JP2003202842A - Method, device and program for image processing and computer-readable recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the picture quality of an image which is chromatically displayed on a liquid crystal display panel by dither matrix shifting. <P>SOLUTION: After the matrix shifting for making matrixes having shifted in a threshold correspond to planes representing images of the primary colors R, G, and B is performed when an image is displayed on the LCD panel 105 having an array of pixels each constituted by putting dots of R, G, and B together in one, pseudo-halftone processing by a multi-valued systematic dither method is carried out and an image based upon image data after the processing is displayed. In this case, an optimum shift quantity determining means 102 finds the simultaneous ON probability of all dots R, G, and B which turn on at the same time as to one pixel and determines a shift quantity for matrix shifting according to statistical values obtained by statistically processing the simultaneous ON probability of all the dots R, G, and B. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a liquid crystal display device, an EL display device and a PD mounted on electronic equipment such as mobile phones, personal digital assistants and personal computers.
Among various display devices such as P, an image processing method applied when displaying an image on a display device configured by arranging dots of a plurality of colors as one pixel and arranging the pixels, The present invention relates to an image processing program and a computer-readable recording medium.

[0002]

2. Description of the Related Art Conventionally, an image display unit of a mobile phone or the like has a monochrome or color liquid crystal display (LCD: Liquid).
Crystal Display) panel is used. This LC
In the D panel, the transmittance of the liquid crystal is changed by turning on / off the driving voltage to the liquid crystal cells arranged in a matrix, and characters or images of two gradations are displayed. Recently, as mobile phones have become more multifunctional such as the Internet function, it has been required to display more information including images with higher gradation and higher image quality on the LCD panel.

When an image is displayed in color on the LCD panel, if the number of gradations that can be displayed on the LCD panel is small, contours called contours on the gradation portion are generated in the gradation portion, and the image quality deteriorates. It is known that this problem is solved and the image quality is improved by performing pseudo-halftone display by a binary or multivalued systematic dither method using a distributed matrix. It is known that the image quality is further improved by (shifting). Here, the matrix is a matrix in which a predetermined number of thresholds are arranged in the row / column direction, and is associated with any one of a plurality of colors forming one pixel. Also,
The matrix shifting refers to a technique of applying one kind of matrix in a state where the fence country is changed, when applying the matrix for each of a plurality of colors forming one pixel.

As a publicly known example to which this matrix shift is applied, there is a description content of Japanese Patent No. 2622429.
This content is R, at a low resolution of 300 dpi.
By performing a process of shifting the matrix of at least one color of G and B by one pixel, a visually preferable result is obtained.

[0005]

By the way, in the conventional image display method for shifting the matrix, only the technical content of performing the process of shifting the matrix of at least one color of R, G, B by one pixel at low resolution is described. It has not been. The effect is that a visually desirable result can be obtained. In other words, the technique does not specifically show what kind of policy should be used for the matrix shift, and how the matrix shift should improve the image quality.

As another known technique, R.I. J. Kle
nsch, Dietrech Meyerhofer,
J. J. Walsh, RCA Corporation
n, Electronically Generate
d Halftone Pictures, TECHN
ICAL ASSOCIATION OF THE G
RAPHIC ARTS PROCESSEDINGS, p
p. 302-pp. The 320 and 1970 papers describe a matrix having different patterns for each color based on the same matrix as a method for minimizing moire and color shift without giving a screen angle.

It is considered that the contents of the publicly known technology correspond to the matrix shifting, but in the technical contents, what kind of policy should be used to perform the matrix shifting, and what kind of action the matrix shifting causes to improve the image quality. This is not a technology that has been specifically shown.

The present invention has been made in view of the above point, and a plurality of dots of a plurality of colors are collectively set as one pixel, and the image quality of an image displayed in color on a display device configured by arranging the pixels is dithered. An object of the present invention is to provide an image processing method, an image processing device, an image processing program, and a computer-readable recording medium that can be improved by shifting.

[0009]

In order to solve the above-mentioned problems, the image processing method of the present invention is to disperse dots of a plurality of colors into one pixel and disperse the dots in a display device in which the pixels are arranged. When displaying an image based on image data that has undergone pseudo-halftone processing in which binarization or multi-value processing is performed using a matrix of type, the matrix of at least one of the plurality of colors is shifted during the pseudo-halftone processing. In the image processing method for performing the processing, a simultaneous ON probability that dots of two or more colors among the plurality of colors are simultaneously turned on adjacent to each other is obtained, and a shift amount at the time of performing matrix shift is calculated, and a simultaneous ON probability is statistically calculated. It is characterized in that it is determined according to the statistical value obtained by processing.

According to this method, the simultaneous ON probability (the lower the probability value, the more difficult it is for a plurality of ON dots to be adjacent to each other at the same time, therefore ON (ON) and OFF (OFF)
, Which indicates that the spatial frequency at which the dots alternate with each other is increased, and the image quality is improved}, and the statistical value of the simultaneous ON probability is further calculated, and the shift amount when the matrix shift is performed is determined from this statistical value. The shift amount corresponding to the statistical value that provides the best image quality can be obtained, and the image quality can be improved by performing the matrix shift with this optimum shift amount.

In the image processing method of the present invention,
The simultaneous on-probability is obtained when the input gradation and the on-dot pattern correspond to each other in each matrix of two or more colors among a plurality of colors forming one pixel for which a matrix shift is performed. , The dot-on probability that a dot is turned on is obtained by dividing each number of times that the dot corresponding to each threshold value of the matrix is turned on for all input tone inputs by the total input tone, It is characterized in that it is obtained by multiplying the dot-on probabilities of the respective matrices of two or more colors among the obtained plurality of colors.

According to this method, the dot-on probability is calculated for each of the matrixes of two or more colors among the plurality of colors which form one pixel and which is subjected to the matrix shift, and these dot-on probabilities are calculated. Since the simultaneous ON probability of dots of two or more colors among a plurality of colors is calculated from, the simultaneous ON probability can be calculated more appropriately.

