JP2003006630A - Device and method for displaying color image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a color image display device, with which color slippage to be sensed in a change in the order of sub-pixel arrangement can be corrected while keeping high spatial resolution. SOLUTION: This device has a storage part 2 for storing a source image having resolution higher than that of a color image to be displayed on a display means 4 and a resolution converting part 3 for generating a reduced image V by converting the resolution of the source image to resolution on the display means 4 and in such a case, the converting part 3 is composed of a source image color sense predicting part 5 for predicting a color to be sensed by a human being from a three-dimensional(3D) color image expressed in a 3D coloring system, a display image color sense predicting part 6 for predicting a color to be sensed by a human being from an image to be obtained when the reduced image V is displayed, and a reduced image optimizing part 7 for comparing both the predicted results and correcting the reduced image V so that the difference of both the results can be minimum.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for displaying a color image on a color display such as a color liquid crystal display or a color plasma display, and in particular, the arrangement of subpixels is from RGB (red, green, blue) to GBR.
The present invention relates to a color image display method and apparatus capable of maintaining a high spatial resolution without changing the color appearance (color shift) even when rearranged in (green, blue, red).

[0002]

2. Description of the Related Art When a color image is displayed on a color display in which one pixel is composed of three types of sub-pixels representing the three primary colors (RGB), it is normally displayed without taking account of the positional deviation of the sub-pixels. To do. However, if the positional deviation of the sub-pixels is not taken into consideration in this way, for example, diagonal lines forming a part of the font will be displayed jagged.

Therefore, the arrangement of the sub-pixels in each pixel is not considered to be fixed, but three types of arbitrarily arranged sub-pixels are regarded as one pixel, so that the above diagonal lines are displayed without jaggedness. Techniques have already been proposed (for example, JP-A-11-305738 and JP-A-2).
000-155551). These known proposals say that no matter how the three types of sub-pixels (RGB) are arranged, the same color must be perceived by the viewer as long as the ratio of the three primary colors is the same. Standing on the premise.

[0004]

However, in reality, when the arrangement (arrangement order) of the three types of sub-pixels is different, the appearance of the color of the pixel is changed. The present inventor discovered such a phenomenon and completed the present invention. This phenomenon will be described. FIG. 10A shows a normal arrangement of subpixels, and FIG. 10B is a view showing the arrangement of subpixels whose arrangement order is changed.

In FIG. 10, P indicates one pixel (Pixel), and p indicates a sub-pixel (subpixel) which is a unit of R, G, and B, respectively. For example, when displaying the end portion of a font, the end portion is usually formed by cutting it in units of one pixel like the vertical line in FIG. However, when the end portion is displayed as in this case (a), as described above, the jagged diagonal lines appear, and a clear display cannot be performed. That is, the spatial resolution is reduced.

Therefore, in order to obtain a high spatial resolution, according to the above-mentioned conventional proposal, in order to prevent the above jaggedness, as shown by the vertical line in FIG. Display one pixel. Even with this, one RGB set can be displayed. However, in that case, the above-mentioned color shift occurs.
This will be explained in a little more detail.

11 (a), (b) and (c),
It is a figure which shows how the conspicuousness of a color differs because of a background and the arrangement order of subpixels. In the figure, (a) shows a case where the arrangement order of RGB is changed on a black background, (b) shows a case where the arrangement order of CMY is changed on a white background, and (c) shows an RGB arrangement on a white background. This is the case when the arrangement order is changed. The above CMY are cyan, magenta, and yellow, respectively.

With reference to FIG. 1A, RG with a black background
When B (left) and GBR (right) are displayed, each color of the left pixel is inconspicuous (a white dot is visible), but each color of the right pixel is conspicuous. In particular, G (green) and R (red) at both ends of the pixel stand out. In other words, you can't see a single white, but you can see green and red separately. This is the color shift described above. The present invention mainly describes correcting the color misregistration in the mode as shown in FIG. 7A, but there are other similar modes of color misregistration. These are shown in (b) and (c) of this figure. The present invention can also be applied to such aspects (b) and (c).

Referring to FIG. 1B, the basic colors are subtractive primary colors (CMY) on a white background. The above (a) (same as (c)) uses the basic colors as additive three primary colors (RG
However, the above-mentioned problem of color misregistration occurs even under the subtractive three primary colors. That is, although the left side of this figure (b) looks like a single black dot, the magenta (M) and cyan (C) at both ends of the pixel are individually visible on the right side.

