JP2013025141A - Image processor, image display device and image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance image quality in visual appreciation by utilizing a color and a position of a sub-pixel corresponding to a white color in a case where a color display is performed by sub-pixels of four colors.SOLUTION: A display device displays an image by sub-pixels of four colors of R (red), G (green), B (blue) and W (white). A sub-pixel G is positioned in a diagonal direction as viewed from a sub-pixel W. Further, a sub-pixel R is adjacent to the sub-pixel W with respect to an X direction, lateral direction in a FIG., and a sub-pixel B is adjacent to the sub-pixel W with respect to a Y direction, longitudinal direction in the FIG. In such configuration, high frequency components of an image signal of R in the Y direction are restricted and high frequency components of an image signal of B in the X direction are restricted. Furthermore, an image signal of G and an image signal of W regard individual sub-pixels as sub-pixels of the same color and perform the same band restriction to the individual sub-pixels. In addition, for colors of R, G and B, the display device adjusts amplitude of high frequency components to be passed through by parameters in response to ratios of the high frequency components of individual colors to high frequency components of W.

Description

  The present invention relates to image processing in the case of performing color display with four color sub-pixels.

  The pixel arrangement in the display device using the three primary colors includes a stripe arrangement and a delta arrangement (see, for example, Patent Document 1). In such a display device, one pixel is constituted by three sub-pixels. In the Bayer array, the number of colors is three, but one pixel is constituted by four subpixels. Two G (green) subpixels, R (red), and B One (blue) sub-pixel is used, and two of these sub-pixels are arranged in the vertical and horizontal directions.

  In a display device in which pixels are configured by a Bayer array, color display is generally performed by image data having a number of pixels that is 1/4 of the number of pixels of input image data. In this case, since the resolution of the actually used image data is lower than the resolution of the input image data, a filter process for limiting the frequency band of the image signal is executed to suppress moire caused by aliasing noise. Is done. For example, in the case of an R or B image signal, in order to prevent moire caused by a high frequency component, it is necessary to limit both the vertical and horizontal frequency bands to half (that is, 1/2). Note that the G image signal has twice the number of subpixels as compared to the R and B subpixels, and therefore the bandwidth limit is smaller than these.

  On the other hand, for the purpose of improving color reproducibility and brightness, color display may be performed using four primary colors (or more primary colors). For example, Patent Document 2 discloses an image display device in which one of G sub-pixels in a Bayer array is replaced with W (white) or C (cyan) and color display is performed with four color sub-pixels. Patent Document 3 discloses a technique for calculating a color image signal of each color when a color image display unit having four or more primary colors is provided. The “primary colors” referred to here are colors that are the sources of (additive) color mixing, and are not limited to the three primary colors of light.

JP 07-006703 A JP 2006-267541 A JP 2000-338950 A

When color display is performed with sub-pixels of four colors, if the band of the image signal is limited independently for each color, the image signal of any sub-pixel is limited to half the frequency band in both the vertical and horizontal directions. The However, when such band limitation is performed, a large amount of information is lost from the original image.
Therefore, the present invention has an object of improving the visual image quality by using the color and position of a sub-pixel corresponding to white when performing color display with four sub-pixels.

According to an image processing apparatus of one embodiment of the present invention, pixels each including four sub-pixels arranged in a first direction and a second direction intersecting the first direction are arranged in a matrix. An image processing apparatus for supplying image signals of four colors including white to a display device in which one of the sub-pixels constituting one pixel corresponds to white, each of the four color image signals A high-frequency component extracting unit that extracts a component in a high-frequency band that is higher in frequency than the center of the frequency band of the image signal, and components of three colors other than white based on the components of each color extracted by the high-frequency component extracting unit A parameter calculation unit that calculates a parameter determined according to a ratio between each component and a white component, and for each of the four color image signals, according to the arrangement of the sub-pixels corresponding to the color A filter processing unit that limits a wave number band and performs a filter process for adjusting the amplitude of the high-frequency band by using the parameter calculated by the parameter calculation unit for each of three color image signals other than white Is provided.
According to this image processing apparatus, it is possible to improve the visual image quality by adjusting the amplitude of the high frequency band by the parameter.

In a preferred aspect, the filter processing unit applies the frequency band in the first direction and the second direction to the white image signal for displaying the first sub-pixel corresponding to white, and the first sub-pixel and the second sub-pixel. The first filter processing unit that restricts the first sub-pixel in a manner corresponding to the arrangement of the second sub-pixels positioned in the diagonal direction with respect to the first sub-pixel, and the image signal for displaying the second sub-pixel in the first direction And a second filter processing unit that restricts the frequency band in the second direction in a manner according to the arrangement of the first subpixel and the second subpixel, and the first subpixel is adjacent to the first direction. A third filter processing unit configured to limit a frequency band in the second direction with respect to an image signal for displaying the third sub-pixel in a manner according to an interval of the third sub-pixel in the second direction; Sub-pixel and A fourth filter that restricts the frequency band in the first direction with respect to the image signal for displaying the fourth sub-pixel adjacent in the second direction in a manner corresponding to the interval in the first direction of the fourth sub-pixel. And a processing unit.
According to this aspect, it is possible to limit the frequency band of the image signal according to the arrangement of the first sub-pixel corresponding to white display and the other sub-pixels.

In another preferred embodiment, at least one of the third filter processing unit and the fourth filter processing unit is a band other than a band to pass among frequency bands of an input image signal, and the first filter The band of the band to be passed by the processing unit is inverted and passed, and the amplitude of the band is adjusted using the parameters.
According to this aspect, it is possible to suppress moire and false color resulting from the display of the first sub-pixel.

