of gradations is lowered. Such a decrease in gradation means that it is impossible to reproduce all the gradation numbers of the input gradation data. Therefore, it is conceivable to improve gradation by shifting the growth of dots in the plurality of basic matrices by using an enlarged threshold matrix in which a plurality of basic matrices are repeatedly arranged.

  However, in the case of such an enlarged threshold matrix, the number of screen lines changes because the growth of dots in a plurality of basic matrices is shifted as the input gradation value increases. This change in the number of screen lines causes generation of a texture (interference pattern) that can be easily recognized by human vision.

  The present applicant has previously filed a method for generating a threshold matrix for suppressing the generation of the texture (Patent Document 2). According to this method, the arrangement position of a plurality of dots in a supercell is determined by granularity evaluation. However, evaluating all the arrangement examples where a plurality of dots are possible has a large number of arrangement examples and requires a large number of calculation steps. Becomes enormous and requires several days of computation, which is not realistic. In addition, since the threshold matrix once generated is designed so that a plurality of dots grow at the same time with respect to an increase of one gradation, the threshold matrix is set corresponding to the change in the characteristics of the print engine of the image forming apparatus. It is difficult to correct.

  Accordingly, an object of the present invention is to provide an image forming apparatus having a threshold matrix that can suppress the occurrence of texture.

In order to achieve the above object, according to a first aspect of the present invention, an image processing unit for generating image reproduction data indicating the presence or absence of dots from input gradation data, and an image based on the image reproduction data are provided. In an image forming apparatus having a printing unit for forming,
In the image processing unit, for the dot group on the displacement vector corresponding to the number of screen lines, the threshold is arranged so that the increasing dots are distributed in correspondence with the increase of each dot in the dot group. The original threshold matrix is corrected based on a threshold matrix and a density characteristic having a relationship between image reproduction data and output image density in the printing unit, and the corrected corrected original threshold matrix is converted to a scale of the input gradation data. A correction unit that generates a normalized threshold matrix that is normalized according to the logarithm; and a screen processing unit that generates the image reproduction data from the input gradation data based on the normalization matrix.

  According to the above image forming apparatus, since the original threshold value matrix is arranged in such a manner that the positions of the dots of the dot group on the displacement vector are distributed corresponding to the dot increase, the occurrence of texture can be suppressed. Gamma correction is performed in accordance with the density characteristics of the printing unit based on the original threshold matrix, and screen processing is performed using the normalized threshold matrix, so that an image can be formed with the target density characteristics while suppressing texture.

In order to achieve the above object, according to a second aspect of the present invention, an image processing unit that generates image reproduction data indicating the presence or absence of dots from input gradation data, and an image based on the image reproduction data are provided. In an image forming apparatus having a printing unit for forming,
In the image processing unit, for the dot group on the displacement vector corresponding to the number of screen lines, the threshold is arranged so that the increasing dots are distributed in correspondence with the increase of each dot in the dot group. A threshold value matrix; a normalization unit that generates a normalized threshold matrix obtained by normalizing the original threshold value matrix corresponding to the number of gradations of the input gradation data; and the input gradation data based on the normalized matrix. A screen processing unit for generating the image reproduction data.

  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 diagram illustrating an example of a basic matrix that is a basic unit of a threshold matrix and an enlarged threshold matrix. FIGS. 1A and 1B show dots having gradation values “1” and “2” generated by the threshold matrix of basic units, and thresholds “0” and “1” are located at the positions of the dots in the threshold matrix. Are arranged respectively. This threshold matrix is based on the displacement vector (4, 4) (−4, 4) set with a screen line number of 106 lpi, and thresholds “0” and “1” are 2 in the matrix of 8 × 8 pixels, respectively. The other threshold values are also arranged in two places. In addition, a small threshold value starting from the minimum threshold value is arranged adjacent to the pixel of the minimum threshold value to constitute a dot concentration type AM screen.