Further, referring to (c) of the figure, since the respective colors are slightly conspicuous on the left and the respective colors are not conspicuous on the right, a color shift occurs, but it is not as remarkable as (a) in the figure. Therefore, it is an object of the present invention to display a color image with a high spatial resolution while visually easing the color shift caused by the difference in the arrangement of subpixels.

[0011]

First, the basic idea of the present invention will be described. Based on the opposite color model for color perception in visual psychology, when a person sees a color image, he or she perceives the color by decomposing the color components into the three axes of the luminance axis, the red-green axis, and the blue-yellow axis. It is said that Furthermore, when the spatial frequency characteristics of each of the three axes are measured, it is known that the high-frequency side of the color image is well perceived in the order of luminance axis → red-green axis → blue-yellow axis.

Based on this opposite color model, in the RGB array shown in FIG. 10A, since the R subpixel and the G subpixel are adjacent to each other, the red component and the green component cancel each other, and It is hard to tell the difference between the two. on the other hand,
In the GBR array of FIG. 6B, since the R subpixel and the G subpixel are separated within one pixel, it is possible to sufficiently cancel the red component and the green component that are perceived up to a relatively high frequency side. It cannot be done, and it can be considered that the difference between the two is easily perceived by the human eye.

As described above, by using the color perception model capable of explaining even the difference in perception due to the difference in the spatial arrangement of the RGB sub-pixels, the color that a person would perceive when a certain color image is presented. Can be predicted (predicted color). If the display device used, such as a color display, expresses the halftone of each subpixel, the predicted color (the color that will actually be seen) and the desired color (the color that the person originally wants to see). By adjusting the display image so as to reduce the difference in color perception between the display image and the display image, it is possible to display a color image close to a desired color on the display device.

A basic device and method according to the basic idea of the present invention described above will be described with reference to the drawings. FIG. 1 is a diagram showing the basic configuration of a color image display device according to the present invention. In the figure, reference numeral 1 represents the entire color image display device. The device 1 includes display means, for example, a color liquid crystal display 4, for displaying a color image in which each image is composed of a plurality of sub-pixels representing a basic color.

Further, the storage unit 2 for storing an original image having a resolution higher than that of the color image to be displayed on the display unit 4, and the resolution of the original image stored in the storage unit 2 are displayed on the display unit 4. And a resolution conversion unit 3 that generates the reduced image V. When the display unit 4 is mounted on a small device such as a mobile terminal, conversion of the original image to the reduced image V is indispensable.

The characteristic components of the present invention are the original image color perception prediction unit 5, the display image color perception prediction unit 6, and the reduced image optimization unit 7 in the resolution conversion unit 3. Further details are as follows. The original image color perception predicting unit 5 is perceived by a person from a three-dimensional color system image represented by a three-dimensional color system obtained by converting the original image stored in the storage unit 2. The color perception processing for predicting a wax color is performed.

Further, the display image color perception predicting section 6 performs the same color perception processing as the above-described color perception processing on the image obtained when the reduced image V is displayed on the display means 4, and the person It predicts the color that will be perceived by. Furthermore, the reduced image optimization unit 7 compares the prediction result of the original image color perception prediction unit 5 with the prediction result of the display image color perception prediction unit 6, and modifies the reduced image V so that the difference between the two is minimized. Is to add.

As a result, the above-mentioned color shift can be made as inconspicuous to the human eye as possible.

[0019]

FIG. 2 is a flow chart showing the basics of the color image display method according to the present invention. In the present invention, the calculation method for minimizing the difference in color perception is not particularly limited, but FIG. 2 shows the flow of processing when the difference in color perception is minimized by using the iterative method.

First, the color perception of the original image is predicted. Next, the original image is reduced by a simple method to use it as the initial value of the reduced image V. Further, the color perception when the reduced image V is displayed is predicted. Determine if the difference in color perception is minimal,
If it is not the minimum, the reduced image V is slightly changed and corrected, and the color perception is predicted for the corrected reduced image.

This process is repeated, and when it is determined that the difference in color perception is the minimum, the process is terminated and the reduced image V is displayed on the display means 4.
Is displayed. More specifically, the method according to the present invention comprises
A color image display method including display means for displaying a color image in which each pixel is composed of a plurality of sub-pixels representing a basic color, and has the following steps.