Further, in the image processing method of the present invention,
It is characterized in that two or more colors of a plurality of colors which form one pixel and are used when obtaining the simultaneous ON probability are determined in descending order of the size that contributes to the luminance.

According to this method, the probability that dots of two or more colors among a plurality of colors forming one pixel are turned on at the same time in the order of the size that contributes to the brightness is calculated. Since the simultaneous ON probability can be obtained, a statistical value indicating that the image quality is higher can be obtained, and an optimal shift amount can be obtained.

In the image processing method of the present invention,
The statistic value is characterized in that it is obtained by any of the average value, standard deviation, maximum value, minimum value of each simultaneous ON probability in the corresponding matrix, alone, or in combination.

According to this method, since the statistic value is obtained by any one of the average value, standard deviation, maximum value and minimum value of the respective simultaneous ON probabilities in the matrix, alone or in combination, the image quality can be obtained as the statistic value. The criterion for judging whether the quality is good or not is properly associated, and a more optimal shift amount can be obtained by using this appropriate statistical value.

In the image processing method of the present invention,
The statistic value is characterized in that the matrix for performing matrix shifting is obtained by calculating the simultaneous ON probability in each shift amount while shifting the matrix by the shift amount of all patterns in this matrix.

According to this method, the shift-target matrix is shifted by the shift amounts of all the patterns in this matrix, the simultaneous ON probabilities at the respective shift amounts are calculated, and the statistical values of the simultaneous ON probabilities are calculated. It is possible to obtain a statistic value indicating that the image quality is the best among the matrix shifts performed with the shift amount, and it is possible to obtain an optimum shift amount.

In the image processing method of the present invention,
The color to which the matrix shift is performed is characterized by being the color that contributes most to the luminance.

According to this method, since the shifted color contributes most to the brightness, the contrast of the image at the dot ON / OFF boundary is less noticeable, and the image quality becomes higher. Also, since the shift itself is relative, it is equivalent when shifting all colors except the color that contributes most to the luminance in the opposite direction.
This case is also included in the present invention in which the color that contributes most to the luminance is shifted.

In the image processing method of the present invention,
The color to be matrix-shifted and the shift amount are characterized in that the matrix for the color that contributes most to the luminance and the matrix for the color that contributes to the next luminance are relatively displaced.

According to this method, since the matrix of the color that most contributes to the luminance and the matrix of the color that contributes to the next luminance are relatively displaced, the contrast of the image at the dot ON / OFF boundary is less noticeable, and higher image quality is achieved. Becomes Further, the image processing apparatus of the present invention shows each image of a plurality of colors when displaying dots on a display device configured by arranging dots of a plurality of colors as one pixel. In an image processing apparatus for displaying an image based on image data that has been subjected to pseudo halftone processing that performs binary or multi-valued processing after shifting a matrix that associates matrices in which the threshold values are shifted for each plane , A statistical value obtained by statistically processing the simultaneous ON probability when determining the simultaneous ON probability that dots of two or more colors among the plurality of colors are simultaneously turned on adjacently and performing the matrix shift. It is characterized in that a pseudo halftone processing means for carrying out the above-mentioned pseudo halftone processing is provided by using the shift amount determined according to the above, or the shift amount which is consequently obtained.

According to this device, the simultaneous ON probability is calculated,
Furthermore, since the statistical value of the simultaneous ON probability is obtained and the shift amount when performing the matrix shift is determined from this statistical value,
The image quality can be improved by obtaining the shift amount corresponding to the statistical value that gives the best image quality and performing the matrix shift with this optimum shift amount.

In the image processing device of the present invention,
The simultaneous on-probability is obtained when the input gradation and the on-dot pattern correspond to each other in the matrix of two or more colors among the plurality of colors forming one pixel, which is subjected to the matrix shift. , The dot-on probability that a dot is turned on is obtained by dividing each number of times that the dot corresponding to each threshold value of the matrix is turned on for all input tone inputs by the total input tone, It is characterized in that it is obtained by multiplying the dot-on probabilities of the respective matrices of two or more colors among the obtained plurality of colors.

According to this apparatus, the dot-on probability is calculated for each of the matrices of two or more colors out of the above-mentioned plurality of colors that are subjected to the matrix shift, and the dot-on probability of the above-mentioned plurality of colors is obtained. Since the simultaneous ON probabilities of two or more colors are calculated, the simultaneous ON probability can be calculated more appropriately.

In the image processing apparatus of the present invention,
The statistic value is characterized in that it is obtained by any of the average value, standard deviation, maximum value, minimum value of each simultaneous ON probability in the corresponding matrix, alone, or in combination.

According to this apparatus, the statistical value is calculated as an average value, a standard deviation, of the simultaneous ON probabilities of each matrix.
Since the maximum value and the minimum value are all, alone or in combination, the criterion for determining the quality of the image quality is properly associated with the statistic value. The amount of smoothing can be calculated.

In the image processing apparatus of the present invention,
The statistic value is characterized in that the matrix for performing matrix shifting is obtained by calculating the simultaneous ON probability in each shift amount while shifting the matrix by the shift amount of all patterns in this matrix.

According to this apparatus, the shift target matrix is shifted by all shift amounts considered in this matrix, the simultaneous ON probabilities at the respective shift amounts are obtained, and the statistical values of the simultaneous ON probabilities are obtained. It is possible to obtain a statistic value indicating that the image quality is the best among the matrix shifts performed with the shift amount, and it is possible to obtain an optimum shift amount.

Further, in the image processing apparatus of the present invention,
The color to which the matrix shift is performed is characterized by being the color that contributes most to the luminance.

According to this apparatus, since the shifted color contributes most to the luminance, the contrast of the image at the dot ON / OFF boundary is less noticeable, and the image quality becomes higher.