  In the case of FIGS. 1A and 1B, since 8 × 8 = 64 dots grows by 2 dots every time one gradation is increased, the number of gradations is “32”. Therefore, compared with 256 gradations of 8-bit input gradation data, the number of gradations is considerably small as “32” and the gradation is not good. However, since a dot is generated at the position of the displacement vector, the screen line number 106 lpi is maintained and the resolution is good.

  FIGS. 1C to 1H show dot generation by an enlarged threshold matrix in which the number of gradations is set to “128” while basically maintaining the number of screen lines in FIGS. 1A and 1B. . Here, every time the gradation value increases, a dot at the position of the arrow is generated. FIGS. 1C to 1H show dots having gradation values “1” to “6”, and threshold values “0” to “5” are arranged at the positions of these dots. This enlarged threshold matrix is composed of 16 × 16 = 256 pixels in order to express 128 gradations, and is obtained by repeatedly arranging four basic matrices shown in FIGS. In order to increase the number of gradations from “32” to four times “128”, the gradation number “1” of the basic matrix corresponds to the gradation number “4” of the expansion matrix. In other words, four basic matrices are arranged in the enlarged matrix, and 8 dots are generated at one gradation in the basic matrix, and 2 dots are generated at 1 gradation in the enlarged matrix to generate all 8 dots in 4 gradations. To be completed.

  In the case of the enlarged matrix shown in FIGS. 1C to 1H, the number of screen lines is 53 lpi (600 dpi / 8√2) at (C) with the gradation value “1”, and the (D) at the gradation value “2”. In (E) with a screen number of 75 lpi (600/8) and a gradation value of “3”, the screen line number of 106 lpi (600 / 4√2). That is, the screen line number fluctuates as the gradation value increases, which causes texture generation.

  The cause of the texture generation is that dots are regularly generated at the positions on the displacement vector based on the number of screen lines as the gradation increases. Therefore, the above-mentioned Patent Document 2 proposes to suppress the generation of texture by making the generation of dots irregular and minimizing graininess. That is, every time the gradation is increased, a plurality of dots are dispersed and irregularly arranged to form a supercell, thereby suppressing texture generation.

  However, according to this, in order to determine the position of the dot group generated as the gradation increases, the granularity is evaluated by the granularity evaluation function for each combination example of the dot group position, and the granularity is determined. The search is for a combination of dot group positions that is minimized, and the calculation time for the search is enormous, which is not realistic.

  Therefore, in this embodiment, instead of searching for the position of the dot group that minimizes the granularity, the search time for the search is reduced by searching for the position of a single dot that minimizes the granularity. shorten. Then, an original threshold matrix that enables generation of dots with suppressed graininess is corrected according to the density characteristics of the image forming apparatus.

  FIG. 2 is a configuration diagram of the image forming apparatus according to the present embodiment. A printer driver PDR is installed in the host computer 10 and supplies print data to the printer 12 which is an image forming apparatus. The printer 12 includes an image processing unit 14 and a print engine 20 that is a printing unit. The image processing unit 14 includes an interface IF that receives print data from the printer driver PDR, a central processing unit CPU, and image reproduction data (binarized data) indicating the presence or absence of dots in the input gradation data included in the print data. 17 has a screen processing unit 16 for converting to 17 and a PWM unit 18 for generating a pulse drive signal 19 by pulse width modulating the image reproduction data 17. The screen processing unit 16 performs binarization processing by the dither method, and refers to the normalized threshold value matrix SCR4 stored in the memory to convert the input gradation data into the image reproduction data 17 that is binarized data. Convert. A pulse drive signal 19 based on the image reproduction data drives a light source in the print engine 20 to form a desired image. The print engine 20 outputs the print medium 23 on which an image is formed.