An original image, which is stored in advance in the storage unit 2 and has a resolution higher than that of a color image to be displayed on the display means 4, is converted into a three-dimensional color system image represented by a three-dimensional color system. It has a step of converting and a step of converting the resolution of the original image into the resolution of the display means 4 to generate the reduced image V. Further, the first step (see step S11 in FIG. 2) of predicting a color that a person would perceive from the above-mentioned three-dimensional color system image and the reduced image V when the reduced image V is displayed on the display means 4 are obtained. The second step (see step S13 in FIG. 2) of predicting the color that a person will perceive from the captured image is compared with the prediction result of the first step and the prediction result of the second step. A third step of modifying the reduced image V to reduce the difference.

In this case, the processing loop of the third step and the second step is repeated until the difference obtained in the third step is minimized (steps S13 → S14 → S15 in FIG. 2). → See S13). The resolution of the display unit 4 described above represents the resolution in pixel units. The arrangement of the sub-pixels (p) in the pixel (P) unit includes a stripe arrangement, a mosaic arrangement, a delta arrangement, etc., but the present invention is applicable to all types of arrangements.

In this case, each sub-pixel is a vertical stripe in the normal stripe arrangement, but the present invention is also applicable to a stripe arrangement composed of sub-pixels in a horizontal stripe. As mentioned above, the basic colors are additive three primary colors (RGB) and subtractive three primary colors (CM
Not limited to Y), as long as it is composed of a plurality of colors,
It can be applied to any color display.

Thus, according to the above-mentioned image display method, it is possible to display a color image having a higher spatial resolution than that in the case where the spatial arrangement of the sub-pixels is not considered (conventional) without causing a sense of discomfort due to color shift. . [Three-dimensional color system image] First, the above-described "three-dimensional color system image represented by the three-dimensional color system" will be described in more detail. This has the following two modes (i) and (ii).

(I) The three-dimensional color system is a uniform perceptual color space. Therefore, the difference between the prediction results is defined within this uniform perceptual color space. (Ii) The three-dimensional color system is a three-dimensional color space defined by the luminance axis, the red-green axis, and the blue-yellow axis that are orthogonal to each other. Therefore, the difference between the above-mentioned prediction results is defined in this three-dimensional color space.

First, a supplementary description will be given of the above (i). Referring to FIG. 1, it is desirable that the original image color perception predicting unit 5 and the display image color perception predicting unit perform filter processing or the like that matches the visual sensitivity of human color perception. This visual sensitivity generally represents the barely perceptible amount of stimulation (threshold value). Therefore, the visual sensitivity of a color represents the stimulus amount at which the color is barely visible.

On the other hand, in the reduced image optimizing unit 7, the above 2
The reduced image V is optimized so that the difference between the outputs of the two color perception prediction units 5 and 6 is minimized. In this case, the difference is determined depending on how to take the parameters of the three-dimensional color system. From the viewpoint of visual sensitivity, it is desirable that the metric of the color in the three-dimensional space is isotropic (direction-independent) and uniform. The space defined by such parameters is the above-mentioned uniform perceptual color space. As this uniform perceptual color space, for example, CIE (Commission
1 recommended by Internationale de 1'Eclairage)
There is one L * u * v * color system. Where L * represents lightness, u * and v * represent chromaticity, respectively, and generally CIEL
Called UV. The u * and v * are mostly u'v '
In the chromaticity coordinates, it is proportional to the difference from the chromaticity of the achromatic color multiplied by the lightness L *, but what is important is that the normal chromaticity is multiplied by the lightness. This is 2 as the lightness decreases.
The difference between the two colors is difficult to distinguish. In addition,
In the two-dimensional space where u'v 'is orthogonal to the brightness, the axis and the metric are selected so that the sensory color difference and the color difference quantified by u'v' are as proportional as possible ( References: Image Analysis Handbook, University of Tokyo Press, 1991, p.
107 ~).

Further, when there is no choice but to approximate with a linear filter due to processing speed constraints and resolution conversion algorithm constraints, the linear filters are designed so that the difference is as uniform as possible. It is desirable to do.
Next, a supplementary description will be given of the above (ii). In this (ii), the above-described three-dimensional color system is expressed in three dimensions defined by the luminance axis, the red-green axis, and the blue-yellow axis, and is included in the processing performed by the original image color perception prediction unit 5. , So that a filtering process that allows a high-frequency component to pass more on the red-green axis than on the blue-yellow axis is included.