In still another preferred aspect, the image processing device is configured such that the frequency response of the band that is passed with the phase inverted is such that the first subpixel, the second subpixel, the third subpixel, and the fourth subpixel are transmitted. It is determined according to the luminance of the pixel.
According to this aspect, it is possible to suppress moire and false colors according to the characteristics of the sub-pixel.

In still another preferred aspect, the filter processing unit refers to an image signal representing 4M pixels in the first direction and 4N pixels in the second direction, and for each color, the M signals in the first direction, An image signal representing the N sub-pixels in the second direction is output, and the parameter calculation unit calculates the number of parameters corresponding to the sub-pixels represented by the image signal output by the filter processing unit.
According to this aspect, calculation of unnecessary parameters can be omitted.

In still another preferred embodiment, the filter processing unit performs the filter processing by adjusting a predetermined specific filter output for each of the three color image signals other than white using the parameters. Run. According to this aspect, it is possible to reduce the memory required for storing the filter as compared with the case of using a plurality of filters.
Alternatively, the filter processing unit, for each of the three color image signals other than white, from a plurality of filters determined for each of the plurality of parameters, according to the parameter calculated by the parameter calculation unit You may make it perform the said filter process by selecting. According to this aspect, it is possible to reduce the amount of calculation at the time of executing the filter process, compared to the case where a specific filter is used.

In still another preferred embodiment, the second filter processing unit performs a filter process on an image signal corresponding to green, and each of the third filter processing unit and the fourth filter processing unit is red or blue. Filter processing is performed on the image signal corresponding to.
According to this aspect, it is possible to increase the resolution of a green image (having high human eye spectral sensitivity) as compared to red or blue.

A display device according to another aspect of the present invention includes a plurality of pixels configured by the first sub-pixel, the second sub-pixel, the third sub-pixel, and the fourth sub-pixel, and arranged in a matrix. A display panel and the image processing apparatus according to claim 1.
According to this display device, the frequency band of the image signal is limited according to the arrangement of the first sub-pixel corresponding to the white display and the other sub-pixels, and it is possible to improve the visual image quality. .

According to another aspect of the present invention, there is provided an image processing method in which each pixel configured by four sub-pixels arranged in a first direction and in a second direction intersecting the first direction is arranged in a matrix. An image processing method for supplying four color image signals including white to a display device in which one of the sub-pixels constituting one pixel corresponds to white, each of the four color image signals To extract a component in a high frequency band that is higher in frequency than the center of the frequency band of the image signal, and based on the extracted color components, for each of the three color components other than white, the respective components and the white component And a frequency band corresponding to the arrangement of the sub-pixel corresponding to the color for each of the four color image signals, and for three colors other than white Each of the image signals Respect, executes the filtering process be adjusted using the parameters of the amplitude is the calculation of the high frequency band.
According to this image processing method, the frequency band of the image signal is limited according to the arrangement of the first sub-pixel corresponding to white display and the other sub-pixels, and it is possible to improve the visual image quality. is there.

1 is a block diagram showing a configuration of a display system 10. FIG. The figure which shows arrangement | positioning of the pixel of a liquid crystal panel. The block diagram which shows the structure of an image process part. The figure which shows the characteristic of the filter of each color. The figure which shows the characteristic of the filter shown as a comparative example. The figure which shows the characteristic of the filter shown as a comparative example. The figure which shows the grating | lattice formed by the subpixel of a Bayer arrangement. The figure for demonstrating the moire which arises in 1st Embodiment. The figure which shows the relationship between the gain adjustment parameter of each color, and the frequency response of a filter. The figure for demonstrating the frequency band which causes a moire. The figure for demonstrating the filter of R and B, and its frequency characteristic. The figure which shows the other example of the filter of R and B. The figure which shows the other example of arrangement | positioning of a subpixel.

[First Embodiment]
FIG. 1 is a block diagram showing a configuration of a display system 10 according to an embodiment of the present invention. A display system 10 shown in FIG. 1 includes an image processing apparatus 100, a projector 200, and a screen 300, and projects an image according to image data supplied from an external apparatus onto the screen 300. The external device is, for example, a personal computer, but may be a digital still camera or the like. Note that the image data supplied by the external device is data in which each pixel is expressed by image signals of three color components of R (red), G (green), and B (blue).

  The projector 200 is a single-plate projector, and includes a light source 210, one liquid crystal panel 220, and a projection lens 230. The liquid crystal panel 220 functions as an optical modulator that modulates incident light, and changes the light transmission state in a manner corresponding to image data supplied from the image processing apparatus 100, thereby irradiating light emitted from the light source 210. Controls the degree of transmission. The projection lens 230 projects the light transmitted through the liquid crystal panel 220 onto the screen 300.

The liquid crystal panel 220 is a flat display panel in which one sub-pixel is constituted by four sub-pixels, and the light transmission state of each sub-pixel is different by a color filter or the like. Each sub pixel corresponds to display of four colors of R (red), G (green), B (blue), and W (white). The arrangement of the pixels of the liquid crystal panel 220 is obtained by replacing one of the two G subpixels in the Bayer array with a W subpixel. In addition, as long as white (white) here contains all the color components of R, G, and B in an appropriate ratio, it may be slightly yellowish or grayish. However, when the luminance values of the R, G, B, and W sub-pixels are Y _R , Y _G , Y _B , and Y _W , respectively, it is ideal that these satisfy the relationship Y _W = Y _R + Y _G + Y _B. In this embodiment, it is assumed that such a relationship is satisfied.