  The image processing unit 14 also has an original threshold value in which the threshold value is arranged so that the dot group on the displacement vector corresponding to the desired number of screen lines is distributed for each dot that increases as the number of dots increases. The matrix SCR2 (screen SCR2) is stored in the memory. The original threshold matrix SCR2 is obtained at the design stage by a method described later and stored in a memory in the image processing unit 14 of the printer 12.

  Then, a calibration process using gradation patch data is executed, and the correction unit 15 corrects the original threshold value matrix SCR2. That is, the patch pattern generated by the print engine 20 is measured by the output density measuring device 21, and the measurement result 22 is supplied to the correction unit 15. Based on the measurement result, the correction unit 15 corrects the original threshold value matrix SCR2 so as to obtain an ideal output density. Then, the corrected original threshold value matrix is normalized corresponding to the gradation value of the input gradation data, and a normalized threshold value matrix SCR4 corresponding to the output density characteristic of the print engine is generated. The above calibration process is executed at initialization or periodically or irregularly according to the aging of the print engine, and the normalization threshold matrix SCR4 is maintained so that an ideal output characteristic is always obtained. The

  3 and 4 are process diagrams showing the relationship between the original threshold value matrix and the normalized threshold value matrix which are screens in the present embodiment. The original threshold value matrix SCR2 and the normalized threshold value matrix SCR4 in the present embodiment will be described with reference to FIGS. In the figure, a basic matrix SCR1, an original threshold matrix SCR2, an original threshold matrix SCR3 corrected by the calibration process, and a normalized threshold matrix SCR4 normalized with the number of gradations of input gradation data are shown.

  3 and 4, the original threshold matrix SCR2 mounted on the image processing unit of the printer is generated by screen design (S12). Then, at the time of initialization of the printer or at a predetermined time, a calibration process is executed based on the original threshold value matrix SCR2, and a corrected original threshold value matrix SCR3 is once generated (S22). The threshold value matrix SCR4 is updated (S24).

  Next, this will be described in detail with reference to FIGS. First, in the screen design, a basic matrix SCR1 is generated based on a desired number of screen lines (S10). That is, the screen line number is determined according to the resolution required for the image forming apparatus, and a basic matrix is generated based on the screen line number.

  FIG. 5 is a diagram showing the relationship between the number of screen lines and the basic matrix. In the screen design, a basic matrix SCR1 for realizing a desired number of screen lines is generated. By dividing the dot density of the image forming apparatus by the screen line number, the distance between the screen line numbers is obtained. For example, as shown in FIG. 5A, when the screen lines SL1 and SL2 are arranged at a distance W corresponding to the determined number of screen lines (number of screen lines per inch), this distance W is set to 1. It is necessary to generate dots at the vertices of the rectangle that is the side. As a result, the displacement vectors B1 and B2 that specify the positions where the dots should be generated are determined. Here, the displacement vector is a vector that specifies the position of an adjacent dot with a certain dot as a reference. In the example of FIG. 5A, the displacement vector B1 is (4, 4) and the displacement vector (-4, 4). Therefore, with reference to the dot DT1, the dot DT2 is generated at the position of the displacement vector B1 (4, 4), and the dot DT3 is generated at the position of the displacement vector B2 (-4, 4). The dot DT4 is at the position of the displacement vector B1 with respect to the dot DT3, and is at the position of the displacement vector B2 with respect to the dot DT2.

  In this screen, a dot group at a position specified by a displacement vector is generated every time one gradation is increased. Therefore, the outer product (4√2 × 4√2 = 32) of the displacement vectors B1 and B2 corresponding to the area of the quadrangle becomes the gradation number “32”. That is, the number of dots that can be generated in the quadrilateral within the four dots DT1 to DT4 corresponds to the number of gradations. The positions of the 32 types of dots are all arranged on the displacement vectors B1 and B2.