That is, in the above-described first step, the high-frequency component of the signal of the three-dimensional color system image is made to pass more on the red-green axis than on the blue-yellow axis. Regarding the above-mentioned luminance axis, it is desirable to pass a higher frequency component than the red-green axis, or to perform no filtering at all. As a three-dimensional color system, YCrCb format or YI
Any colorimetric system represented by luminance and two-dimensional chromaticity having different band characteristics such as Q format is included.

More preferably, the step of detecting the edge part of the brightness change in the image composed of the brightness component represented by the brightness axis, and the step other than the detected edge part in the image. In the area, a step of limiting a change in the color specification value of the pixel is further included. By this edge detection, the color shift can be removed more accurately. This is for the following reason.

According to the combined MRF (Markov Random Field) model proposed as a calculation theory for vision, at a place where light and dark discontinuity is generated, that is, at an edge portion in an image of a luminance component, other color etc. Discontinuities regarding clues are likely to occur. According to this concept, it is rational to consider that in a region where no edge is detected in the luminance component, discontinuity of colorimetric values does not occur, and only the smooth change is limited.

Based on this idea, the colors of the GBR array are conspicuous in the black background (GB on the right side of FIG. 11A).
R), it can be explained that the colors of the GBR array are not so noticeable in the white background (the GBR on the right side of FIG. 11C). This will be described with reference to the drawings. FIG. 3 is a diagram for explaining edge detection. GBR especially on a black background
It is a schematic diagram simulated by the model in the state where one pixel is displayed in the array.

FIG. 3A shows the process of edge detection for the luminance axis component, and FIG. 3B shows the process of edge detection for the red-green axis component. However, for simplification, only one dimension along a line crossing the sub-pixel p is shown. In the simulation shown in FIG. 3, the luminance values Y _R , Y _G, and Y _{B of} R, G, and B used as the three primary colors are normally
It is used to satisfy the relation of Y _G > Y _R > Y _B.

The upper part (a) of the figure shows a graph about the luminance axis, and the lower part (b) shows a graph about the red-green axis. GBR displayed at the left end of each of (a) and (b)
The central graph is obtained by applying the filter process F to the graph representing the state of. Further, with respect to the graph at the center of the luminance axis, edge detection processing E is further performed to obtain an edge portion indicated by a black circle at the right end.

On the other hand, with respect to the graph at the center of the red-green axis, the correction process M is performed so that it changes discontinuously in the edge part of the black circle and changes smoothly in the other regions, and the graph at the right end is obtained. The graph at the right end obtained here is a graph that is considerably swayed along the red-green axis (red-green is conspicuous). As a result, a more accurate perceptual prediction value can be obtained. Continuing to refer to FIG.

FIG. 4 is a diagram for explaining the edge detection. Contrary to the case of FIG. 3, the GBR is especially on a white background.
FIG. 12 is a schematic diagram simulated by a model in a state where one pixel is displayed in an array (right GBR in FIG. 11C). This Figure 4
Is exactly the same as that of FIG. 3 described above. Figure 4
In (a), since only two edge parts are detected from the left end graph of the luminance axis (right end of this figure (a)), the right end graph of the red-green axis of this figure (b) is the center part. The two peaks cancel each other out, resulting in a graph that does not move much along the red-green axis (red-green is not noticeable). As a result, a more accurate perceptual prediction value can be obtained.

Next, a method of generating a binary image, for example, a black and white image as an original image in the storage unit 2 of FIG. 1 will be described, which is different from the above-described mode. According to this method, the above-mentioned color shift is inconspicuous and a binary image with high spatial resolution can be obtained. Specifically, a plurality of sub-pixels p
Is divided into a first sub-pixel group that becomes yellow when driven simultaneously and a second sub-pixel group that exhibits blue when driven, and these first and second sub-pixel groups are individually driven and controlled. The binary image is displayed on the display means 4.