In the image processing device of the present invention,
The color to be matrix-shifted and the shift amount are characterized in that the matrix for the color that contributes most to the luminance and the matrix for the color that contributes to the next luminance are relatively displaced.

According to this apparatus, since the matrix of the color that contributes the most to the luminance and the matrix of the color that contributes to the next luminance are relatively displaced, the contrast of the image at the dot ON / OFF boundary is less noticeable, and the image quality is higher. Becomes

Further, the image processing program of the present invention uses a dispersion type matrix for each color of a plurality of colors in a display device configured by arranging dots of a plurality of colors as one pixel and arranging the pixels. When displaying an image based on image data that has been subjected to pseudo-halftone processing for performing binary or multivalued processing, a matrix of at least one of the above-mentioned plurality of colors In an image processing program for performing shift processing on a matrix, a simultaneous ON probability that dots of two or more colors among the plurality of colors are simultaneously turned on adjacently is calculated, and the simultaneous ON probability is calculated when performing the matrix shift. To the computer to perform the above pseudo halftone processing using the shift amount determined according to the statistical value obtained by statistically processing It is characterized in that to the row.

According to this program, the simultaneous ON probability is calculated for each of the matrixes of two or more colors among the above-mentioned plural colors that are subjected to matrix shifting, and the plural colors are calculated from these dot-on probabilities. , The statistical value of the simultaneous ON probability of two or more colors is obtained, and the shift amount when the matrix shift is performed is determined from this statistical value. Therefore, the shift amount corresponding to the statistical value that gives the best image quality is obtained, and this optimum The image quality can be improved by performing the matrix shift with the shift amount.

A computer-readable recording medium of the present invention is characterized by recording the above-mentioned image processing program.

According to this recording medium, the simultaneous ON probability is calculated, the statistical value of the simultaneous ON probability is further calculated, and the shift amount when the matrix shift is performed is determined from this statistical value. The image quality can be improved by obtaining the shift amount corresponding to and performing the matrix shift with this optimum shift amount.

[0038]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Before describing the embodiments, a binary and multi-valued systematic dither method for color images will be described. The dither method is a method in which a device having a small number of display gradations has a pseudo display of an intermediate gradation between the display gradations. Called multi-valued dither. Multi-valued dither is composed of a combination of binary dithers.

First, the binary dither will be described.
First, the dither matrix showing the order of threshold values shown in FIG. 9E is determined, and the threshold matrix shown in FIG. 9B is calculated from the dither matrix. Next, the input image shown in (a) is compared with the threshold matrix shown in (b), and if the value of a pixel in the input image is greater than or equal to the corresponding threshold value, 255 is output, and if smaller, 0 is output. Threshold processing for output is performed, and a pseudo halftone image (output image) consisting of binary values of 0 and 255 shown in (d) is output. Now, even if the input image (a) is an image with 128 uniform densities, the average value of the pseudo halftone image shown in (d) which is the output image is 128, and the pseudo image is simulated by the binary pattern of 0, 255. Therefore, the halftone 128 is expressed.

The multi-value dither is executed by dividing the range of the input gradation according to the multi-value to be used and performing the binary dither within each range. When the multi-value is four-valued and the corresponding output gradation is 0,85,170,255, the input gradation is divided into three ranges of 0-85,86-170,171-255.

First, the threshold matrix shown in (b) to (d) is generated from the dither matrix shown in FIG. 10A depending on which gradation range the input image belongs to. In the case of the input image shown in FIG. 9A, the gradation 128 is included in the range, so that the threshold matrix shown in FIG. 9C is generated. Next, using the matrix of (c), binary dither with gradations 85 and 170 is executed, and a pseudo halftone image as shown in (e) is output. Here, the threshold matrices shown in (b) to (d) do not have to be generated each time, and may be generated in advance. Further, other than this, a procedure for realizing multi-valued dither is known, but the present invention relates to a dither matrix and does not depend on a specific method of applying a dither threshold value.

The above is the case of performing dithering on an image in which one pixel is one color (gray). When dithering a color image, one pixel of the color image is R,
When it is composed of three colors of G and B, one pixel is divided into R, G, and B color components, and each image is regarded as an image of one color, and dithering is performed. At this time, the same matrix may be used for each of R, G, and B colors, or different matrices may be used.

Further, when associating different dither matrices with each color, as shown in FIG.
The G and B color matrices are individually stored in the memories 1101 to 11
03, and dither processing may be performed by the dither processing means 1104 to 1106. However, when the separate dither matrices are composed of a single matrix and a matrix shifted by a certain offset, one matrix is stored in the memory 1110 as shown in (b). Different dither matrices may be associated with the respective colors by the addressing means 1111 to 1113 which are independent for R, G, and B at the time of reading.

The addressing means 1111 to 1113 for shifting the matrix can be expressed by the following equations (0-1) and (0-2). Here, the matrix size is x_size in the x direction and y_size in the y direction.
The shift amount of the matrix is (x_size, y_size
In the case of e), it is sufficient to make the (m, n) threshold value of the matrix correspond to the (i, j) pixel of the input image. Also, addressing means 1111 to 11
If 13 is set to be the same, the same matrix can be associated with each color of R, G, and B.

M = (ix−shift)% x_size (0−1) n = (j−y_shift)% y_size (0−2) In the above, one pixel of the color image is the three primary colors of R, G, and B. However, the color composition of the color image may be composed of a plurality of other colors such as C, M, Y, and K, and is not limited to a specific color. .

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(Embodiment) FIG. 1 is a block diagram showing a configuration of an image processing apparatus according to an embodiment of the present invention.

The image processing apparatus 100 includes an input means 10
1, a dither processing means 104 having an optimum shift amount determining means 102 and a matrix moving means 103, and an LCD panel 105 for color display.

The input means 101 takes in an image obtained by a scanner or an image pickup means as input image data.