  FIG. 2 is a diagram illustrating an arrangement of pixels of the liquid crystal panel 220. In the liquid crystal panel 220, sub-pixels are arranged side by side in the X direction in the figure and in the Y direction perpendicular to the X direction, and two sub-pixels adjacent to each other in the X direction, the two sub-pixels, and the Y-direction One sub-pixel is composed of two sub-pixels adjacent in the direction. In the liquid crystal panel 220, sub-pixels corresponding to G and B are arranged in order in odd-numbered rows along the X direction, and sub-pixels corresponding to R and W are arranged in order in even-numbered rows along the X direction. It is out. In the liquid crystal panel 220, subpixels corresponding to G and R are arranged in order in the odd-numbered columns along the Y direction, and subpixels corresponding to B and W are arranged in the even-numbered columns along the Y direction. Are lined up. Here, the X direction corresponds to the first direction in the present invention, and the Y direction corresponds to the second direction in the present invention.

  Here, it is assumed that each sub-pixel is a square of the same size, and the interval between them is equal in both the X direction and the Y direction. In this way, the four subpixels constituting one pixel are arranged in a square shape. The liquid crystal panel 220 is configured by arranging such square pixels in a matrix in the X and Y directions. Here, the G sub-pixel is located in the diagonal direction as viewed from the W sub-pixel, is adjacent to the B sub-pixel in the X direction, and is adjacent to the R sub-pixel in the Y direction. The R subpixel is adjacent to the W subpixel in the X direction and is adjacent to the G subpixel in the Y direction. The B subpixel is adjacent to the W subpixel in the Y direction and is adjacent to the G subpixel in the X direction.

  The W sub-pixel is a sub-pixel corresponding to white, and changes the light transmission state according to the image signal corresponding to white. Similarly, the R, G, and B sub-pixels are sub-pixels corresponding to red, green, and blue, respectively, and change the light transmission state according to the image signal corresponding to the display of each color. These sub-pixels of the liquid crystal panel 220 are driven by a drive circuit (not shown) to change the light transmission state.

  In the following, each sub-pixel corresponding to white is referred to as “sub-pixel W”, and an image signal for displaying the sub-pixel W is referred to as “W signal”. It is assumed that the sub-pixel and the image signal to be distinguished are distinguished. Here, the sub-pixel W corresponds to a first sub-pixel, and the sub-pixel G corresponds to a second sub-pixel. The sub pixel R corresponds to the third sub pixel, and the sub pixel B corresponds to the fourth sub pixel.

  The image processing apparatus 100 acquires image data from an external apparatus, converts the image data into image data suitable for display on the projector 200, and supplies the image data. That is, the image processing apparatus 100 generates image data including four color image signals of R, G, B, and W from image data including three color image signals of R, G, and B, and outputs the image data to the projector. 200.

  More specifically, the image processing apparatus 100 includes an input unit 110, an image processing unit 120, and an output unit 130. The input unit 110 acquires image data including image signals of three colors R, G, and B from an external device and inputs the image data to the image processing unit 120. The image processing unit 120 generates image signals of four colors R, G, B, and W by performing image processing to be described later on the image signal, and supplies this to the output unit 130 as image data. The output unit 130 supplies the image data generated by the image processing unit 120 to the projector 200.

  FIG. 3 is a block diagram illustrating a configuration of the image processing unit 120. As shown in the figure, the image processing unit 120 includes a color conversion processing unit 121, a high frequency component extraction unit 122, a parameter calculation unit 123, a filter processing unit 124, and a thinning processing unit 125. In more detail, the filter processing unit 124 includes a filter processing unit 124R, a filter processing unit 124G, a filter processing unit 124B, and a filter processing unit 124W.

  The color conversion processing unit 121 converts the three color image signals of R, G, and B into four color image signals of R, G, B, and W. The color conversion processing unit 121 can convert an image signal by a known method (for example, a method disclosed in Patent Document 3). The image signal is, for example, 8-bit data expressing the luminance (that is, brightness) with 256 gradations, but the number of gradations is not limited to this.

The high frequency component extraction unit 122 extracts high frequency components from each of the four color image signals R, G, B, and W generated by the color conversion processing unit 121. Here, the high frequency component refers to a component on the high frequency side from the center of the frequency band of the input image signal. In the following, the white high-frequency component extracted by the high-frequency component extraction unit 122 is expressed as H _w, and the other three colors are similarly expressed as H _r , H _g , and H _b , respectively. Note that the high frequency component extraction unit 122 may extract a high frequency component in each of the X direction and the Y direction, or may extract a high frequency component only in one direction.

  In the following, a band on the higher frequency side than the center of the frequency band of the input image signal is referred to as a “high frequency band”, and a band on the lower frequency side than the center of the frequency band of the input image signal. This is called “low frequency band”. That is, the high frequency component mentioned above is a component in the high frequency band here. In addition, the frequency here is a spatial frequency of an image.

The parameter calculation unit 123 calculates a parameter used in the filter processing unit 124 (hereinafter referred to as “gain adjustment parameter”) for each sub-pixel. The parameter calculation unit 123 calculates gain adjustment parameters G _r , G _g , and G _b represented by the following formulas for the three color components R, G, and B, respectively. That is, the gain adjustment parameters G _r , G _g , and G _b are determined according to the ratio of the high-frequency component of each color and the high-frequency component of W, respectively.

