KR101788681B1 - Color correction to compensate for displays' luminance and chrominance transfer characteristics - Google Patents

Abstract

Display devices are provided with circuitry for performing color correction to compensate for the luminance and chrominance transition characteristics of the display devices. Certain techniques are suitable for RGBW display devices and subpixel-rendered display devices. Certain display devices include an external light source (e.g., a backlight unit of an LDC device), and the color correction is adjusted by dynamic control of the light source.

Description

{COLOR CORRECTION TO COMPENSATE FOR DISPLAYS 'LUMINANCE AND CHROMINANCE TRANSFER CHARACTERISTICS}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to image processing, and more particularly, to color correction for image display.

In digital image processing, an image is typically represented by the number of pixels. The color of each pixel is defined, for example, by the coordinates of the color in a color space, such as sRGB. The display device converts the color coordinates into "gradation levels ", which are used to define electrical signals (e.g., voltages) that determine the brightness of corresponding areas on the screen of the display device. (Sometimes, the color coordinates may be used as the gray level levels themselves.) The brightness displayed in the screen area is a function of the corresponding gray level (the "gamma function "," gamma transfer function & Characteristic, " or "gamma curve").

For many display devices, the gamma function is non-linear and can be approximated by a power function relationship.

L = v ^? (1)

Here, L is luminance, v is a gradation level, Is a constant relating to the display. And 2.2 for various types of devices such as CRT (cathode ray tube) and LCD (liquid crystal display).

In the various display devices, the relation (1) above is only an approximation, and can be modified using look-up tables (LUTs) which are stored in the display and include experimental data obtained for the display device.

A typical color representation relates to a combination of primary colors including, for example, red, green and blue. The display device uses the graded levels of gradation for each of the primary colors (e.g., for each "channel"). For each different channel, the gamma functions may be different and thus separate LUTs are provided for each channel.

For example, a color LCD may include a myriad of red, green, and blue subpixels. The subpixels may contain the same liquid crystal cells, but have color filters of different colors (red, green and blue). However, the liquid crystal cells have different optical characteristics with respect to the hue, depending on the wavelength. Depending on the bandwidth of the spectrum of the color filters, this optical characteristic causes non-identical luminance gamma transfer characteristics between the red, green and blue channels. In addition, the optical characteristics cause color difference deviations (e.g., tone deviations) within each channel. The non-identical gamma transfer characteristics may be modified using the separate look up tables (LUTs) for each channel. In the RGB LCD, the color difference deviations are not recognized as a serious problem because the color filters have a narrow spectral bandwidth.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and it is an object of the present invention to provide an image data processing method capable of accurately displaying hue and color with respect to transition characteristics such as luminance and chrominance.

According to another aspect of the present invention, there is provided an image data processing apparatus for performing the image data processing method.

According to another aspect of the present invention, there is provided an image data processing method for displaying an image using a plurality of primary colors including a plurality of first primary colors and at least one second primary colors, Obtaining a color signal representing one or more colors and representing an amount of each of the primary colors in one or more colors, merging a first calibration of the color signal to correct for a transition characteristic of the display device, Obtaining a first signal dependent on the amount of each of the first primary colors, merging a second calibration signal of the color signal to be calibrated with respect to a transition characteristic of the display device, Obtaining a second signal that is independent of the amount of the at least one second primary colors but is dependent on the amount of the at least one second primary colors, The method comprising the steps of: obtaining a relative mass signal indicative of a relative mass of at least one or more second primary colors associated therewith; and acquiring the first and second signals and the first and second calibrations, And obtaining a combined calibration signal indicative of the combined calibration.

According to an embodiment of the present invention, the first signal may be independent of the amount of one or more second primary colors in the representation.

According to one embodiment of the present invention, the dependency on the relative amount is such that if the combined calibration is equal to the value between the first calibration, the second calibration or the first and second calibrations and the relative amount is small The combined calibration may be close to the first calibration.

According to one embodiment of the present invention, the combined calibration may be a result of linear interpolation of the first and second calibrations.

According to one embodiment of the present invention, the relative amount is an increasing function of the amount of one or more of the second primary colors, and may be a decreasing function of at least one of the first primary colors.

According to an embodiment of the present invention, the amount of each color may be a value of color coordinates in a linear color space.

According to an embodiment of the present invention, the method may further comprise using the combined calibration for the image display.

According to one embodiment of the invention, the combined calibration signal may define the combined calibration as one or more tone values in a non-linear color space.

According to an embodiment of the present invention, at least some calibration of each second primary color that calibrates the transition characteristics of the display device may be performed by calibrating two or more of the first primary colors.