The LCD panel 105 has a screen in which color filters of three primary colors of R, G, and B are arranged, and a full-color image is displayed on this screen by mixing the color filters. Here, as shown in FIG. 2, the LCD panel 105 includes pixels 201 each of which is a group of sub-pixels (also referred to as dots) in which each of R, G, and B is a rectangle, and the same color of the pixels 201 is displayed. Images are displayed on a screen that is arranged in a line (arranged in stripes).

The dither processing means 104 is the input means 101.
The input image data captured by is separated into planes showing images of R, G, and B, and a threshold matrix set based on the dither matrix is associated with each of the R, G, B planes. Performs pseudo-halftone processing by the multi-valued or multi-valued systematic dither method. Then, the image data after the pseudo halftone processing is collected and output to the LCD panel 105.

The matrix moving means 103, when the dither processing means 104 associates the threshold matrix for each of the R, G, B planes, the threshold is optimally shifted for at least one color plane of R, G, B ( The process of associating the threshold value matrix in the state of being shifted by the pixel (described later) is performed. That is, when performing pseudo halftone processing, R, G, B
The process of shifting the matrix of at least one color is performed, and when the matrix shifting is performed, the matrix is shifted by the optimum shift amount (described later) determined by the optimum shift amount determining means 102 which is a characteristic element of the present embodiment. .

Here, the matrix shift will be described in detail. First, the general matrix shift is shown in FIG.
As shown in, when the matrix is 4 × 4, the image 3
01 for R, G, B dither matrices Rm, G
When m and Bm are laid out in a grid pattern 302, the matrix of at least one color of R, G, and B is shifted from the grid of the grid pattern 302 by parallel movement. FIG. 3 shows a case where the matrix is not shifted and a case where only the G dither matrix Gm is shifted by (1, 1). Here, (1, 1) represents the amount of matrix shift by (x, y) of the coordinates, the state without the dither matrix shift is (0, 0), and the dither matrix is shifted by one pixel in the right direction. In this case, x is a positive 1 and y is a positive 1 when the pixel is shifted downward by 1 so that the shifted number is represented by adding a positive or negative sign to the shifted direction. ing.

FIG. 4 shows a state in which only the G dither matrix is shifted by (0, 1) due to such matrix shifting. FIG. 4A is a dither matrix for R, G, and B without matrix shifting,
(B) shows a dot ON pattern in the case where data of uniform densities such that half of the pixels of R, G, and B are lighted are input. FIG. 4C shows G
Is a matrix of G when the matrix is shifted by (0, 1) only, and (d) is data of a density such that R, G, and B at that time illuminate half of all pixels. 7 shows a dot ON pattern when is input. Here, since the dither matrix in the case of no matrix shift is used for R and B, the dot positions to be turned ON are the same for R and B as in the case of no matrix shift, but for G, (0, 1 ) Only the matrix is shifted, so G
For, the dot at the position shifted by (0, 1) is O
N will be done. Note that the matrix is repeatedly made to correspond to the image in a grid pattern, so the matrix of G shown in (c) after matrix shifting is
The bottom row of the dither matrix shown in (a) is “15, 7, 1
"3, 5" is shifted to the top row.

ON without matrix shifting in FIG.
O when the matrix is shifted from the dot pattern (a)
Compared with the N dot pattern (b), the ON dot pattern (b) is ON when the matrix is shifted.
The spatial frequency in the x direction of the dot pattern becomes high. Generally, the relationship between the spatial frequency and the visual sensitivity is shown by the curve 501 in FIG.
As shown in, the higher the spatial frequency is, the lower the visual sensitivity is, and the less the contrast is felt. Therefore, the ON dot pattern of (b) having a higher spatial frequency is ON.
The cluster of dots and the cluster of OFF dots are not recognized as separate clusters but are felt more uniformly, and are perceived by the human eye as a halftone of the average density. Since the input was assumed to be uniform gray data such that half of the dots would light up, it would be said that the image quality would be better if it was felt as a halftone of the smoothed average density. It can be seen that the image quality is improved. Here, the colors for matrix shifting are R,
It is particularly effective when G, which has the highest brightness indicating the intensity of brightness, is shifted. Further, since the matrix shift is a relative problem between colors, the same effect can be obtained by shifting R and B by the same amount in opposite directions instead of shifting only G. Furthermore, since the influence of the relative positional relationship between the color most related to the luminance and the color next related to the luminance becomes the most important, it is necessary to set the relative positions of these two colors to be displaced. is important.

Looking at the relationship between the pattern of ON / OFF dots and the spatial frequency, under the same number of ON dots, as shown in FIG. It can be seen that the larger the number of pixels that are simultaneously turned on in one pixel, the lower the spatial frequency and the lower the image quality, and as shown in (d), the smaller the number, the higher the spatial frequency and the higher the image quality.

For this reason, in the present embodiment, the optimum shift amount determining means 102 has a probability that the R, G, and B three-color subpixels are simultaneously turned ON in one pixel (hereinafter, R, G, and B dots are turned ON simultaneously). Probability) is obtained, a statistical value is obtained from the R, G, and B dot simultaneous ON probabilities, and the optimal shift amount is determined by using this statistical value as an evaluation value.

Here, based on the dither matrix (without shifting) shown in FIG. 4A, as shown in FIG. 6, the input gradation (17 gradations) and the output pattern (ON dot pattern).
When and correspond, the number of times that any dot of the pixels corresponding to each threshold value is turned on for the input of 0 to 16 gradations =
If N is set, N / 17 (17 is 17 of 17 gradations) is a probability that a dot of any one of R, G, and B is turned ON for all input gradation inputs (hereinafter, referred to as dot ON probability). Will be shown. The dot ON probability is higher as the dither matrix threshold value is smaller, and is lower as the threshold value is larger.

This dot ON probability is calculated by the following equation (1) for each dither matrix of R, G, B, and the product of these dot ON probabilities of R, G, B is calculated based on equation (2). Simultaneously turning on the R, G, and B dots by calculating
Probability is calculated.