G _r = | H _w | / | H _r |
G _g = | H _w | / | H _g |
G _b = | H _w | / | H _b |

  The filter processing unit 124 performs filter processing on the four color image signals of R, G, B, and W. More specifically, the filter processing unit 124 includes filter processing units 124R, 124G, 124B, and 124W. Each of the filter processing units 124R, 124G, 124B, and 124W performs filter processing on the image signal corresponding to the code. For example, the filter processing unit 124R performs filter processing on the R signal. The filter processing unit 124W corresponds to the first filter processing unit, and the filter processing unit 124G corresponds to the second filter processing unit. The filter processing unit 124R corresponds to a third filter processing unit, and the filter processing unit 124B corresponds to a fourth filter processing unit.

  The filter processing unit 124 limits the frequency band corresponding to the arrangement of sub-pixels corresponding to the color display of each of the four color image signals of R, G, B, and W, and R, G Filter processing for adjusting the amplitude of the high frequency band using the gain adjustment parameter calculated by the parameter calculation unit 123 is performed on each of the three color image signals B and B. That is, unlike the filter processing units 124R, 124G, and 124B, the filter processing unit 124W does not perform gain adjustment (that is, amplitude adjustment) using the gain adjustment parameter.

  The filter processing unit 124 performs filter processing on a luminance linear signal (a signal in which the actual luminance change is linear with respect to the signal value change). When the input image signal is not luminance linear, the filter processing unit 124 converts the image signal so that the luminance is linear, executes the filter processing, and performs the inverse conversion after executing the filter processing to obtain the gamma value. Is restored. For example, when the color space of the image signal is defined by sRGB, the gamma value is “2.2”, so the image signal is not luminance linear. In this case, the filter processing unit 124 performs gamma conversion so that the gamma value of the image signal is changed from “2.2” to “1.0”, and then the gamma value is applied to the image signal that has been subjected to the filter processing. The gamma conversion is executed again so that the value changes from “1.0” to “2.2”.

  The following two methods can be given as a filtering method by the filter processing units 124R, 124G, and 124B. Note that the first method can reduce the memory required for storing the filter (that is, the set of filter coefficients) as compared to the second method. On the other hand, the second method can reduce the amount of calculation at the time of executing the filter process, compared with the first method.

  The first method is a method of executing filter processing by adjusting the output of a specific filter determined in advance using a gain adjustment parameter for each of the R, G, and B image signals. is there. The filter in this case is designed on the basis of a case where gain adjustment is not performed (for example, when a gray gray is displayed). In this case, the filter processing units 124R, 124G, and 124B attenuate the reference output obtained by the filter using a gain adjustment parameter as necessary.

  On the other hand, in the second method, the gain adjustment calculated by the parameter calculation unit 123 from each of a plurality of filters determined for each of a plurality of gain adjustment parameters for each of the three color image signals of R, G, and B. In this method, filtering is performed by selecting the one corresponding to the parameter. In this case, a plurality of filters including filters corresponding to the first method described above are stored in the memory that stores the filters, and the filter processing units 124R, 124G, and 124B are calculated for the image signals of the respective sub-pixels. The filter coefficient corresponding to the gain adjustment parameter is read out for each pixel.

  The thinning processing unit 125 executes thinning processing for reducing the number of pixels of the image signal of each color whose band is limited by the filter processing unit 124. The thinning processing unit 125 reduces the number of pixels of each color image signal to ¼ for each of the X direction and the Y direction. That is, the thinning processing unit 125 converts image data of 4M × 4N pixels on one side into image data of M × N pixels. However, the image data after conversion by the thinning-out processing unit 125 is data in which one pixel is composed of four subpixels. That is, when the number of pixels of the image data before conversion is compared with the total number of sub-pixels of the image data after conversion, the former is four times the latter (twice in the X direction and twice in the Y direction).

  The configuration of the display system 10 is as described above. With this configuration, the display system 10 converts the image data of 3 colors (RGB) 4M × 4N pixels into image data of 4 colors (RGBW) M × N pixels by the image processing apparatus 100, and the projector 200 An image corresponding to the image data is displayed on the screen 300. At this time, the filter processing unit 124 performs filter processing based on the configuration of the liquid crystal panel 220 in the projector 200.

  The filter processing executed by the filter processing units 124R, 124G, 124B, and 124W is determined according to the arrangement of the sub-pixels. Specifically, the filtering process for each color is performed in such a manner that the frequency band in each direction is limited and the gain is set to 0 in the band to be limited in a manner corresponding to the interval in the X direction and the Y direction of the sub-pixel of the color. To do. The filter processing of the present embodiment has a first feature, in particular, in that the display by the sub-pixel W (that is, white display) uses the color components of RGB colors. The filter processing of the present embodiment has a second feature in that a new problem (details will be described later) caused by the first feature is solved by a gain adjustment parameter. The following description is for the case where the filter processing unit 124 executes the filter processing by the above-described first method.

FIG. 4 is a diagram illustrating the characteristics of the filters of the respective colors according to the present embodiment. 5 and 6 show the characteristics of the filter shown as a comparative example of the present embodiment, and FIG. 5 is a general diagram assumed in a Bayer array (subpixel W in this embodiment is replaced with subpixel G). FIG. 6 shows the characteristics of other filters when sub-pixels having the same arrangement as in this embodiment are used. In the figure, the horizontal axis (f _x ) represents the frequency in the X direction, and the vertical axis (f _y ) represents the frequency in the Y direction. In the figure, a square indicated by a solid line represents a frequency band of an input image signal, and a hatched portion represents a band that the filter passes.