The main items of the color image display apparatus and method according to the present invention have been described above, and a detailed example thereof will be supplemented below. FIG. 5 is a diagram showing the relationship between the original image, the reduced image, and the display image. 1A is the original image data stored in the storage unit 2 of FIG. 1, and FIG. 1B is being optimized by the reduced image optimizing unit 7 of FIG. 1 (to minimize the difference described above). The data of the reduced image V, as shown in FIG. 6C, is the data of the display image in the display means 4 that should be actually displayed on the screen.

Referring to FIG. 5, the display means 4 has 3 pixels ×
A color liquid crystal display of a stripe arrangement capable of displaying a color image of 1 pixel, on which a white dot having a size of 1 pixel (456 of FIG.
White reference) is assumed to be displayed slightly to the right. In FIG. 5, the row of pixels is shown in the horizontal direction, and the RGB values of each pixel are shown in the vertical direction. In this case, the storage unit 2 stores a desired original image of 9 pixels × 1 pixel. In the original image, the pixels 4 to 6 are made white as described above. The reason that the value is 1/3 instead of 1 is to make the normalization process unnecessary in the subsequent process.

The reduced image in FIG. 5B is one candidate as a result of reducing the original image. The color shift when it is assumed that the image is reduced like this candidate image is calculated as follows. Note that the display image in FIG. 5C is an image showing a reduced image displayed on the display unit 4. However, the hatched part in the figure must be 0. For example, pixel 0 is red (R), so GB is always 0,
Since pixel 1 is green (G), RB is always 0, and pixel 2 is blue (B), so RG is always 0. The same applies hereinafter.

FIG. 6 is a diagram showing the processing of the perception prediction units 5 and 6 (FIG. 1). Here, the RGB values are used as they are as the three-dimensional color system. However,
It is desirable to perform γ correction so that the RGB values are proportional to the brightness when displayed. Left column of this figure (a)
5A and 5A 'show the pixel values of the original image of FIG. 5A and the display image of FIG. 5C as a graph of RGB values.

The signals shown in FIGS. 5A and 5A 'are filtered according to the visual sensitivity characteristic. The filters used here are visual sensitivity characteristics with respect to luminance, and are all the same filter. The result of the filtering process is represented by the graphs in the right columns (b) and (b ') of FIG. (B) represents a state of a color (predicted value of a color desired to be shown to a person) that is expected to be perceived when the original image is presented on the high-resolution display unit 4, while (b ′) ) Represents a color state that a person may actually perceive when displaying reduced image candidates. Therefore, for example, for the pixel 5, the RGB value of A in FIG. 6B and the B value of B in FIG.
The color difference between the R, G, and B values represents the color shift in question.

Therefore, in the reduced image optimizing unit 7, the perceived predicted color of the original image is obtained by the sum of squares of the differences in RGB values at all sampling points (sampling points corresponding to each pixel in FIGS. 6B and 6B '). The difference from the perceived predicted color of the display image is set, and the loop (S1
3 → S14 → S15 → S13) to determine the reduced image.

In the above description, the RGB values are used as they are in order to obtain the color difference. However, the present invention is not limited to this, and the color difference may be obtained after conversion into LUV values. The case of using this LUV value will be described below. However, up to the point of performing the filtering process according to the visual sensitivity characteristic, it is the same as the detailed example in which the RGB value is used as it is.

After the filtering process, as a final step of each process in the original image color perception predicting unit 5 and the display image color perception predicting unit 6, the filtered RGB value is converted into the CIELUV value for each sampling point. And the two-color perception predicting units 5 and 6 with this conversion value.
Get each prediction result of. The reduced image optimizing unit 7 determines the Euclidean distance between the two colors based on the prediction result in the CIE LUV color space as the color difference at each of the sampling points. Then, the sum of squares of the color differences at all sampling points is used as the difference between the prediction results of the perceptual prediction units 5 and 6.

Further, a full search is performed to find an optimum reduced image V that minimizes the above difference. That is, the above differences are calculated for all the candidates for the reduced image composed of combinations of pixel values, and the candidate for the reduced image when the minimum value of the differences is taken is determined as the reduced image to be displayed on the display means 4. . Although it has been described above that the three-dimensional color system is the uniform perceptual color space, in this case, the above-mentioned LUV is used.
Unlike the detailed example using the value, the above difference can be calculated directly from the result of the filtering process. This is because when the uniform perceptual color space is used, the data of the original image of FIG. 5A is subjected to LUV conversion in advance to be the data of the original image.