  FIG. 5B shows a basic matrix SCR1 of the screen designed from the above screen line number. This is the same as FIGS. 1A and 1B. The basic matrix SCR1 has a total size of 64 pixels of 8 × 8 pixels. FIG. 5B shows the position of the dot generated for the gradation value “1” and the position of the dot (arrow position) generated for the gradation value “2”. The basic matrix SCR1 in which is arranged has thresholds “0” and “1” at the positions of the dots. In the example of FIG. 5B, 2 dots are generated every time the gradation is increased, and the number of gradations is 32 (= 64/2). That is, as shown in FIGS. 3 and 4, in the basic matrix SCR1, threshold values corresponding to the gradation values 0 to 31 are arranged at the positions of the displacement vectors.

  Next, the size of the enlarged matrix in which a plurality of basic matrices are repeatedly arranged is determined (S11 in FIGS. 3 and 4). The size of the expansion matrix corresponds to the scale of the supercell in this embodiment, and can be set to an arbitrary size in the screen design process.

  FIG. 6 is a diagram illustrating an example of an enlarged matrix. FIG. 6 shows the positions of dots generated by the expansion matrix when the gradation value is “1” and the positions of dots generated by the expansion matrix when the gradation value is “31”. In the enlarged matrix SCR1EX, 144 basic matrixes SCR1 in the horizontal direction and 12 in the vertical direction, 144 in total, are repeatedly arranged. Therefore, the size is 96 pixels × 96 pixels. In the gradation value “1”, since two dots are generated in each basic matrix SCR1, 288 (2 dots × 144) dots are generated on the displacement vector in the enlarged matrix SCR1EX. Similarly, at the next gradation value “2”, 288 dots are generated at the position of the arrow shown in FIG. That is, the dot corresponding to the next gradation value is generated around the dot generated at the gradation value “1”, and the screen is a dot-concentrated screen that maintains the number of screen lines. Then, at the gradation value “31”, the last 288 dots are generated and all the dots are generated. As a result, all dots are generated in the enlarged matrix having the gradation value “31”.

  In the present embodiment, the positions of 288 dots generated for each gradation in this enlarged matrix are determined so that the granularity becomes the smallest when a single dot of 288 dots is generated. To go. That is, for 288 dots corresponding to the gradation value “1” located on the displacement vector, the granularity when each dot is generated is evaluated by the granularity evaluation function described later, and the granularity is the smallest. A threshold value is assigned so that a dot is generated at a position (S12 in FIGS. 3 and 4). As shown in FIG. 4, threshold values “0” to “287” are assigned to 288 dots at the gradation value “1” in the gradation number 32. Further, threshold values “288” to “575” are assigned to 288 dots in the gradation value “2”. Further, threshold values “576” to “863” are assigned to 288 dots at the gradation value “3”. Then, threshold values “8928” to “9215” are assigned to 288 dots at the maximum gradation value “31”. As a result, threshold values “0” to “9215” are assigned to 96 pixels × 96 pixels = 9216 pixels of the enlarged matrix. As a result, an original threshold matrix SCR2 is generated.

  FIG. 7 is a diagram showing an arrangement example of 288 dots in the original threshold matrix SCR2. Threshold values (Vth) “0” to “287” are given to the positions of 288 dots. The position of the 288 dots is determined by evaluating which position has the minimum granularity among the 288 positions on the displacement vector by using the granularity evaluation function. Therefore, based on the granularity evaluation function, the position of the threshold value “0” is selected from the 288 candidate positions with the minimum granularity. As the position of the threshold “1”, one having the minimum granularity is selected from the 287 candidate positions. Then, the position of the threshold value “2” is selected from the 286 candidate positions that has the minimum granularity. Similarly, one of the threshold values “3” to “287” is selected from the candidate positions of 285 to 1 that gradually decrease, each having the minimum granularity. The positions indicated by the arrows in FIG. 7 are the positions where the minimum granularity is obtained, and the corresponding threshold values are arranged at these positions, and as a result, the generated dots become the positions of the arrows. As described above, when searching for the position of the dot having the minimum granularity, it is only necessary to search for the position having the minimum granularity from 288 types of candidate positions at the maximum. Therefore, the calculation time man-hour required for the search is not so high.