  As shown in FIG. 5, in the Bayer array, for the sub-pixel R and the sub-pixel B, the bands in both the X direction and the Y direction are generally limited to half of the low frequency side (that is, the low frequency region). It is. This is because the sub-pixel R and the sub-pixel B are arranged every other pixel in both the X direction and the Y direction, and an image can be expressed only at half the resolution of the input image signal. is there. On the other hand, since the filter for the sub pixel G includes two sub pixels G in one pixel, the band to be limited can be half that of the sub pixel R or the sub pixel B. Specifically, the filter applied to the G signal is configured to cut a region where the frequencies in both the X direction and the Y direction are high.

  FIG. 7 is a diagram illustrating a lattice formed by the sub-pixels in the Bayer array. As shown in the figure, the grid composed of the sub-pixels G is a square with one side shorter than the grid composed of the sub-pixels R (or sub-pixels B), and the sub-pixels R (or sub-pixels B). ) With a shape inclined by 45 ° with respect to the lattice constituted by Further, when comparing the lengths of the sides of each lattice between the case of the sub-pixel G and the case of the sub-pixel R (or sub-pixel B), the former is the latter √2 / 2 (the square root of 2 is divided by 2). Number) times. Therefore, in the Bayer array, display with higher resolution is possible with the G signal, and the pass band of the G signal can be wider than the pass band of the R signal or the B signal.

  On the other hand, in this embodiment, as shown in FIG. 4, the filter applied to the G signal and the W signal is the same as the filter applied to the G signal in the Bayer array of FIG. The reason for performing such band limitation is that the display by the sub-pixel W includes the G component. In the present embodiment, the sub-pixel W is regarded as the sub-pixel G, and filters to be applied to the G signal and the W signal are determined.

  Similarly, the display by the sub-pixel W includes an R component and a B component. In the present embodiment, the characteristics of the filter applied to the R signal and the B signal are also determined using this fact. Specifically, when the sub-pixel W is regarded as the sub-pixel R, the sub-pixel R is arranged without a gap (without sandwiching other sub-pixels) in the X direction, and the input image signal Since it can be reproduced with the same resolution, the filter is configured not to limit the band in the direction. On the other hand, even when the sub-pixel W is regarded as the sub-pixel R, the sub-pixel R is arranged every other pixel in the Y direction, so that the band in the direction is limited as in the case of FIG. . That is, the filter applied to the R signal is configured to pass the low frequency region and not pass the high frequency region in the Y direction.

  Note that the arrangement of the sub-pixel B when the sub-pixel W is regarded as the sub-pixel B is the same as that obtained by rotating the arrangement of the sub-pixel R when the sub-pixel W is regarded as the sub-pixel R by 90 °. For this reason, the filter applied to the B signal in this embodiment has a characteristic in which the X direction and the Y direction of the filter characteristic applied to the R signal are interchanged.

  As described above, each color filter of the present embodiment limits only the band in the Y direction for the R signal in a manner corresponding to the interval of the sub-pixels R, and only the band in the X direction for the B signal. Limit in a manner according to the interval. In addition, the filter for the G signal and the W signal limits the band of each signal in a manner corresponding to the arrangement of the sub pixels when both the sub pixel G and the sub pixel W are identified and combined. In other words, in the present embodiment, the frequency band of the image signal of each color is set by utilizing the fact that the sub-pixel W includes an element having the same optical property as any of the sub-pixel R, the sub-pixel G, and the sub-pixel B. Thus, it is possible to ensure a wider area than in the comparative examples of FIGS.

  When the filter of the present embodiment is compared with the filter of FIG. 6, there is a difference in the bandwidth limit. Specifically, since the R signal and the B signal do not require any restriction in the X direction or the Y direction by regarding the sub pixel W as the sub pixel R or the sub pixel B, the passing bandwidth is 2 Double. In addition, since the G signal and the W signal do not have to be treated as different sub-pixels by regarding the sub-pixel W as the sub-pixel G, the same band limitation as in the Bayer array in FIG. 5 is used. Can do.

  Therefore, according to the display system 10 of the present embodiment, the band limit width is smaller than that in the case where band limitation is performed independently for each color image signal (that is, without considering the positions of sub-pixels of other colors). Can be reduced, and the luminance of the projected image can be improved as compared with the case where the sub-pixel G is used instead of the sub-pixel W. Therefore, according to the display system 10, it is possible to improve the visual image quality compared to the case of any of the comparative examples in FIGS.

  On the other hand, a new problem arises by applying the band limitation as in the present embodiment. The problem is the generation of moiré due to the difference in brightness of each color of R, G, B, and W. The principle that moire and false colors occur in the configuration of this embodiment is specifically as follows.

  The filter of the present embodiment is designed based on a specific color. When the filter according to the present embodiment is designed so that moire does not occur in gray, which is an achromatic color, moire is generated in at least one of R, G, and B in other colors (in this case, chromatic colors). Occurs. More specifically, moire occurs in the configuration of the present embodiment because the color component of the color signal of any one of R, G, and B and the color component of any one of the colors of the W signal (that is, white) This is a case where the color components (when the light is split into three colors of R, G, and B) are not equal, and the brightness is different.

  FIG. 8 is a diagram for explaining moire generated in the present embodiment. Here, a case where attention is paid to the R component is shown. In the figure, a broken straight line represents the position of each of the R and W subpixels, and a sine wave represented by a solid line represents an image signal of a certain high frequency component. A plot (point) on the sine wave represents the brightness displayed by each sub-pixel.

  The example illustrated in FIG. 8A illustrates a case where the sub pixel R and the sub pixel W have the same brightness. In this case, when attention is paid only to the display by the sub-pixel R, the brightness changes like a mountain as shown by the alternate long and short dash line, so that moire occurs. However, if attention is paid only to the display by the sub-pixel W in this case, the brightness changes so that a moire (that is, a moire having a phase shift) opposite to that generated by the sub-pixel R is generated. Therefore, when these sub-pixels are observed at the same time, the moires cancel each other, and as a result, the moire becomes invisible.