According to another aspect of the present invention, there is provided an image display apparatus for displaying a color signal including at least one color included in an image and including a color coordinate value for each color, Includes calibration circuit. Wherein the color coordinates are associated with a plurality of primary colors, wherein the primary colors comprise a plurality of first primary colors and one or more second primary colors, and wherein the color coordinates of each color are associated with first coordinates And second coordinates associated with the one or more second primary colors. The circuit includes a first circuit, a second circuit, and an interpolator. The first circuit merges a first calibration of the color signal to be calibrated for the transition characteristics of the display device and is dependent on the values of one or more of the first coordinates in the color signal. Said second circuit merging a second calibration of said color signal to be calibrated with respect to the transition characteristics of said display device and wherein said second circuit is independent of values of one or more said first coordinates in said color signal, And generates a second signal that is dependent on the values of the two coordinates. The interpolator interpolates the first and second signals based on an indication of the relative amount of at least one of the one or more of the second primary colors associated with the first primary colors to generate a combined calibration signal .

According to an embodiment of the present invention, the first signal may be independent of the value of one or more second coordinates in the color signal.

According to an embodiment of the invention, the interpolation may be linear.

According to one embodiment of the present invention, the relative amount is an increasing function for one or more of the one or more of the second coordinates, and may be a decreasing function for one or more values of the first coordinates.

According to one embodiment of the present invention, the combined calibration signal may define the combined calibration as one or more tone values in a non-linear color space.

According to an embodiment of the present invention, the color coordinates for each color may be located in a linear color space.

According to an embodiment of the present invention, each of the second primary colors may be represented by a mixture of two or more of the first primary colors.

According to another aspect of the present invention, there is provided an image display apparatus for displaying an image using a plurality of primary colors including a plurality of first primary colors and at least one second primary colors, And circuitry for expressing one or more colors in the image and calibrating a color signal comprising values of one or more color coordinates of each color. The circuit includes a first data path for processing the hue signal and a second data path for processing the hue signal. The first and second data paths comprise a shared circuit. Wherein the first data path is a separation that corrects each value of the color coordinates of the plurality of first sets of color coordinates, wherein the first data path comprises at least one but less than all of the color coordinates of each color for the transition characteristics of the display device Wherein the separate color correction circuits of the first data path perform a correction of the color signal with respect to a transition characteristic of the display device, Represents one of the primary colors. Wherein the second data path is a separation that corrects each value of the color coordinates of the plurality of second sets of color coordinates, wherein the second data path comprises at least one but less than all of the color coordinates of each color for the transition characteristics of the display device Wherein the separate color correction circuits of the second data path perform a correction of the color signal with respect to a transition characteristic of the display device, Represents one of the two primary colors. Wherein the sharing circuit comprises an interpolation circuit for interpolating the signal obtained using the first color correction circuit and the suppression signal using the second color correction circuit to produce a combined calibration signal, Based on an indication of the relative amount of at least one of the one or more second primary colors associated with at least one of the first primary colors included in the colors.

According to an embodiment of the present invention, the first and second sets may be composed of one color coordinate.

According to one embodiment of the present invention, the sharing circuit includes a third calibration circuit connected to the lower portion of the interpolation circuit, and the sharing circuit can calibrate the amount of the primary colors with respect to the transition characteristics of the display device .

According to an embodiment of the present invention, the shared circuit may further comprise a subpixel rendering circuit coupled between the interpolator and the third calibration circuit.

According to an embodiment of the present invention, the calibration circuit may further include a control circuit connected between the interpolation circuit and the third calibration circuit to determine an output power of a light source used to display an image, Wherein the output power indicative of the output power is determined dynamically by one or more images displayed by the imaging device and the light source is selected from the group of signals generated by the first and second data paths Lt; RTI ID = 0.0 > subpixels < / RTI >

According to an embodiment of the present invention, the sharing circuit may further include a subpixel rendering circuit connected to a lower portion of the interpolation circuit.

According to an embodiment of the present invention, the sharing circuit includes a control circuit connected to a lower portion of the interpolator to determine an output power of a light source used to display an image, Wherein the light source is dynamically determined by one or more images displayed by the imaging device and the light source is optically coupled to subpixels generated by the first and second data paths and controlled by a signal dependent on the output power, Can be provided.

According to another aspect of the present invention, there is provided a method of processing image data processing for displaying images on a screen unit of a display device including a plurality of subpixels, Pixel has a luminance dependent on the state of a subpixel defined using a subpixel value and the images are displayed using a light source that provides light to the screen unit. The image data processing method includes (1) obtaining image signals representing respective images using respective color coordinates, and (2) generating the sub pixel values from the image signals. (2) generating subpixel values for at least one image IM1 of the images comprises: (2A) generating information about the output power generated by the light source and displaying the image IM1 Generating a scaling-information signal that is dependent on at least one of the images, the scaling information including scaling information including a definition of how the image signals are scaled corresponding to the output power of the light source; (2B) Scaling the color coordinate values of the image IM1 to produce a scaled image signal, and (2C) using at least one of the sub-pixel values to express the color correction of the scaled image signal of the image IM1 And obtaining a color-correcting signal.