Next, the above-mentioned "three-dimensional color system is a three-dimensional color space defined by the luminance axis, the red-green axis and the blue-yellow axis which are orthogonal to each other, and the difference between the prediction results is A detailed example of the resolution conversion unit 3 under the configuration example of "defining in three-dimensional color space" will be described. As the color system at this time, the conversion formula M ₁ = Y−X M ₂ = Y−Z W = Y of the router / Niberg color solid is used.

Here, M ₁ and M ₂ represent color moments,
W represents brightness. Further, X, Y, and Z represent tristimulus values in the CIEXYZ space. Furthermore, the two color moments,
To M _1, M _2, and _{M 1 '= λ 1 M 1} M 2' = λ 2 M 2 is multiplied by lambda ₁ and lambda ₂ representing the brightness and the ratio of a measure of the chromaticity axis. In addition, the color difference

[0050]

[Equation 1]

## EQU1 ## When the above is determined, λ ₁ and λ ₂ are determined in advance by trial and error so that the degree of contribution to the color difference from each of the luminance axis, the red-green axis and the blue-yellow axis is as uniform as possible. Keep it. In this case the transformation is a linear transformation from RGB values, where W, M ₁ 'and M ₂ ' represent approximate brightness, red-green and blue-yellow axes. As the filter processing applied to each axis, Gaussian filter processing having different spatial characteristics is performed for each axis.

In the reduced image optimizing unit 7, the reduced image candidates are sequentially updated using the sum of squares of the differences in colorimetric values as an evaluation function,
Find the optimal reduced image. Next, for the image composed of the luminance component represented by the luminance axis described above, the step of detecting the edge portion of the luminance change in the image, and in the area other than the detected edge portion in the image The step of limiting the change in the colorimetric value of the pixel, and the display image color perception prediction unit in the “color image display method including further steps” will be described with reference to the drawings.

FIG. 7 is a diagram showing a specific example of the processing of the display image color appearance prediction unit 6. First, a display image 11 having a sufficiently high resolution as if the display surface of the display means 4 was photographed is created. At this time, in the case of a color liquid crystal display, it is desirable to take into consideration the space occupied by the black matrix that fills the space between adjacent sub-pixels with black.

This display image 11 is converted into a luminance image 12 and a chromaticity image 13, and a filtering process is applied to each. Edges are further detected from the filtered luminance image 14 by using a Laplacian operator or the like, and regions such as white and yellow are divided (15). Then, for each of the divided areas, the average value of the luminance value and the chromaticity value in the area is obtained from the three filtered images, and the area is filled with the average value. The obtained image is the perception expected image 17.

Note that each pixel value is as close as possible to the pixel value of the filtered image under the constraint that the luminance and chromaticity in the area change smoothly with this image as the initial value. To update. This makes it possible to obtain a more accurate perceptual prediction image, that is, a prediction image that is closer to the appearance of human eyes. Finally, an application example using the method of the present invention will be described. [Application 1] FIG. 8 is a diagram for explaining the application 1.

The application example 1 can solve the problem that the color is visible at the jagged edges of the diagonal lines when a completely black-and-white font is displayed. Consider, for example, the case of creating a font used to display monochrome characters on a color liquid crystal display. A normal font is designed in pixel units as shown in FIG. However, with the present invention,
A font can be designed with three times the degree of freedom as shown in FIG. As a result, the jaggedness of the diagonal line is removed. However, if it is left as it is, it will not be completely monochrome, and it will appear colored. Therefore, the storage unit 2 of FIG.
Then, the image is displayed and corrected using the image display method of the present invention. [Application Example 2] Consider a case where a font used to display Morokuro characters is created on a color liquid crystal display, for example. For this reason, the above-described color image display method, that is, "a plurality of subpixels are divided into a first subpixel group that becomes yellow when driven simultaneously and a second subpixel group that exhibits blue when driven, A color image display method in which the first and second sub-pixel groups are individually driven and controlled to display a binary image on the display unit 4 "is applied.

In the case of the application example 1 described above, a complete morokuro font can be obtained. On the other hand, however, the monochrome font was found to cause blurring of characters. The reason for this is that the fonts are created without considering the effects of red and green, so that a large amount of modification is required to erase unnecessary colors. In other words, this drastic correction causes blurring of characters.