  The evaluation formula E used in the granularity measurement is as follows.

ここで，

here,

である。

It is.

  The above-described VTF (Visual Transfer Function) is an approximate function indicating the sensitivity of human vision to the spatial frequency component included in the image. Further, m and n indicate the spectral orders in the coordinate position x and y directions of the threshold matrix, and Am and n indicate the energy of the frequency located at the coordinate position (m, n). Fm, n indicates a VTF obtained by projecting a two-dimensional plane to one dimension, rate is a correction factor of VTF, reso is a printing resolution, T is a vertical and horizontal size of the matrix, and reso / T is a fundamental frequency of the screen. Indicates. INT is a function that rounds off the decimal point to an integer.

  As the evaluation value E of the granularity evaluation function E is smaller, the image has a frequency characteristic that is less perceptible to human vision. Therefore, the smaller the evaluation value E is, the more concentrated the energy is on the high frequency component, and the lower the frequency component is, the smaller the granularity is. In other words, small granularity means that dots are dispersed and arranged, and there are few low frequency components and many high frequency components in the spatial frequency components of the image. The possibility of texture generation is reduced, and the image quality can be improved.

  Therefore, the position of 288 dots to be generated at each gradation value shown in FIG. 7 is a position where the granularity evaluation value E for the image after generation for each dot is the minimum value from among the candidate positions. To be determined. This dot position search is performed for all 9216 positions having threshold values “0” to “9215”. The calculation man-hour for determining the position is not so large because there are 288 candidate positions for each dot at most.

  As a result, the original threshold value matrix SCR2 in which the threshold values “0” to “9215” are assigned in the 96 × 96 pixel enlarged matrix is generated. This original threshold value matrix SCR2 is a screen without a supercell because 288 dots are generated at the displacement vector position for each gradation for the gradation number 32. The original threshold value matrix SCR2 is designed so that the position of 9216 dots has the minimum granularity of the image in which each dot is generated. Therefore, if screen processing is performed using this original threshold matrix, dots can be grown without changing the number of screen lines, and the granularity is selected to be the minimum value. Therefore, it is possible to form an image having a larger number of gradations with a desired number of screen lines and suppressing the occurrence of texture.

  As shown in FIGS. 2 and 3, the original threshold value matrix SCR2 generated by the screen design is mounted on the image processing unit of the printer as the image forming apparatus (S13). Then, a calibration process is performed using the patch pattern formed by the original threshold matrix SCR2, and the threshold matrix is corrected so that an ideal density characteristic can be obtained (S20, S22).

  In the calibration step (S20), a patch pattern image is formed using a threshold matrix in which the thresholds of the original threshold matrix SCR2 are normalized to the number of gradations “256” of the input gradation data. In other words, when the input gradation data consists of 8 bits, the number of gradations is “256”. Therefore, by normalizing the original threshold value matrix to this number of gradations, the input gradation data is compared with the threshold value. And binarized data can be generated. Therefore, the threshold values “0” to “9125” of the original threshold value matrix SCR2 are normalized to the threshold values “0” to “255”. This normalization is performed by the following arithmetic expression.

すなわち，オリジナル閾値マトリクスＳＣＲ２の閾値ＴＨに２５５／９１２５を乗算することで，閾値「０」〜「９１２５」を閾値「０」〜「２５５」に正規化することができる。ただし，乗算結果ＴＨ₂₅₅が小数点以下を有する場合は切り上げられる。

That is, by multiplying the threshold value TH of the original threshold value matrix SCR2 by 255/9125, the threshold values “0” to “9125” can be normalized to the threshold values “0” to “255”. However, when the multiplication result TH ₂₅₅ has a decimal part, it is rounded up.