  On the other hand, the example shown in FIG. 8B shows a case where the sub pixel R and the sub pixel W do not have the same brightness, and the sub pixel W becomes brighter than the sub pixel R. Even in this case, the moire due to the sub-pixel R acts to reduce the moire due to the sub-pixel W (that is, to make the moire inconspicuous), but the moire cannot be completely canceled, resulting in a moire. The moire generated in this manner is more easily perceived by the human eye as the difference in brightness between the subpixel R and the subpixel W (difference in high-frequency component) increases.

  The gain adjustment parameter of the present embodiment acts to reduce the moiré that occurs in this way. In other words, the filter processing unit 124 adjusts the amplitude of the high-frequency component of the other color signal according to the gain adjustment parameter according to the difference between the high-frequency component of the W signal and the high-frequency component of the other color signal, so It is possible to reduce the difference in brightness that occurs between the sub-pixels.

FIG. 9 is a diagram showing the relationship between the gain adjustment parameter of each color and the frequency response of the filter. As shown in the figure, the filter processing unit 124R has a frequency response of the X-direction of the high-frequency component is varied by the gain adjustment parameter G _r. For example, if G _r = 1, the filter processing unit 124R outputs the high frequency component without changing the amplitude of the high frequency component, while if G _r = 0, the filter processing unit 124R sets the amplitude of the high frequency component to zero. That is, when G _r = 0, the filter processing unit 124R functions as a low-pass filter that passes low-frequency components in the X direction and cuts high-frequency components. Further, when 0 <G _r <1, the filter processing unit 124R attenuates the amplitude of the high frequency component according to the magnitude of the gain adjustment parameter G _r . Note that the filter processing unit 124R does not change the frequency response with the gain adjustment parameter G _{r in} the Y direction. This is because the high frequency component in the Y direction is cut from the R signal.

On the other hand, the filter processing unit 124B is the frequency response of the Y-direction of the high-frequency component is changed by the gain adjustment parameter G _b, not the frequency response of the X-direction of the frequency components is changed by the gain adjustment parameter G _r. Further, filtering section 124G is, X-direction, the frequency response of the high-frequency component of the Y-direction both varied by the gain adjustment parameter G _b.

  As described above, according to the display system 10 of the present embodiment, when color display is performed with four color sub-pixels, it is possible to narrow the bandwidth limit and to generate moiré in a specific color. In addition to preventing the moiré, it is possible to suppress moire even in colors other than the specific color. Therefore, according to the display system 10, it is possible to improve the visual image quality compared with the case where the gain of the high frequency component is not adjusted using the gain adjustment parameter.

[Second Embodiment]
This embodiment has the same configuration as that of the first embodiment described above except that the filter processing for the image signal is different. Therefore, in the present embodiment, the same reference numerals used in the first embodiment are used for the same components as those in the first embodiment, and overlapping descriptions are omitted as appropriate.

  This embodiment improves the so-called resolution and further improves the visual image quality by suppressing moire and false colors that may occur in the configuration of the first embodiment. The principle of generating moire and false colors in the configuration of the first embodiment is specifically as follows. Note that the moire of interest in the present embodiment is generated for a reason different from the moire reason illustrated with reference to FIG.

  The filtered R signal or B signal has lost information on the frequency band not restricted in the W signal, or contains information on the frequency band restricted in the W signal, etc. The frequency band included as information is different from the W signal. The information of the frequency band that is included in the W signal and not included in the R signal or the B signal can be regarded as extra information from a certain point of view. Since the frequency band of such information exceeds the Nyquist frequency (1/2 of the sampling frequency), it becomes aliasing noise and causes moire. Here, the sampling frequency is determined based on the size of one pixel of the input image signal. Note that the G signal after the filter processing has the same frequency band as information as the W signal, and therefore moire due to the difference in the passing frequency band does not occur.

  In the configuration of the first embodiment, such moire occurs only in the R or B color component. That is, the moire caused by the difference in the passing frequency band is a colored moire. Therefore, in the configuration of the first embodiment, particularly when an image including a high-frequency component is to be displayed, a color different from the original color of the image, that is, a false color is present in a so-called edge (contour) portion or the like. Occur.

  FIG. 10 is a diagram for explaining a frequency band that causes moire. In the figure, a band that is not limited in the W signal but limited in the R signal or B signal is indicated by hatching, and a pass band by the filter (see FIG. 4) is indicated by a broken line.

  As shown in FIG. 10, when comparing the characteristics of the R filter and the W filter, there is a band that passes in the W signal and does not pass in the R signal in the high frequency band in the Y direction. On the other hand, comparing the characteristics of the B filter and the W filter, there is a band that passes in the W signal and does not pass in the B signal in the high frequency band in the X direction.

  Therefore, in the present embodiment, the frequency response is adjusted so as to cancel the influence of the band in at least one (preferably both) filters of R and B. Specifically, the filter processing units 124R and 124B perform the filtering process so that the phase of the band other than the band that the filter processing unit 124R passes through and the band that the filter processing unit 124W passes through is reversed, thereby Counter the effects of signals. Note that “offset” here does not only mean that they cancel each other out completely, but it means that the influence of the other is reduced by one.