According to an embodiment of the present invention, the scaling information included in the scaling-information signal for the image IM1 may be obtained from at least a partial representation of the color correction for at least one image.

According to an embodiment of the present invention, the scaling information for each image IM1 may be dependent on the values of the color coordinates of one or more images before the image IM1.

According to an embodiment of the present invention, the scaling information for each image IM1 may be independent of the values of the color coordinates of the image IM1.

According to an embodiment of the present invention, the scaling information for each image IM1 may depend on the color correction of one or more images prior to the image IM1.

According to an embodiment of the present invention, the scaling information for each image IM1 may be independent of the color correction for the image IM1.

According to an embodiment of the present invention, generating the scaling-information signal for each image IM1 comprises generating an unscaled signal of at least a portion of the adjusted color correction for the at least one image . Wherein the adjusted color calibration is a calibration for a transition characteristic of the display and is the color calibration adjusted for an output power level that is independent of the scaling information for the at least one image, The scaling information of the image of the at least one image and at least a portion of the color correction of the at least one image.

According to an embodiment of the present invention, the method further comprises generating a subpixel-value signal representing at least one of the subpixel values for the image IM1 from the color correction signal, The subpixel rendering process may be performed using the color correction signal.

According to another aspect of the present invention, there is provided an apparatus for processing image data including a plurality of subpixels, the apparatus comprising: Pixel has a luminance value dependent on the state of the subpixel defined using the corresponding subpixel value, and the images are displayed using a light source that provides light to the screen unit. The image data processing apparatus includes a circuit for obtaining an image signal representing an image using color coordinates and generating the subpixel values from the image signals. (A) a method for displaying, for the images, the information about the output power generated by the light source in displaying the images, and scaling the image signals representing the image to match the output power of the light source Light-source-control DLSC circuit for determining scaling information that includes information on the color of the image, (B) correcting the color coordinates of the image according to the scaling information provided by the DLSC, A scaling circuit for generating a scaled value of the coordinates; and (C) a color-correcting circuit for calibrating the scaled value and generating the calibrated scaling value of the color coordinates used to provide the sub-pixel values.

According to one embodiment of the present invention, the DLSC circuit may generate the scaling information in response to the color-correcting circuit based on at least a portion of the information on the calibrated scaling value.

According to one embodiment of the present invention, in displaying each image IM1 for one or more images, the scaling information may be dependent on the calibrated scaling values of one or more images prior to the IM1.

According to an embodiment of the present invention, for each image IM1, the scaling information may be independent of the color correction of the image IM1.

According to one embodiment of the invention, the DLSC circuit comprises an un-scale circuit for determining the adjusted values of the color coordinates, the adjusted values being corrected for the transition characteristics of the display device And adjusted for an output power level that is independent of the scaling information, and may be determined from at least a portion of the color correction determined by the scaling information and the color-correcting circuit.

According to an embodiment of the present invention, the apparatus may further include a subpixel rendering circuit that performs subpixel rendering on the data including the corrected scaling value.

According to an embodiment of the present invention, for each of the images IM1, the scaling information may be independent of the values of the color coordinates of the image IM1.

According to another aspect of the present invention, there is provided a method of processing image data for displaying an image, the method comprising the steps of: calculating values of color coordinates of one or more colors included in the image, And obtaining a color signal containing the color signal. Obtaining, for each color, a calibrated-color signal comprising calibrated values from the color signal, the calibrated values merging at least one of the values of the color coordinates of the color coordinates with respect to the characteristic before the display Generating an adjustment signal from the calibrated-color signal, the adjustment signal merging information about the adjustment performed on the calibrated values of the hue; and modifying the calibrated-color signal with the adjustment signal to adjust the color Calibrated color signal comprising an adjustment of the color coordinates and a calibrated value of the color signal. Wherein the color coordinates include at least one color coordinate where the calibrated and calibrated values are greater than the calibrated values and at least one color coordinate whose calibrated and calibrated values are less than the calibrated values.

According to an embodiment of the present invention, each color coordinate is associated with a respective primary color, and the primary colors include a plurality of first primary colors and one or more second primary colors, Can be expressed as a mixture of first primary colors. In addition, the color coordinates may include first color coordinates associated with the first primary colors and one or more second color coordinates associated with the one or more second primary colors. Wherein the color coordinates include at least one first color coordinate whose adjusted and calibrated values are greater than the calibrated values and at least one second color coordinate whose calibrated and calibrated values are less than the calibrated values, The color coordinates may include at least one of the second color coordinates for which the adjusted and calibrated values are greater than the calibrated values and at least one of the first color coordinates for which the calibrated and calibrated values are less than the calibrated values.

According to an embodiment of the present invention, the method may further include performing a subpixel rendering process on the adjustment-corrected color signal.

According to an embodiment of the present invention, modifying the calibrated-color signal with the adjustment signal may further comprise clamping the adjusted and calibrated values within a possible range.