  FIG. 8 is a diagram illustrating an example of a normalized original threshold matrix SCR2NORM. The normalized gradation value “1” corresponds to the gradation value “36” in the original threshold matrix, and the normalized gradation values “2”, “3”, “4”, and “5” are gradations, respectively. It corresponds to the values “72” “108” “144” “180”. The normalized threshold “255” corresponds to the original threshold “9125”. Therefore, according to the normalized threshold value matrix, 36 dots are generated every time one gradation is increased. Moreover, since the positions of each of the 36 dots change irregularly, the normalized threshold value matrix SCR2NORM becomes a supercelled screen. The positions of the dots based on the original threshold matrix are all set at positions where the granularity is minimized. Therefore, the 36 dot positions based on the normalized threshold value matrix SCR2NORM are also arranged at positions where the granularity becomes small. In addition, since each of the 32 dots is always arranged at 288 positions on the displacement vector, there is no variation in the number of screen lines.

  The patch pattern data is screen-processed by the threshold matrix SCR2NORM thus normalized, the patch pattern is generated by the printing unit 20, the output density is detected by the output density measuring device 21 (see FIG. 2), and the detected measurement is performed. The density 22 is given to the correction unit 15 (S20 in FIG. 3).

  FIG. 9 is a diagram illustrating measured density characteristics and target density characteristics. In FIG. 9A, density characteristics obtained from patch patterns generated by binarizing the patch pattern data using the normalized threshold matrix SCR2NORM are plotted with black squares. That is, patch pattern output density (optical density) values 0 to 255 with respect to input gradation values 0 to 255 are shown. The target ideal density characteristics are indicated by black circles for the black square measurement results. The target density has a slightly convex characteristic.

  The correction unit 15 obtains a gamma correction table so that the measured density characteristic becomes the target density characteristic, corrects the threshold value of the original threshold value matrix SCR2 based on the gamma correction table, and corrects the corrected original threshold value. A matrix SCR3 is generated (S22 in FIGS. 3 and 4). By performing gamma correction not on the normalized threshold matrix SCR2NORM but on the original threshold matrix SCR2 having a resolution of thresholds “0” to “9215”, correction can be performed without reducing the number of gradations.

  In the gamma correction described above, the correction unit 15 normalizes the gradation values “0” to “255” of the density characteristic in FIG. 9A to “0” to “9215” to obtain the density in FIG. 9B. Find characteristics and target concentration. That is, the horizontal axis and the vertical axis in FIG. 9B are converted to 9216 gray levels. A gamma correction table is obtained from the density characteristics and the target density.

  FIG. 10 is a diagram illustrating a method for obtaining a gamma correction table. In the first quadrant, an input tone value (threshold value) is assigned to the horizontal axis, and an output density is assigned to the vertical axis, and a measured density characteristic 30 of the print engine 20 is shown. In the second quadrant, the horizontal axis represents the correction threshold value and the vertical axis represents the output density, and the target density characteristic 32 is shown. In the third quadrant, a fold line 36 is simply shown. A correction curve 34 is shown in the fourth quadrant. As shown in the first quadrant, an output density OUT for a certain input gradation value IN is obtained by measuring a patch pattern. However, according to the target density characteristic 32 in the second quadrant, the input gradation value CIN is required to be a larger value in order to obtain the output density OUT. Therefore, by changing the threshold value IN of the threshold value matrix to a larger correction threshold value CIN, the input tone value IN can be handled as a substantially smaller input tone value, and as a result, the density characteristics of the print engine 20 can be handled. Can be matched with the target density characteristic 32.

  That is, as shown in FIG. 4, the correction unit 15 corrects the thresholds of the original threshold matrix SCR2 based on the gamma correction table 34 of FIG. 10 obtained from the density characteristics of the printer, and generates a corrected original threshold matrix SCR3. (S22). According to the correction table 34 of FIG. 10, the low threshold value is corrected largely, and the high threshold value is corrected slightly large. In addition, by using density characteristics normalized to the threshold values “0” to “9215”, it is possible to generate a corrected original threshold value matrix without reducing the number of gradations.