FIG. 11 is a diagram for explaining the R and B filters of this embodiment and their frequency characteristics. In the R and B filters of this embodiment, the processing for the band B ₁ in the figure is the same as that of the first embodiment, and the processing for the band B ₂ is different from that of the first embodiment. The R filter of the present embodiment is different from the first embodiment in that the phase of the high frequency band (B ₂ ) in the Y direction is inverted and passed. Similarly, the B filter of this embodiment is different from that of the first embodiment in that the phase of the high frequency band (B ₂ ) in the X direction is inverted and passed. Here, the negative frequency response means that the image signal is output with the phase of the component of the band B ₂ inverted.

In addition, the filter according to the present embodiment also performs the adjustment using the gain adjustment parameter for the band in which the phase is inverted and passed. Specifically, filtering section 124R also perform the adjustment of the gain for the band B ₂ is a Y-direction of the high-frequency band as well X-direction, the filter processing unit 124B is X-direction high-frequency as well as the Y direction The gain adjustment is also performed for the band B ₂ which is the band. Note that the behavior when the gain adjustment parameter is 0 is common to the present embodiment and the first embodiment.

  According to the filter processing of the present embodiment, it is possible to suppress chromatic moire and false colors. In addition, the effect of suppressing moire and false colors becomes more prominent in achromatic images than in chromatic images (in the human visual sense, chromatic false colors produced in achromatic images are However, it is easier to perceive than the chromatic fake colors that appear in chromatic images.)

Note that the band for inverting the phase is not necessarily limited to the band illustrated in FIG. 11 because it is sufficient that the band to be passed by the filter processing unit 124W is included.
FIG. 12 is a diagram illustrating another example of the R and B filters of the present embodiment. The filter shown in the figure has a band B ₂ smaller than that of the filter shown in FIG. 11, but the band includes a band that is passed through by the filter processing unit 124W, and therefore has an effect of suppressing moire and false color. Play.

[Modification]
The present invention is not limited to the embodiment described above, and can be implemented in various forms exemplified below. In addition, the modification shown below may combine each suitably as needed.

(1) The pixel arrangement of the present invention may not be square as in the above-described embodiment. For example, when the sub-pixel is rectangular, the arrangement of the entire pixel is also rectangular. Further, the first direction and the second direction do not necessarily have to be orthogonal to each other, and it is sufficient if they have a crossing relationship. That is, the specific shape of the pixel of the present invention is not limited as long as four sub-pixels in two rows and two columns are arranged in a quadrilateral shape.

  Further, the positional relationship between the sub-pixels in the pixel is not limited to that in the above-described embodiment. In the pixel of the present invention, if a sub-pixel (sub-pixel W in the embodiment) that performs the same band limitation as the sub-pixel W is located in the diagonal direction of the quadrilateral when viewed from the sub-pixel W, the adjacent sub-pixel is implemented. It may be different from the form.

  FIG. 13 is a diagram illustrating another example of the arrangement of sub-pixels. The arrangement shown in FIG. 13A is obtained by replacing the positions of the sub-pixel R and the sub-pixel B from the arrangement of the above-described embodiment (see FIG. 2). In this case, the filtering process for the R signal is the same as the filtering process performed for the B signal in the embodiment, and the filtering process for the B signal is the same as the filtering process performed for the R signal in the embodiment. It should be. Further, the arrangement shown in FIG. 13B is obtained by replacing the positions of the sub-pixel G and the sub-pixel W from the arrangement of the above-described embodiment. In this case, the filtering process for the image signal of each color is the same as that of the embodiment. In addition, the arrangement of the sub-pixels may be an arrangement obtained by rotating each arrangement in a point-symmetric manner or a line-symmetric manner, as in the examples shown in FIG. 13 (c) and FIG. 13 (d).

  Furthermore, the positional relationship between the sub-pixels of each color may not be an arrangement in which the sub-pixel W and the sub-pixel G face each other in the diagonal direction. For example, when the display device displays an image with a strong redness or an image containing a lot of red, or when it is desired to display red more easily than other colors, the sub-pixel facing the sub-pixel W in the diagonal direction is set as a sub-pixel. The pixel R may be used. That is, the arrangement of the sub-pixels may be determined in consideration of the image displayed on the display device and the image quality required for the display device.

(2) The above-described embodiment is an example in the case where the relationship Y _W = Y _R + Y _G + Y _B is satisfied. However, an actual liquid crystal panel does not necessarily satisfy such a relationship, and Y _W may not be equal to the sum of Y _R , Y _G , and Y _B. In such a case, when the phase is inverted by filtering the R signal or the B signal, the frequency response is determined according to the actual luminance of the sub-pixel and balanced with the luminance of the sub-pixel of each color. It is also possible. That is, the filter processing by the filter processing unit 124 may be determined based on the characteristics of the liquid crystal panel 220 that is actually used. Therefore, the filter processing unit 124 may be configured such that the user can adjust the characteristics of the filter in consideration of the characteristics of the liquid crystal panel 220.

(3) In the embodiment described above, the image signal output from the filter processing unit 124 is thinned out by the thinning processing unit 125. However, the filter processing unit according to the present invention is configured to have the functions of the filter processing unit 124 and the thinning-out processing unit 125 of the above-described embodiment, and refers to an image signal of 4M × 4N pixels. It may be configured to output image signals for pixels. In this case, the gain adjustment parameter calculated by the parameter calculation unit 123 is sufficient for M × N pixels. Accordingly, the parameter calculation unit 123 in this case does not calculate the gain adjustment parameters for 4M × 4N pixels for each pixel of the input image signal, and the number corresponding to the sub-pixels of the image signal output by the filter processing unit. Only the gain adjustment parameter may be calculated.