According to an embodiment of the present invention, for each color, the adjusted and calibrated values may represent a metamer of the calibrated values.

There is provided an apparatus for processing image data for displaying an image on a screen unit of a display device including a plurality of subpixels according to an embodiment of the present invention, And circuitry for processing a color signal comprising values of one or more color coordinates contained in the image. The circuit includes a color correction circuit and an adjustment circuit. The color correction circuit obtains, for each color, a calibrated value of the color coordinates from which the one or more of the color coordinate values of the color are merged from the value of the color coordinates for the transition characteristic of the display. The adjustment circuit determines whether to adjust the calibrated value and to modify the calibrated value by performing the adjustment to determine whether to generate calibrated and calibrated values of the color coordinates of the hue. Wherein the step of processing the image data of the image data processing apparatus includes the steps of modifying the calibrated value by increasing at least one of the calibrated values of the color coordinates and decreasing at least one of the calibrated values of the other color coordinates .

According to an embodiment of the present invention, the method may further include performing a subpixel rendering process on the adjustment-corrected value.

According to another aspect of the present invention, there is provided a method of processing image data to display one or more images on a screen unit of a display device including a plurality of subpixels, Pixel has a luminance value dependent on the state of the subpixel defined using the corresponding subpixel value, and the images are displayed using a light source that provides light to the screen unit. Wherein the image data processing method for each of at least one of the images IM1 includes obtaining an image signal including a value of one or more color coordinates of a color included in the image IM1, The method comprising the steps of: obtaining an adjusted image signal representing the image IM1 using adjusted values of the image IM1, information about the output power generated by the light source in displaying the image IM1, Generating a scaling information signal including information on a method of scaling video signals to match the output power of the light source, the scaling information signal including scaling information dependent on at least one of the images, The color coordinates of IM1 are adjusted so that the scaled value of the color coordinates is And generating a color-corrected signal from the scaled video signal, the color-corrected signal representing a color correction for the scaled video signal for the image IM1 and correcting according to a transition characteristic of the display do. The color-correcting signal is used to provide at least one of the sub-pixel values. Wherein for at least one color of the image IM1, the color coordinates are determined by at least one color coordinate whose adjusted value is greater than the calibrated value by a first amount, and a second amount At least one other color coordinate less than the calibrated value, the first and second quantities being dependent on the scaling information for at least one image.

According to one embodiment of the present invention, for at least one color included in the image IM1, the adjusted values represent a metamer of the color, and the values of the color coordinates are represented by another metamer metamer. < / RTI >

According to one embodiment of the present invention, the color-correcting signal may comprise a calibrated and scaled value of the color coordinates. The adjustment process may also include obtaining an approximate indicator of the amount of change of at least one calibrated and scaled value associated with the corresponding scaled value and adjusting a corresponding value in the video signal. If the calibrated and scaled value is greater than the corresponding scaling value then the corresponding calibrated value is less than a corresponding value in the video signal and if the calibrated and scaled value is less than the corresponding scaling value And making the corresponding calibrated value greater than a corresponding value in the video signal.

According to another aspect of the present invention, there is provided an apparatus for processing image data including a plurality of subpixels, the method comprising the steps of: Pixel has a luminance value dependent on the state of the subpixel defined using the corresponding subpixel value, and the images are displayed using a light source that provides light to the screen unit. Wherein the image data processing device comprises: an adjustment circuit for obtaining values of one or more color coordinates contained in the images, and adjusting the values of one or more of the colors to obtain an adjusted value of the color coordinates; Information about an output power generated by the light source and information about a method of scaling the image signals representing the images to match the output power of the light source and generating scaling information that is dependent on one or more of the images Circuit, circuitry for scaling the adjusted value of the color coordinates of the hues according to the scaling information to produce a scaled value of the color coordinates, and circuitry for performing color correction on the scaled values of the images from the scaled values And the like. Wherein the color calibration is a calibration of the transition characteristics of the display and is used to provide at least one of the sub-pixel values and for at least one color included in the at least one image, At least one color coordinate greater than the calibrated value and at least one other color coordinate wherein the calibrated value is less than the calibrated value by the second quantity corresponding to the first quantity. The first and second quantities are dependent on the scaling information for at least one image.

According to an embodiment of the present invention, for at least one color contained in at least one image, the adjusted values represent a metamer of the color, metamer. < / RTI >

According to one embodiment of the present invention, the color correction comprises a calibrated and scaled value of the color coordinates, and the calibration circuit is operable to calculate a correction value of at least one calibrated and scaled value associated with the corresponding scaled value An approximated index can be obtained. Further, a corresponding value in the video signal is adjusted. If the calibrated and scaled value is greater than the corresponding scaling value then the corresponding calibrated value is less than a corresponding value in the video signal and if the calibrated and scaled value is less than the corresponding scaling value , The corresponding calibrated value may be greater than a corresponding value in the video signal.