FIG. 9 is a diagram for explaining the application example 2. In the application example 1 described above, when it is required to eliminate the above-described character blur, as shown in FIG.
Each pair of adjacent red and green pixels is treated as a yellow group (yellow in the figure). Then, the pixel value can be controlled independently of the blue group including only the remaining blue pixels.

In this case, the degree of freedom in font design is twice that of the normal method, but since the blue-yellow axis is perceived only on the low spatial frequency side, the luminance distribution is relatively preserved, and the image display method of the present invention. Even if the color perception is optimized by using, the blurring is less likely to occur. The background color and the character color are not limited to white and black, and monochrome display using any two colors is possible.

The embodiments of the present invention described above are as follows. (Supplementary Note 1) In a color image display device including a display unit for displaying a color image in which each pixel is composed of a plurality of sub-pixels representing a basic color, a resolution higher than the resolution of the color image to be displayed on the display unit is set. And a resolution conversion unit that converts the resolution of the original image stored in the storage unit into the resolution of the display unit to generate a reduced image. Here, the resolution conversion unit may be perceived by a person from a three-dimensional color system image represented by a three-dimensional color system, which is obtained by converting the original image stored in the storage unit. An original image color perception predicting unit that performs a color perception process that predicts a deep color and an image obtained when the reduced image is displayed on the display unit perform the same color perception process as the color perception process. , People perceive from the image A display image color perception prediction unit that predicts a color, compares the prediction result of the original image color perception prediction unit and the prediction result of the display image color perception prediction unit, and reduces the reduced image so that the difference between them is minimized. A color image display device, comprising: a reduced image optimizing unit that corrects the image.

(Supplementary Note 2) A color image display method including a display means for displaying a color image in which each pixel is composed of a plurality of sub-pixels representing a basic color, wherein the display means is stored in advance in a storage unit. A step of converting an original image having a resolution higher than that of the color image to be displayed into a three-dimensional color system image represented by a three-dimensional color system, and the resolution of the original image by the display means. To the resolution of
A first step of predicting a color that a person will perceive from the three-dimensional color system image, and a step of generating a reduced image, and a step of obtaining the reduced image when the reduced image is displayed on the display means. The second step of predicting a color that a person will perceive from the image displayed, the prediction result of the first step, and the prediction result of the second step are compared to reduce the difference between the two. And a third step of modifying the reduced image.

(Supplementary Note 3) The third step and the second step are performed until the difference obtained in the third step is minimized.
3. The color image display method according to attachment 2, wherein the processing loop of steps is repeated. (Supplementary note 4) The color image display according to supplementary note 2, wherein the three-dimensional color system is a uniform perceptual color space, and the difference between the prediction results is defined in the uniform perceptual color space. Method.

(Supplementary Note 5) The three-dimensional color system is a three-dimensional color space defined by a luminance axis, a red-green axis and a blue-yellow axis which are orthogonal to each other, and the difference between the prediction results is 3. The color image display method according to attachment 2, wherein the color image is defined in the three-dimensional color space. (Supplementary Note 6) In Supplementary Note 5, wherein the first step includes a filtering process that allows the high-frequency component of the signal of the three-dimensional color system image to pass through more in the red-green axis than in the blue-yellow axis. Color image display method described.

(Supplementary Note 7) A step of detecting an edge portion of a luminance change in the image composed of the luminance component represented by the luminance axis, and a region other than the detected edge portion in the image. 6. The method for displaying a color image according to Supplementary Note 5, further comprising: limiting the change in the colorimetric value of the pixel. (Supplementary Note 8) The plurality of sub-pixels are divided into a first sub-pixel group that becomes yellow when driven simultaneously and a second sub-pixel group that exhibits blue color when driven, and the first and second sub-pixels are divided. 3. The color image display method according to appendix 2, wherein the sub-pixel groups are individually driven and controlled to display a binary image on the display means.

[0065]

As described above, according to the present invention, (1) By using the original image color perception prediction unit 5, the display image color perception prediction unit 6, and the reduced image optimization unit 7,
It is possible to display a color image with a higher spatial resolution, as compared with the conventional example in which the spatial arrangement of the subpixels is not considered without giving the viewer a feeling of strangeness due to the color shift.

(2) As a three-dimensional color system, a uniform perceptual space or a three-dimensional color space defined by mutually orthogonal luminance axis, red-green axis and blue-yellow axis is used.
It is possible to eliminate the color shift caused by the difference in the arrangement of the sub-pixels (RGB or GBR). (3) According to the processing that incorporates the edge detection described above,
Color shift can be eliminated more accurately.