  In the above correction, for example, the input gradation value “7” when the gradation number is “256” becomes the normalized input gradation value IN “252.96”, and the normalized output density OUT at that time is “ 397.65 ”(“ 11 ”for the number of gradations“ 256 ”), and the input gradation value CIN for obtaining the normalized target density“ 397.65 ”is“ 728.4 ”. Therefore, the correction table needs to be corrected so that the threshold value “252.96” before correction becomes the threshold value “728.4” after correction. In this way, the correction gamma table 34 is a table of gradation values for which the output densities “0” to “9215” should be obtained. Therefore, the corrected original threshold value matrix SCR3 is generated by correcting the threshold value of the original threshold value matrix SCR2 according to the corrected gamma table 34.

  Then, the correction unit 15 normalizes the corrected original threshold matrix SCR3 to thresholds “0” to “255” corresponding to the number of gradations “256” of the input gradation data, and the normalized threshold matrix SCR4 is Stored in memory. In this normalization step, the threshold values “0” to “35” are set to the threshold value “1”, the threshold values “36” to “71” are set to the threshold value “2”, and the threshold values “72” to “107” are set. The threshold value is set to “3”, and similarly, the remaining threshold values are normalized to a threshold value of 256 gradations. The screen processing unit 16 in the image processing unit 14 refers to the updated normalized threshold value matrix SCR4 and binarizes the input gradation data to generate the image reproduction data 17, and the printing unit 20 thereby forms an image. I do.

  FIG. 11 is a diagram showing an example of the normalized threshold value matrix SCR4 in the present embodiment. Compared to the normalized threshold matrix SCR2NORM before correction in FIG. 8, the number of pixels having threshold values “0”, “1”, “2”, “3”, and “4” is reduced. That is, the threshold value in the original threshold value matrix is changed to a larger threshold value by the correction by the correction curve 34 shown in FIG. Even if corrected in this way, the dots generated at each gradation value are limited to positions on the displacement vector, and the granularity after dot generation is designed to be small. Therefore, it is possible to realize a screen that achieves both resolution and gradation and suppresses the generation of texture.

  As described above, according to the present embodiment, when designing the original threshold matrix, the position of each dot in the enlarged matrix is searched based on the granularity evaluation accompanying the generation of a single dot. Then, gamma correction is performed based on the original threshold matrix so that the target density characteristic is obtained, and screen processing is performed using a normalized threshold matrix obtained by normalizing the corrected original threshold matrix to the number of gradations of the image forming apparatus. Since the occurrence position of each dot by the original threshold matrix is designed to suppress the granularity, the suppressed granularity can be maintained even if it is gamma corrected, and such an original threshold matrix is provided. By doing so, calibration can be performed so that an image can be formed with a target density corresponding to the secular change of the printing unit of the image forming apparatus.

  In the above embodiment, the image processing unit generates a normalized threshold matrix after performing gamma correction on the original threshold matrix, and performs screen processing with reference to the normalized threshold matrix. However, the normalized threshold value matrix may be generated by normalizing without correcting the original threshold value matrix, and screen processing may be performed with reference to the normalized threshold value matrix. In this case, correction cannot be performed following the density characteristics of the print engine, but screen processing is performed based on an original threshold matrix that has both resolution and gradation and can suppress texture generation. be able to. In this case, the correction unit 15 has a function of normalizing the original threshold matrix to generate a normalized threshold matrix. The correction unit 15 can be realized by either a hardware unit or a software module.