In another aspect of the present invention, there is provided a method of processing image data including P pixels according to an embodiment of the present invention, wherein P is an integer of 1 or more, and the image data specifies a color of each pixel . Wherein the image is displayed on the plurality of subpixels comprising at least one subpixel for each of a plurality of primary colors and wherein the plurality of subpixels comprises a subpixel of the at least one primary color, . Wherein each subpixel emits one of the plurality of primary colors and has a dependent luminance of a state of the subpixel defined using the subpixel value of the subpixel, And generating a sub-pixel value for displaying the image. Wherein the step of generating the subpixel value comprises the steps of obtaining a color-coordinates signal representing the color coordinates of at least one pixel, merging from the color coordinate signal a calibration of a transition characteristic of the display in which the image is displayed, Obtaining a signal of calibrated coordinates representing calibrated color coordinates of at least one pixel, and performing a subpixel rendering process on the signal of the calibrated coordinates to generate the subpixel values. The color coordinates of the pixel include color coordinates for each of the primary colors.

According to another aspect of the present invention, there is provided an apparatus for processing image data of an image according to an embodiment of the present invention, wherein the image data specifies a color of each pixel and the image includes at least one Wherein the plurality of subpixels comprises the at least one primary color subpixel that is smaller than the number of pixels. Each of the subpixels emits one of the plurality of primary colors and has a dependent luminance of the state of the subpixel defined using the subpixel value of the subpixel. The image data processing apparatus includes a circuit for generating a sub-pixel value for displaying the image for each sub-pixel. The circuit comprising: circuitry for receiving the color coordinates of each pixel comprising color coordinates for each of the primary colors, circuitry for generating calibrated color coordinates of the pixel from the color coordinates, And a circuit for performing a pixel rendering process. The calibrated color coordinates of the pixels merge the calibration of the transition characteristics of the display in which the image is displayed.

It is hereinafter referred to as "automatic color correction" or ACC in which color correction is performed to adjust the color data for the transition characteristics of the display device. As used herein, the term "transition characteristic" is a term that generally includes both gamma transfer characteristics (i.e., luminance characteristics) and color difference characteristics (e.g., chrominance shifts of white subpixels).

In RGBW indications, the white subpixels have a high spectral bandwidth, and the deviation of the chrominance within the white channel is important. The deviation of the chrominance within the white channel is determined by Hamer et al. Can be compensated for through the brightness adjustment of the red, green and blue gradation levels as described in U.S. Patent Application Publication No. 2008/0252797 A1, filed October 16, 2008, and incorporated herein by reference. The above patent application describes a white color difference correction for an organic light emitting diode (OLED) display device and mentions that the invention of this application can also be applied to an LCD. The OLED display includes red, green, blue, and white phosphors. First, an experiment is performed to measure white color difference deviations. Especially. The white luminous bodies were turned on at different intensities, and for each intensity, the hue corresponding to the actual representation was measured and recorded in linear RGB coordinates. In this method, three curves can be obtained by R (W), G (W) and B (W) which provide linear R, G and B coordinates for each W value. The three curves and the corresponding inverse functions W (R), W (G) and W (B) are written in a table and stored in the display device.

The display receives image data in some color space, such as sRGB, and converts this data to linear RGB. Then, the linear RGB data is converted into RGBW in a specific method to correct the white color difference. In particular, let us denote the linear RGB coordinates of a pixel by Rp, Gp and Bp. The display calculates the corresponding RGBW coordinates Rw, Gw, Bw, and Ww as follows. The white coordinate Ww may be set to a minimum value of W (R), W (G) and W (B) or a specific number below the minimum value. (The specific number below the minimum value can be used to increase the spatial uniformity of luminance in displaying colors close to white). The Rw, Gw and Bw coordinates are calculated from the corresponding RGB coordinates by R (W) G (W) and B (W). That is, it is as follows.

Rw = Rp-R (Ww)

Gw = Gp-G (Ww)

Bw = Bp-B (Bw)

OLED displays are also described by Lee et al. Quot ;, U.S. Patent Application Publication 2006/0262053 A1, published November 23, 2006.

In some embodiments of the present invention, the ACC is performed by selecting an appropriate calibration of the color coordinates for each combination of values of the color coordinates (such as RGBW). For example, the preferred tables may be generated by providing corrected coordinates for each combination of the color coordinate values. However, the tables generated by this method can be very large. If the circuit of the computer is used instead of the tables, the circuit can be very complex. Thus, in some embodiments, the following techniques are used. In some RGBW embodiments, it is calculated using the white color difference corrected RGB coordinates. Conventional tables may also be used to calibrate the brightness of the red, green, blue and white colors. If the input color is dark gray, the white color difference calibration table and the white luminance calibration table can be used to accurately display the color. If the color is fully saturated, the white coordinate is zero, and the red, green, and blue luminance tables cause the saturated color to be accurately displayed. If the input color is not white and a fully saturated color, then (i) interpolates between the white color defined by the white coordinate and (ii) the fully saturated color defined by the RGB coordinate.