[0060]

[Equation 1]

[0061]

[Equation 2] However, Xmat: dither matrix for X (X = R, G, B), N: maximum threshold value, THmat (i, j): threshold value of dither matrix mat, Pmat_on (i, j):
dot on probability of mat, PR, G, B_on (i,
j): R, G, B dot simultaneous ON probability.

Further, a method of obtaining the R, G, B dot simultaneous ON probability will be specifically described with reference to FIG. This Figure 7
For a matrix of size 4 × 4, a matrix without a matrix shift is used for R and B, and a matrix in which a dither matrix is shifted downward by one pixel by (0,1) is used for G, R, The state of the matrix for obtaining the simultaneous ON probability of G and B dots is shown.

First, the non-shifting R dither matrix shown in FIG. 7A, the non-shifting B dither matrix shown in FIG. 7B, and the (0, 1) shifted G shown in FIG. 7C.
In each of the dither matrix of FIG.

Since this dot ON probability is N / 17 described above, the dot ON probability of the pixel corresponding to each threshold value of the R dither matrix shown in (a) is as shown in (d).
Each number N = 1 to 16 described in the matrix 1/17
It is the value multiplied by. For example, the dot ON probability of the pixel corresponding to the threshold 0 shown in (a) is 16/17, the dot ON probability of the pixel corresponding to the threshold 2 is 14/17, and the dot ON probability of the pixel corresponding to the threshold 15 is 1/17. Becomes

The dot ON probability of the B dither matrix shown in (b) is obtained in the same manner as the above R dither matrix and is the same. The dot ON probability of the G dither matrix shown in (c) is different from the above R and B in the arrangement of the thresholds, but similarly, each threshold is described in the matrix shown in (f) N = 1 to 1 times. It is a value obtained by multiplying 16 by 1/17.

The R, G, and B dot simultaneous ON probabilities are obtained by multiplying the R, G, and B dot ON probabilities thus obtained at the same pixel positions in the matrix. That is, (0,0) in (d), (e), and (f)
The value obtained by multiplying the number of times N = 16,16,1 at
Since the value at the position of (0,0) shown in (g) is 256, it is multiplied by (1/17) ³ to obtain 256/49.
The simultaneous ON probability of R, G, and B dots of 13 can be obtained.

Next, in order to calculate an evaluation value representing the likelihood of simultaneous ON of R, G, and B dots, the average value, maximum value, minimum value, standard deviation, etc. are calculated from the probability of simultaneous ON of R, G, and B dots. Calculate the statistical value of. The evaluation value is the following formula (3), the average value is the formula (4), the standard deviation is the formula (5), and the maximum value is the formula (6).
The minimum value is calculated by the equation (7).

[0068]

[Equation 3]

[0069]

[Equation 4]

[0070]

[Equation 5]

[0071]

[Equation 6]

[0072]

[Equation 7] However, Eval: evaluation value, αn: weighting coefficient (n =
1, ..., 4), M: size of matrix in x direction, N:
Size of matrix in y direction, Pave (Gmat, G
mat, Bmat): standard deviation of PR, G, B_on (i, j), Pmax (Gmat, Gmat, Bma)
t): maximum value of PR, G, B_on (i, j), Pmi
n (Gmat, Gmat, Bmat): PR, G, B_
The minimum value of on (i, j).

Although the evaluation value is a total value obtained by weighting a plurality of statistical values in the equation (3), one statistical value may be used alone or in combination. When only the average value is used alone as the evaluation value, the local high probability and the variation in the value are not considered, but if the maximum value and the standard deviation are added to form a linear sum of them, they are considered. The evaluation value, which is the total value, represents the likelihood of simultaneous ON of the R, G, and B dots, and the smaller the value, the better the image quality. In other words, the smaller the value, the more R, G,
It means that simultaneous ON of B dots is unlikely to occur and the image quality is improved.

Further, as shown in the following table (1), each statistic value has a smaller average value, maximum value and standard deviation. , The larger the minimum value is, the R, G, B dots are turned on simultaneously.
Is less likely to occur and the image quality is good. Since the minimum value tends to be opposite to the other statistical values as it is, it may be used as a value indicating that the smaller the value is, the less likely the simultaneous ON of R, G, and B dots occurs.

[0075]

[Table 1] Next, an image processing method by the image processing apparatus 100 having such a configuration will be described with reference to the flowchart shown in FIG. However, the matrix of size 4 × 4 shown in FIG. 7 described above is applied to this image processing.

(Step S801) First, when the dither processing means 104 separates the input image data obtained by the input means 101 into planes showing images for R, G, and B, the optimum shift amount determining means 102. Then, it is determined whether the dither matrix of R, G, or B is to be shifted. Here, it is assumed that it is decided to shift the G dither matrix.

(Step S802) The optimum shift amount determining means 102 determines the shift amount of G with respect to the dither matrix. At this point, it is assumed that the G dither matrix is shifted downward by one pixel (0, 1).

(Step S803) G is determined by the determination.
The dither matrix of is the shifted dither matrix shown in FIG. Further, in step S804, the R and B dither matrices become the non-shifted dither matrices shown in (a) and (b).

(Step S805) The dot ON probability matrix is calculated based on the G dither matrix for which the shift amount has been determined by the optimum shift amount determining means 102. That is, as described above, the number N of times when any of the dots of the pixels corresponding to each threshold value is turned ON with respect to the input of all the input gradations is as shown in the relationship between (c) and (f) of FIG. ,
The dot-on probability matrix is calculated by applying the values corresponding to the respective threshold values of the G dither matrix and multiplying each N by 1/17.

(Step S806) A dot-ON probability matrix is calculated based on the R and B dither matrices without shift determined by the optimum shift amount determining means 102. That is, the number of times N is as shown in FIGS.
As shown in the relationships between (d) and (e), R and B are applied corresponding to the respective threshold values of the dither matrix, and N of each is multiplied by 1/17 to obtain R and B.
Each dot ON probability matrix is calculated.