(4) The present invention does not require the optical modulator to be configured with transmissive pixels as in the above-described embodiment, and has a pixel arrangement similar to that in the above-described embodiment. If present, the present invention can be applied to a display panel having a self-luminous pixel such as an organic EL (electroluminescence) display or a plasma display. Therefore, the display device according to the present invention is not limited to the projector. Further, even when the light modulator is a liquid crystal panel, this is not limited to the transmission type, but may be a reflection type.

  The image processing apparatus according to the present invention may be realized by an image processing circuit incorporated in a display device, or may be realized by software processing executed by a computer device such as a personal computer. Furthermore, the present invention can also be provided in the form of an image processing method for executing image processing corresponding to each of four colors, a program for causing a computer device to execute such image processing, and a recording medium on which such a program is recorded. is there.

DESCRIPTION OF SYMBOLS 10 ... Display system, 100 ... Image processing apparatus, 110 ... Input part, 120 ... Image processing part, 121 ... Color conversion processing part, 122 ... High frequency component extraction part, 123 ... Parameter calculation part, 124, 124R, 124G, 124B, 124W: Filter processing unit, 125 ... Thinning processing unit, 130 ... Output unit, 200 ... Projector, 210 ... Light source, 220 ... Liquid crystal panel, 230 ... Projection lens, 300 ... Screen

Claims (10)

Each pixel composed of four subpixels arranged in two in the first direction and in the second direction intersecting the first direction is arranged in a matrix, and one of the subpixels constituting one pixel Is an image processing device for supplying four color image signals including white to a display device corresponding to white,
A high-frequency component extraction unit that extracts a component in a high-frequency band that is on the high-frequency side from the center of the frequency band of the image signal, from each of the four color image signals;
Based on the components of each color extracted by the high-frequency component extraction unit, a parameter calculation unit that calculates parameters determined according to the ratio of each component to the white component for the three color components other than white,
For each of the four color image signals, the frequency band is limited according to the arrangement of the sub-pixels corresponding to the color, and for each of the three color image signals other than white, the high frequency band An image processing apparatus comprising: a filter processing unit that executes a filter process for adjusting an amplitude using a parameter calculated by the parameter calculation unit. The filter processing unit
For the white image signal for displaying the first sub-pixel corresponding to white, the frequency bands in the first direction and the second direction are set diagonally to the first sub-pixel and the first sub-pixel. A first filter processing unit that restricts in a manner according to the arrangement of the second sub-pixels positioned;
A second that limits a frequency band in the first direction and the second direction with respect to an image signal for displaying the second sub-pixel in a manner corresponding to an arrangement of the first sub-pixel and the second sub-pixel; A filter processing unit;
A mode in which a frequency band in the second direction is set according to an interval in the second direction of the third sub-pixel with respect to an image signal for displaying the first sub-pixel and a third sub-pixel adjacent in the first direction. A third filter processing unit restricted by
A mode in which a frequency band in the first direction is set according to an interval in the first direction of the fourth sub-pixel with respect to an image signal for displaying the first sub-pixel and a fourth sub-pixel adjacent in the second direction. The image processing apparatus according to claim 1, further comprising: a fourth filter processing unit restricted by At least one of the third filter processing unit and the fourth filter processing unit is
The frequency band of the input image signal is a band other than the band to be passed, and the first filter processing unit passes the band with the phase inverted, and the amplitude of the band is used as the parameter. The image processing apparatus according to claim 2, wherein the image processing apparatus is adjusted. The frequency response of the band that is passed with the phase inverted is determined according to the luminance of the first sub-pixel, the second sub-pixel, the third sub-pixel, and the fourth sub-pixel. Item 4. The image processing apparatus according to Item 3. The filter processing unit
Reference is made to an image signal representing 4M pixels in the first direction and 4N pixels in the second direction, and for each color, an image representing M subpixels in the first direction and N subpixels in the second direction. Output signal,
The parameter calculator is
5. The image processing apparatus according to claim 1, wherein the number of parameters corresponding to the sub-pixels represented by the image signal output from the filter processing unit is calculated. The filter processing unit
6. The filter processing is executed for each of three color image signals other than white by adjusting the output of a predetermined specific filter using the parameter. The image processing apparatus according to any one of the above. The filter processing unit
For each of the three color image signals other than white, the filter processing is performed by selecting a plurality of filters determined for each of the plurality of parameters according to the parameter calculated by the parameter calculation unit. The image processing apparatus according to claim 1, wherein the image processing apparatus is executed. The second filter processing unit performs filter processing on an image signal corresponding to green,
The image according to any one of claims 1 to 7, wherein the third filter processing unit and the fourth filter processing unit each perform a filter process on an image signal corresponding to red or blue. Processing equipment. A display panel including a plurality of pixels arranged in a matrix, the first sub pixel, the second sub pixel, the third sub pixel, and the fourth sub pixel;
A display device comprising: the image processing device according to claim 1. Each pixel composed of four subpixels arranged in two in the first direction and in the second direction intersecting the first direction is arranged in a matrix, and one of the subpixels constituting one pixel Is an image processing method for supplying four color image signals including white to a display device corresponding to white,
From each of the four color image signals, extract a component of a high frequency band that is on the high frequency side from the center of the frequency band of the image signal,
Based on the extracted components of each color, for three color components other than white, a parameter determined according to the ratio of each component to the white component is calculated,
For each of the four color image signals, the frequency band is limited according to the arrangement of the sub-pixels corresponding to the color, and for each of the three color image signals other than white, the high frequency band An image processing method comprising: performing a filter process for adjusting an amplitude using the calculated parameter.

Images