(4) According to the processing divided into the yellow group and the blue group described above, the color shift is not noticeable, and
A binary image with high spatial resolution can be obtained.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of a color image display device according to the present invention.

FIG. 2 is a flowchart showing the basics of a color image display method according to the present invention.

FIG. 3 is a diagram for explaining edge detection.

FIG. 4 is a diagram for explaining edge detection.

FIG. 5 is a diagram showing a relationship between an original image, a reduced image, and a display image.

FIG. 6 is a diagram showing processing of perceptual prediction units 5 and 6 (FIG. 1).

FIG. 7 is a diagram showing a specific example of processing of a display image color appearance prediction unit 6.

FIG. 8 is a diagram for explaining an application example 1;

FIG. 9 is a diagram for explaining an application example 2;

FIG. 10 (a) shows a typical arrangement of sub-pixels,
(B) is a figure showing the arrangement of the sub-pixels whose arrangement order is changed.

11 (a), (b) and (c) are diagrams showing that colors are conspicuous differently due to the background and the arrangement order of sub-pixels.

[Explanation of symbols]

1 ... Image display device 2 ... memory 3 ... Resolution converter 4 ... Display means (color liquid crystal display) 5 ... Original image color perception prediction unit 6 ... Display image color perception prediction unit 7 ... Reduced image optimization unit

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04N 1/46 H04N 9/64 Z 5C079 1/60 1/40 D 5C082 9/12 1/46 Z 9 / 64 G09G 5/36 520G F Term (reference) 5B057 AA01 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC02 CD05 CE16 CH01 DA16 5C060 BB01 BC01 HB00 JA20 5C066 AA03 CA09 GA01 QC06 5PP06P20P20Q5C0620 PP5A0622 5C079 HB01 HB09 HB11 LA24 LA37 LB01 MA17 NA03 PA05 5C082 BA34 BD02 CA12 CA34 DA51 MM10

Claims (5)

[Claims] 1. A color image display device including a display means for displaying a color image in which each pixel is composed of a plurality of sub-pixels representing a basic color, wherein the resolution is higher than the resolution of the color image to be displayed on the display means. A storage unit for storing an original image having the following, and a resolution conversion unit for converting the resolution of the original image stored in the storage unit into the resolution of the display unit to generate a reduced image. Here, the resolution conversion unit is obtained by applying a conversion to the original image stored in the storage unit, from a three-dimensional color system image represented by a three-dimensional color system,
An original image color perception prediction unit that performs a color perception process that predicts a color that a person will perceive, and an image obtained when the reduced image is displayed on the display unit, similar to the color perception process. A display image color perception prediction unit that performs color perception processing and predicts a color that a person will perceive from the image, a prediction result of the original image color perception prediction unit, and a prediction result of the display image color perception prediction unit And a reduced image optimizing unit that corrects the reduced image so that the difference between the two is minimized, and a color image display device. 2. A color image display method including a display means for displaying a color image in which each pixel is composed of a plurality of sub-pixels representing a basic color, the display means pre-stored in a storage section. A step of converting an original image having a resolution higher than that of the color image to be converted into a three-dimensional color system image represented by a three-dimensional color system, and the resolution of the original image in the display means. A step of converting the image into a resolution to generate a reduced image, a first step of predicting a color that a person may perceive from the three-dimensional color system image, and the display means for displaying the reduced image. The second step of predicting a color that a person would perceive from the image obtained when displayed on the screen, the prediction result of the first step and the prediction result of the second step are compared, and both of them are compared. To reduce the difference Color image display method characterized by further comprising a third step of applying a correction to the serial reduced image. 3. The color image display according to claim 2, wherein the processing loop of the third step and the second step is repeated until the difference obtained in the third step is minimized. Method. 4. The color according to claim 2, wherein the three-dimensional color system is a uniform perceptual color space, and the difference between the prediction results is defined in the uniform perceptual color space. Image display method. 5. The three-dimensional color system is a three-dimensional color space defined by a luminance axis, a red-green axis, and a blue-yellow axis that are orthogonal to each other, and the difference between the prediction results is the cubic order. The color image display method according to claim 2, wherein the color image is defined in the original color space.