(Step S807) The optimum shift amount determining means 102 multiplies the values of the dot ON probabilities of R, G, and B obtained above by those at the same pixel positions in these matrices. , The R, G, B dot simultaneous ON probability matrix shown in FIG. 7 (g) is calculated.

(Step S808) The optimum shift amount determining means 102 calculates an evaluation value from each value of the R, G, B dot simultaneous ON probability matrix. For these, the respective statistical values of the average value, the standard deviation, the maximum value, and the minimum value are calculated by the above equations (4) to (7). Further, the evaluation value is calculated by the equation (3) using these statistical values. Here, for simplicity of explanation, it is assumed that only the average value is used as the evaluation value.

(Step S809) The optimum shift amount determination means 102 determines whether or not the calculation of the evaluation value has been completed for the shift amounts of all patterns in the G dither matrix to be shifted.
As a result, if not completed, the process returns to step S802 and the processes up to step S808 are repeated. On the other hand, the evaluation values (=
If the calculation of the average value is completed, the process proceeds to step S810.

(Step S810) The optimum shift amount determining means 102 detects the best evaluation value from the calculated total evaluation values (= total average value) of Table 2.

[0085]

[Table 2] As can be seen from the results of Table 2, the average value = 9.15, which is the minimum average value, is detected as the best evaluation value.

(Step S811) The optimum shift amount determining means 102 determines the shift amount (0, 1) corresponding to the best evaluation value = 9.15 as the optimum shift amount.

However, the above steps S801 to S811
In the above process, it is assumed that only the G dither matrix is shifted, and the optimum shift amount is determined. However, any one of R and B, any two colors of R, G, and B, or R. , G,
Even when all B colors are shifted, the optimum shift amount can be determined by the same process.

Next, the dither processing means 104 performs R, G, B
When the dither matrix is associated with each of the planes, the matrix moving unit 103 associates the G plane with the dither matrix in which the threshold value is shifted by the optimum shift amount of pixels. That is, when the pseudo halftone process is performed, the process of shifting the G matrix by the optimal shift amount is performed. Then, in the dither processing means 104,
The image data of the R, G, and B color planes that have undergone the pseudo halftone processing after the shift processing are collected and output to the LCD panel 105, and an image of high quality based on the image data is displayed on the LCD panel 105. It

As described above, according to the image processing apparatus of the present embodiment, when the dots of the three primary colors of R, G and B are collectively set as one pixel and the image is displayed on the LCD panel 105 in which the pixels are arranged, the image is displayed. Then, after shifting the matrixes in which the matrices in which the threshold values are shifted for the planes showing the R, G, and B images are associated with each other, pseudo halftone processing by the binary or multivalued systematic dither method is performed. An image based on the applied image data is displayed. In this case, the optimum shift amount determination means 102 obtains the R, G, B all-dots simultaneous ON probability that the R, G, B dots are simultaneously turned on in one pixel,
The shift amount at the time of performing the matrix shift is determined according to the statistical value obtained by statistically processing the R, G, B all-dots simultaneous ON probability.

According to this structure, R, G, B all-dots simultaneous ON probability {The lower the probability value, the more difficult it is for a plurality of ON dots to be adjacent to each other at the same time.
N) and OFF (OFF) alternate spatial frequencies are high, which indicates that the image quality is improved}, and a statistical value of the simultaneous ON probability is calculated, and a matrix shift is performed from this statistical value. The shift amount is determined. That is, since the shift amount corresponding to the statistical value that gives the best image quality is obtained, the image quality can be improved by performing the matrix shift with this optimum shift amount. Therefore, LC
The image quality of the image displayed in color on the D panel 105 can be improved by shifting the dither matrix.

Next, an image processing program according to the present embodiment and a computer-readable recording medium recording the image processing program will be described. The recording medium is capable of transmitting the data content of the program to the reading function of the computer, and corresponds to, for example, various magnetic disks, optical disks, ROMs, RAMs, etc. that are built in, externally attached or detached to the computer.

An image processing program is recorded in the program recording area of such a recording medium. This program causes a computer to collect dots of the three primary colors of red, green, and blue into one pixel, and a liquid crystal display panel in which the pixels are arranged is subjected to pseudo-halftone processing by a binary or multivalued systematic dither method. When displaying an image based on the applied image data, processing for shifting the matrix of at least one color of red, green, and blue is executed at the time of pseudo halftone processing, and the above-mentioned red, green, and blue dots are 1 Simultaneous on-probability that pixels are turned on at the same time is obtained, and pseudo halftone processing is performed using the shift amount when performing the above matrix shift determined according to the statistical value obtained by statistically processing the simultaneous on-probability. Let it run.

Further, when the simultaneous ON probability is obtained in the above determination of the shift amount, the input tone and the ON dot pattern are present in each of the red, green and blue matrices to which the matrix shift is applied. When and correspond, the number of times that the dots corresponding to each threshold value of the matrix are turned on for the input of all input gradations is divided by all the input gradations to turn on the dots. Alternatively, the dot-on probability may be obtained, and the dot-on probabilities of the respective red, green, and blue matrices may be multiplied to obtain the dot-on probability.

Further, the statistic value obtained in the above shift amount determination is obtained by any of the average value, the standard deviation, the maximum value and the minimum value of the respective simultaneous ON probabilities in the corresponding matrix, alone or in combination. It may be one. Further, the statistic value may be obtained from each of the simultaneous ON probabilities by obtaining the simultaneous ON probability in each displacement amount while shifting the matrix for performing the matrix displacement by the displacement amounts of all patterns in this matrix. . Further, it is desirable that the color for which the matrix shift is performed is the color that contributes most to the luminance. Furthermore, it is desirable that the color for which the matrix shift is performed and the shift amount be such that the matrix for the color that contributes most to the luminance and the matrix for the color that contributes to the next luminance are relatively displaced.