Further, in the RGBW indications, the present invention has observed that when the display device uses a dynamic control light source, the color correction provides a special challenge. For example, since the LCD subpixels do not emit light, the LCD may include a backlight panel that includes a separate light source (110 of FIG. 1) that illuminates subpixels in the screen unit 120. The light source 110 is disposed on the back surface of the sub-pixels. In the LCD, the screen unit 120 is called an LCD panel and includes a screen driver 124 for driving the sub pixels (not shown) in the screen 126. The light source 110 may be dynamically controlled (i.e., controlled based on the displayed image) to reduce power consumption. If the image is dark, the light source is driven with low output power, and the sub-pixels compensate for the low output of the light source by increasing the transmittance. Increasing the transmittance of the subpixels can be achieved by increasing the subpixel values (gradation levels). For example, if the sRGB input data is converted to a linear RGBW representation (by block 130 in FIG. 1), then the power of the light source is reduced to a specific constant BL (i. E., Multiplied by BL < And multiplies the RGBW values by the reciprocal of the BL (indicated by INVy and determined at block 140). In FIG. 1, the multiplication is performed in a scaling block 150. The scaled RGBW values are converted into gradation levels (denoted as Rg, Gg, Bg and Wg for corresponding red, green, blue and white subpixels) in an "output gamma" Dynamic control of the light source is sometimes referred to as dynamic backlight control or DBLC.

Some embodiments of the present invention calibrate the white chrominance and the luminance of RGB while performing the RGBW transform 130. However, the calibration may not be correct for the scaled RGBW values (output of block 150). Therefore, display quality degradation may be high. Other embodiments perform the color correction on the scaled data (output of block 150). However, once the scaled data is color corrected, the scaling constants BL and INVy may not be optimized due to unnecessarily high power consumption and / or brightness distortion. In some embodiments, various techniques are used to prevent degradation of display quality and to consume less power. For example, in some embodiments, the interpolation technique mentioned above may be used to interpolate to output gamma 160.

Another challenge is in the subpixel-rendering (SPR) representation regardless of whether the display uses an external light source. In the non-SPR representation, each image pixel is mapped to a distinct combination of subpixels capable of displaying any primary color. In the SPR display, in order to reduce the number of subpixels, a pixel is mapped to a subpixel region where one or more primary colors are lacking. In addition, a subpixel of a certain primary color can be shared with several pixels. In this case, the value of the subpixel may be determined through information on several pixels in the subpixel rendering (SPR) process. In the present invention, it has been found that some ACC techniques have a better effect when applying the subpixel values generated by the SPR process. In particular, some ACC techniques have a better effect when applied to the RGBW data prior to the SPR process. The SPR process is very surprising since it can change the RGBW data to invalidate the color corrections performed before the ACC.

Some embodiments include a novel method of generating the ACC tables to reduce the size of the table and increase the maximum possible image brightness for many images. Increased brightness is very useful for portable and battery operated devices such as cellular phones. However, the present invention is not limited to these devices.

This summary is a brief introduction to some of the features of the present invention. The present invention is not limited to RGBW display devices. However, it may include display devices using other primary colors such as, for example, yellow, magenta, and cyan. The invention is not limited to the features and advantages described above except as defined by the appended claims.

According to the present invention, the hue and chrominance characteristics of a display device can be accurately displayed by adjusting and correcting hues included in an image in the color space.

1 is a block diagram of a display device according to the prior art.
2 is a block diagram of a display device according to the first embodiment of the present invention.
3 is a block diagram of a display device according to a second embodiment of the present invention.
4 is a conceptual diagram illustrating mapping of an image composed of conventional pixels to a display device including a subpixel.
5 is a block diagram of a display device according to a third embodiment of the present invention.
6 is a block diagram of a display device according to a fourth embodiment of the present invention.
7 is a block diagram of a display device according to a fifth embodiment of the present invention.
8 is a block diagram of a display device according to a sixth embodiment of the present invention.
9 is a block diagram of a display device according to a seventh embodiment of the present invention.
10 is a block diagram of a display device according to an eighth embodiment of the present invention.
11 is a block diagram of a display device according to a ninth embodiment of the present invention.
12 is a block diagram of a display device according to a tenth embodiment of the present invention.
13 is a block diagram of a display device according to an eleventh embodiment of the present invention.
14 is a block diagram of a display device according to a twelfth embodiment of the present invention.

The embodiments described in this section illustrate the invention but do not limit it. The invention is defined by the appended claims.