It is a figure which shows the example of the basic matrix which is a basic unit of a threshold value matrix, and the expanded threshold value matrix. 1 is a configuration diagram of an image forming apparatus in the present embodiment. It is process drawing which shows the relationship between the original threshold value matrix which is the screen in this Embodiment, and the normalization threshold value matrix. It is process drawing which shows the relationship between the original threshold value matrix which is the screen in this Embodiment, and the normalization threshold value matrix. It is a figure which shows the relationship between a screen line number and a basic matrix. It is a figure which shows the example of an expansion matrix. It is a figure which shows the example of arrangement | positioning of 288 dots in the original threshold value matrix SCR2. It is a figure which shows the example of the normalized original threshold value matrix SCR2NORM. It is a figure which shows the measured output density characteristic and the target output density characteristic. It is a figure which shows the method of calculating | requiring a gamma correction table. It is a figure which shows an example of the normalization threshold value matrix SCR4 in this Embodiment.

Explanation of symbols

12: printer, image forming apparatus 14: image processing unit 15: correction unit 16: screen processing unit 20: printing unit, print engine 21: output density measuring device SCR2: original threshold matrix SCR4: normalized threshold matrix

Claims (8)

An image processing unit for generating image reproduction data indicating the presence or absence of dots from the input gradation data;
In an image forming apparatus having a printing unit for forming an image based on the image reproduction data,
The image processing unit includes:
An original threshold value matrix in which threshold values are arranged so that the increasing dots are dispersedly arranged corresponding to the increase of each dot in the dot group on the displacement vector corresponding to the screen line number;
The original threshold value matrix is corrected based on density characteristics having a relationship between image reproduction data and output image density in the printing unit, and the corrected original threshold value matrix corresponds to the number of gradation levels of the input gradation data. A correction unit for generating a normalized threshold matrix normalized,
An image forming apparatus comprising: a screen processing unit that generates the image reproduction data from the input gradation data based on the normalization matrix. In claim 1,
The original threshold value matrix is generated by arranging each dot of the dot group at any position on the displacement vector so that the granularity after generation of the dot is minimized. An image forming apparatus. In claim 2,
The outer product of displacement vectors corresponding to the number of screen lines is L (L is plural), the basic matrix is composed of L × M (M is a positive integer) number of pixels, and the original threshold matrix is N (N is plural). ) Having the number of pixels in which the basic matrix is arranged,
In the original threshold value matrix, an M × N dot group is generated on the displacement vector every time one gradation of L gradations is increased. In claim 2 or 3,
The image forming apparatus according to claim 1, wherein the normalized threshold value matrix is such that the number of dots generated each time the number of gradations of the input gradation data increases by one gradation is not constant. In claim 1,
The correction unit corrects the original threshold matrix so that an output density of an image formed by the printing unit has a desired density characteristic corresponding to a gradation value of the input gradation data, thereby correcting the corrected original threshold value. An image forming apparatus that generates a matrix. In claim 1,
The correction unit generates patch pattern data having a plurality of gradation values based on a normalized original threshold matrix obtained by normalizing the original threshold matrix corresponding to the number of gradations of the input gradation data, and An image forming apparatus, wherein the printing unit generates a patch pattern based on patch pattern data, and the density characteristic is detected based on the density of the generated patch pattern. An image processing unit for generating image reproduction data indicating the presence or absence of dots from the input gradation data;
In an image forming apparatus having a printing unit for forming an image based on the image reproduction data,
The image processing unit includes:
An original threshold value matrix in which threshold values are arranged so that the increasing dots are dispersedly arranged corresponding to the increase of each dot in the dot group on the displacement vector corresponding to the screen line number;
A normalization unit that generates a normalized threshold matrix obtained by normalizing the original threshold matrix corresponding to the number of gradations of the input gradation data;
An image forming apparatus comprising: a screen processing unit that generates the image reproduction data from the input gradation data based on the normalization matrix. In claim 7,
The original threshold value matrix is generated by arranging each dot of the dot group at any position on the displacement vector so that the granularity after generation of the dot is minimized. An image forming apparatus.

Images