Although the embodiments of the present invention have been described above, the present invention is not limited to such embodiments, and may be implemented in various modes without departing from the scope of the invention. Is possible. For example, the following modifications are possible.

Modification 1: The LCD panel is applied as the image display means in the above-mentioned embodiment, but the invention is not limited to this as long as the image display means has one pixel composed of sub-pixels of a plurality of colors. . For example, EL panel or P
It can also be applied to a display panel such as a DP.

Modification 2 In the above embodiment, the image display means is applied to the LCD panel in which the color filters of the same color are arranged in a row in the vertical direction in a stripe type arrangement, but one pixel is a subpixel of a plurality of colors. However, the present invention is not limited to this as long as the color display is performed by the color mixture. For example, it can be applied to a panel that does not use a color filter, or can be applied to a panel in which the arrangement of sub-pixels of a plurality of colors forming one pixel is a zigzag arrangement.

Modification 3: In the above embodiment, the case where a plurality of colors forming one pixel are R, G and B is applied, but the present invention is not limited to this. For example, multiple colors are C,
It can also be applied to cases such as M, Y, and K.

Modified Example 4 In the above-described embodiment, the case where the sub-pixel has a rectangular shape is applied, but the present invention is not limited to this. Need not be an exact rectangle,
It can also be applied to a panel in which the vertical and horizontal directions are not uniform, for example, a panel having elliptical subpixels. In the present invention, these shapes are referred to as a rectangular shape.

Modification 5: In the above embodiment, the statistical value is calculated for the dot simultaneous ON probability in order to obtain the optimum shift amount, but the present invention is not limited to this. Dot on
Since the probability and the dot OFF probability have a complementary relationship of 1 and the statistical values obtained from the dot simultaneous ON probability and the dot simultaneous OFF probability are always the same, the dot simultaneous OFF probability is used instead of the dot simultaneous ON probability. May be.

Modified Example 6 In the above embodiment, the probability that all the colors of a plurality of colors forming one pixel are simultaneously turned on in one pixel is used, but the present invention is not limited to this. For example, one pixel is formed. It suffices that the dots of two or more colors among a plurality of colors are adjacently turned on at the same time.

Modification 7: In the above embodiment, the average value of the dot simultaneous ON probabilities is used as the evaluation value, but it is not limited to this. For example, a weighting coefficient is applied to each of the statistical values such as the average value, standard deviation, variance, maximum value, and minimum value obtained from the dot simultaneous ON probabilities, and they are obtained as the sum of any of them, alone, or in combination. Any evaluation value can be used.

Modification 8: In the above embodiment, one of a plurality of colors forming one pixel in the optimum shift amount determining means is used.
Although the case where only the color matrix is shifted is applied, the present invention is not limited to this. For example, a matrix of two or more colors among a plurality of colors forming one pixel may be shifted.

In addition, in the image processing apparatus and the image processing program described in the embodiments of the present invention, a matrix in a shifted state or one pixel corresponding to at least one of a plurality of colors forming one pixel is formed. The matrix common to a plurality of colors and the shift amount for each color are illustrated in the case of being stored in the memory or recorded in the recording medium, but the matrix in the shifted state, the common matrix, and the shift amount for each color are fixed. Need not be stored in the memory, but may be acquired through a communication system or the like by a user operation. Alternatively, the matrix may be attached to the image data by incorporating it in the header portion of the image file, and the matrix may be acquired each time the image is browsed, without the user's conscious operation. Further, the information incorporated in the communication system or the header may be the matrix itself in the state of being shifted for each color, or the information regarding the shift amount for each color.

Finally, the present invention records the configuration as the invention of the image processing method, the configuration as the image processing apparatus, the program for realizing these, the data provided for the program, and the program as described above. The present invention can be realized in various forms such as a recording medium, a program, or a data signal including data provided for the program and embodied in a carrier wave.

[0106]

As described above, according to the present invention,
The dots of the three primary colors of red, green, and blue are grouped together into one pixel, and an image based on image data that has been subjected to pseudo halftone processing by the multivalued systematic dither method is displayed on the liquid crystal display panel in which the pixels are arranged. When performing a process of shifting the matrix of at least one color of red, green, and blue during the pseudo halftone process, the simultaneous on-probability that the red, green, and blue dots are turned on simultaneously in one pixel is obtained. The shift amount when performing the matrix shift is determined according to the statistical value obtained by statistically processing the simultaneous ON probability.

That is, the simultaneous ON probability (indicating that the lower the probability value, the higher the spatial frequency at which the ON / OFF dots alternate) is to improve the image quality}, and the statistical value of the simultaneous ON probability is obtained. Is calculated, and the shift amount when the matrix shift is performed is determined from this statistical value. In other words, since the shift amount corresponding to the statistic value that gives the best image quality is obtained, the image quality can be improved by performing the matrix shift with this optimum shift amount. Therefore, the image quality of the image displayed in color on the liquid crystal display panel can be improved by shifting the dither matrix.

[Brief description of drawings]

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

FIG. 2 is a diagram showing pixels of an LCD panel in the image processing apparatus.

FIG. 3 is a diagram for explaining matrix shifting in a 4 × 4 Bayer matrix.

FIG. 4 is a diagram showing a state in which only a G dither matrix is displaced due to matrix shifting.

FIG. 5 is a characteristic diagram of spatial frequency and visual sensitivity.

FIG. 6 is a dot ON for input of all input gradations when an input gradation corresponds to an output pattern (ON dot pattern) in a 4 × 4 Bayer matrix without shifting.
It is a figure which shows a state.

FIG. 7 is an explanatory diagram of a process of obtaining an R, G, B dot simultaneous ON probability matrix when the G dither matrix of R, G, B is shifted.

FIG. 8 is a flow chart for explaining a process for determining an optimum shift amount by the optimum shift amount determination means.

FIG. 9 is a matrix diagram for explaining binary dither.

FIG. 10 is a matrix diagram for explaining multi-valued dither.