Some embodiments of the present invention modify the representation as in Figure 1 as follows. The color correction is incorporated in the output gamma block 160. The display device may or may not use a DBLC, and in particular, the DBLC block 140 and the scaling block 150 may or may not be included. The light source 110 may or may not be included, and in particular, the display device may not be of the LCD type. The display device may use LUTs, a circuit of a computer, or a combination of the circuits of the LUTs and a computer. For example, some values may be stored in the LUTs, and white and other values may be interpolated from the LUT values. For convenience of explanation, the examples described below use tables, but are not limited thereto.

Figure 2 shows a block diagram of a display device in which the output gamma block 160 includes tables such as 220, 230,

The color correction technique shown in Fig. 2 will be referred to herein as a "vector method ". The output gamma block 160 receives the subpixel values Rs, Gs, Bs, and Ws for the subpixels and outputs the gray values (Rg, Gg, Bg, and Wg) . When the subpixels emit corresponding primary colors, each subpixel value corresponds to the luminance of the corresponding subpixel. The input values Rs, Gs, Bs, and Ws may be linear with respect to brightness, although this is not required. In the embodiment of FIG. 2, the display device receives sRGB data and converts it to linear RGBW coordinates Rw, Gw, Bw and Ww (not required). The digital processing block 210 then processes the coordinates to generate the subpixel values Rs, Gs, Bs, and Ws. The present invention is not limited to the process described in sRGB or block 210. The block 210 may include the scaling block 150 or the DBLC block 140 as shown in FIG. 1, or may include an SPR rendering unit and / or another unit. Or the display device may not include the block 210 (i.e., the subpixel values Rs, Gs, Bs, and Ws may be the same as the linear RGBW coordinates Rw, Gw, Bw, and Ww) have.). The display device may be an SPR display device, or may not be an SPR display device.

The output gamma block 160 includes ACC tables including "ACC0 " for luminance calibration and" ACC1 "for white color difference correction. The table ACC0 includes separate tables 220 and 230 for each primary color. The table ACC0 220 is the three tables for the red, green and blue primary colors, and the table ACC0 230 is the table for the white primary colors. The RGB tables 220 may be tables that have been used in conventional RGB displays. The tone values stored in these tables are represented by Rg0, Gg0 and Bg0. For example, for each red value Rs, the ACCO table 220 of the red primary color reduces the corresponding tone value Rg0. The Green and Blue ACC0 tables are similar. The ACCO table 220 can be created using conventional techniques. For example, for the red primary color, the gamma function for the display device may turn on the red pixels (i.e., Rg, Gg, Bg, and Wg values with Gg = Bg = Wg = (Which is supplied to the display 124) by measuring the luminance of the image on the screen 126. The ACCO table of the red primary color can be obtained by inversely transforming the gamma function. The green and blue ACC0 tables can also be made in a similar manner.

The white ACC0 table reduces the white gradation value Wg for each input white subpixel value Ws. Table 230 can be made in the same manner as in Table 220 above.

The ACC1 table 240 converts the white subpixel value Ws into three RGB gradation values Rg1, Gg1 and Bg1. The three RGB gray level values Rg1, Gg1 and Bg1 are used to compensate for the white chrominance shift appearing when the white tone value Wg of the white ACC0 table 230 is supplied to the screen unit 120. [ Green and blue gradation values.

For a fully saturated color (i. E., A color with at least one sRGB coordinate of 0), the white subpixel value Ws may typically be zero, and the RGB ACC0 table 220 may be interpolated by interpolation blocks 250, Gg0 and Bg0) that are transmitted unaltered by the gray scale values (described below). These values (Rg, Gg, and Bg appearing at the output of the interpolation block 250) are transmitted to the screen unit 120 through the dither block 270. When the white subpixel value Ws is 0, the white ACCO table 230 reduces the tone value Wg of 0. Thus, the color correction in block 160 is highly effective for fully saturated colors.

The interpolation block 250 interpolates the tone values Rg, Gg, and Bg, which are outputs of the tables 220 and 240, based on the saturation degree parameter Sat and the color saturation index output from the block 260. For example, in some embodiments, the parameter Sat is as follows.

Sat = Ws / max (Rs, Gs, Bs, 1) (2A)

The 1 on the right was included to avoid dividing by zero. In this case, the parameter Sat represents the reciprocal of the degree of saturation. In some embodiments, the subpixel values Rs, Gs, Bs, and Ws may be weights of relative brightness. For example, let Yr be the luminance of the screen 126 when the red subpixels emit maximum light and all green, blue, and white subpixels emit at a minimum. Yg, Yb and Yw are defined in a similar manner for the green, blue and white subpixels. Then, in some embodiments, the parameter Sat is as follows.

_{Sat = (Ws * Y w)} / max [(Rs * Y r), (Gs * Y g), (Bs * Yb), 1] (2B)

The interpolation at block 250 may be linear or some other form. For example, in some embodiments, if linear interpolation is used, the Sat varies from 0 to 1 or less. For example, in the case of 2A, the RGBW conversion of the block 130 is performed so that Ww does not exceed max (Rw, Gw, Bw) and the digital processing of the block 210 is performed so that Ws is max (Rs, Gs, Bs. In conclusion, the parameter Sat has a minimum value of 0 and a maximum value of 1. In some embodiments, the tone values (Rg, Gg, and Bg) output from the block 250 are as follows.