FIG. 11 is a block diagram of an apparatus for associating different dither matrices for R, G, and B colors.

[Explanation of symbols]

100 image processing device 101 input means 102 optimal shift amount determining means 103 matrix moving means 104 dither processing means 105 LCD panel 201 1 pixel 501 by a group of sub-pixels of three primary colors of red, green and blue Spatial frequency-visual sensitivity characteristic curve Rm of red Dither matrix Gm Green dither matrix Bm Blue dither matrix

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04N 1/405 G09G 5/00 520J 5C081 1/52 H04N 1/40 C 1/60 DF term (reference) 5B057 AA20 CA01 CA08 CA12 CA16 CB01 CB07 CB12 CB16 CE13 CE16 CH01 DA16 5C006 AA12 AA21 BB11 FA56 5C077 LL19 MP08 PP32 PP43 PP46 PQ12 PQ22 RR02 RR05 5C079 H01 01B29 CC07 H01 01 KK47 5C082 BA12 BA34 BA35 BD02 CA12 MM10

Claims (15)

[Claims] 1. A binarization or multi-value conversion process using a distributed matrix for each color of the plurality of colors in a display device in which dots of a plurality of colors are collectively set as one pixel and the pixels are arranged. When displaying an image based on image data that has been subjected to pseudo-halftone processing, performing a process of shifting the matrix of at least one of the plurality of colors with respect to the matrix of any other color during the pseudo-halftone processing. In the image processing method, the simultaneous on-probability that dots of two or more colors among the plurality of colors are simultaneously turned on adjacently is obtained, and the shift amount at the time of performing the matrix shift is statistically calculated by the simultaneous on-probability. An image processing method characterized in that it is determined according to a statistical value obtained by performing the processing. 2. The simultaneous on-probability corresponds to an input gradation and an on-dot pattern in each matrix of two or more colors out of the plurality of colors to which the matrix shift is applied. In doing so, by dividing the number of times each dot corresponding to each threshold value of the matrix is turned on with respect to the input of all input gradations by the total input gradations,
The dot-on probability that the dot is turned on is obtained, and the dot-on probability of each matrix of two or more colors of the obtained plurality of colors is multiplied to obtain the dot-on probability. Image processing method. 3. The method according to claim 1, wherein two or more colors of the plurality of colors used when calculating the simultaneous ON probability are determined in order of magnitude contributing to luminance. Image processing method. 4. The statistical value is an average value, a standard deviation, a maximum value of the simultaneous ON probabilities of each of the corresponding matrices,
The image processing method according to claim 1, wherein all of the minimum values are obtained individually or in combination. 5. The statistical value is obtained from each of the simultaneous ON probabilities by obtaining the simultaneous ON probability at each displacement amount while shifting the matrix for performing the matrix displacement by the displacement amounts of all patterns in this matrix. The image processing method according to claim 1 or 4, characterized in that. 6. The color for which the matrix shifting is performed is
The image processing method according to claim 1, wherein the color contributes most to brightness. 7. The matrix shift according to claim 1, wherein the shift amount is set so that at least a matrix for a color that contributes to the most luminance and a matrix for a color that contributes to the next luminance are shifted. The image processing method according to claim 1. 8. When displaying dots on a display device configured by arranging dots of a plurality of colors as one pixel, a threshold value is set for each plane showing each image of the plurality of colors. In the image processing apparatus for displaying the image based on the image data on which the pseudo halftone processing for performing the binary or multivalued processing is performed after shifting the matrixes in which the matrices in the shifted state are associated with each other, Among the two or more dots, the simultaneous ON probability that adjacent dots are simultaneously ON is obtained, and when the matrix shift is performed, it is determined according to the statistical value obtained by statistically processing the simultaneous ON probability. An image processing apparatus comprising: a pseudo halftone processing unit that performs the pseudo halftone processing using a shift amount or a shift amount that results. 9. The simultaneous on-probability corresponds to an input gradation and an on-dot pattern in each matrix of two or more colors out of the plurality of colors to which the matrix shift is applied. In doing so, by dividing the number of times each dot corresponding to each threshold value of the matrix is turned on with respect to the input of all input gradations by the total input gradations,
9. The dot-on probability that the dot is turned on is obtained, and the dot-on probability of each matrix of two or more colors among the obtained plurality of colors is obtained and the dot-on probability is obtained. Image processing device. 10. The statistical value is obtained by any one of an average value, a standard deviation, a maximum value and a minimum value of each simultaneous ON probability in the corresponding matrix, alone or in combination. 8. The image processing device according to item 8. 11. The statistical value is obtained from each of the simultaneous ON probabilities by obtaining the simultaneous ON probability in each of the displacement amounts while shifting the matrix for performing the matrix displacement by the displacement amounts of all patterns in the matrix. The image processing apparatus according to claim 8 or 10, characterized in that. 12. The image processing apparatus according to claim 8, wherein the color for which the matrix shift is performed is the color that contributes most to the brightness. 13. The shift amount of the matrix is set such that at least a matrix for a color that contributes most to luminance and a matrix for a color that contributes next to luminance shifts. The image processing device according to any one of claims. 14. A binarization or multi-value conversion process using a distributed matrix for each color of the plurality of colors in a display device in which dots of a plurality of colors are collectively set as one pixel and the pixels are arranged. When displaying an image based on image data that has been subjected to pseudo-halftone processing, a process of shifting the matrix of at least one of the plurality of colors with respect to the matrix of any other color is performed during the pseudo-halftone processing. In the image processing program, the simultaneous on-probability that dots of two or more colors among the plurality of colors are simultaneously turned on adjacently is obtained, and the simultaneous on-probability is statistically processed when the matrix shift is performed. Image processing, characterized in that the computer is caused to execute a process for performing the pseudo-halftone processing using the shift amount determined according to the statistical value or the shift amount that is determined as a result. Program. 15. A computer-readable recording medium on which the image processing program according to claim 14 is recorded.