Rg = Sat * Rg1 + (1-Sat) * Rg0 (3)

Gg = Sat * Gg1 + (1-Sat) * Gg0

Bg = Sat * Bg1 + (1-Sat) * Bg0

In some embodiments, if the color is white or gray, the parameter Sat is defined by (2A), Rs = Gs = Bs = Ws and Sat is one. Therefore, in the case of (3), the interpolation outputs (Rg, Gg, Bg) = (Rg1, Gg1, Bg1). That is, the white or gray color can be displayed more accurately.

A polynomial or other kind of interpolation method may be used and it may be a different kind of method for the combination of the outputs of the tables 220 and 240. For example, in some embodiments,

Rg = (Sat * Rg12.2 + (1-Sat) * Rg02.2) ^{1 / 2.2}

Bg = (Sat * Gg12.2 + (1-Sat) * Gg02.2) ^{1 / 2.2}

Rg = (Sat * Bg12.2 + (1-Sat) * Bg02.2) ^{1 / 2.2}

Note that although the parameter Sat appears as the reciprocal of the degree of saturation, the parameter Sat value for a given color depends on the selection of the RGBW representation of the color (Rs, Gs, Bs and Ws) The choice of the same metamer). As is well known, for most colors, the RGBW representation is not the same. Therefore, the parameter Sat is an index of the magnitude of the Ws coordinate with respect to the Rs, Gs, and Bs coordinates.

In some embodiments, to reduce the size (number of gates) of the circuit, the interpolation is performed for a portion that is not all of the RGB grayscale values. For example, the red and blue gradation values Rg and Bg can be obtained by the equation (3), but the green gradation value Gg is set to Gg1 or Gg0.

In some embodiments, the interpolation is also performed for the white tonality values. For example, the table 220 may be modified to provide an appropriate white tone value Wg0 for each combination of Rs, Gs, Bs, and Ws values. The interpolation block 250 may generate Wg by interpolating the Wg0 value and the Wg value generated by the table 230. [ The present invention is not limited to the above example.

The dithering block 270 is optional and can be used when the outputs Rg, Gg, Bg, and Wg of the output gamma 160 need to be reduced. For example, in some embodiments, the width of the tone values received by the picture unit 120 is 8-bit, but the width of the tone values generated by the output gamma 160 is 10-bit . Temporal, spatial, or temporal dithering can be performed using well-known methods and other cutting methods can be used.

In some embodiments, the white ACCO table 230 may be approximated with an inverse of the gamma function of the screen unit 120 for at least the white subpixels.

Specifically, in some embodiments, the ACC0 and ACC1 tables can be made as follows.

The combination of the optical measurement values is measured when supplying and displaying approximate values of the tone values (Rg, Gg, Bg, and Wg) in the screen unit 120. Specifically,

1. Obtain the color difference and luminance of the brightest R, G, B and W primary colors. Particularly, for any one of the primary colors, the corresponding tone values Rg, Gg, Bg and Wg at the input of the picture unit 120 are set to a maximum value (for example, 255) Are set to zero for the entire screen. If a backlight unit is used, the power consumption of the backlight unit is maximized. Then, the color difference (i.e., at CIE xy coordinates) and the luminance are measured. In some embodiments, if there is a large amount of noise in the luminance measurement, luminescence instead of luminance may be measured for the R, G, and B coordinates.

2. For each of the four primary colors, while the tonal values of the other primary colors remain at zero, the tonal value of the primary color increases from zero to the maximum value. For each increment value, the brightness of the primary color is measured. In some embodiments, if the width of the tone values is 8-bit, each primary color is set to 0, 8, 16 ... 240, 248, 255 (33 points), and the value gradually increases.

3. For the W increment, the color difference for each Wg value in the CIE xy coordinates is recorded.

For example, when such measurements are performed on a sample digital information display (DID) LCD panel, the following results are obtained. The CIE xy measurements for the brightest primary colors are as follows.

1.0, the brightness measurement values of the brightest primary colors are as follows.

These values are used with well-known colorimetric logarithms to produce the following two matrices. The first matrix converts the normalized RGBW measurements into CIE XYZ coordinates. The first matrix is as follows.

The second matrix defined by "R2X " which includes the first column of the first matrix and converts the linear RGB to CIE XYZ is as follows.

The measured value of the increased primary color brightness can be used to produce an output gamma curve for each primary color using well known methods. The gamma curve can be inversely transformed to produce the ACC0 tables 220 and 230. [

The ACC1 table 240 can be created in the following manner.

1. For a particular range of Ws values, for example, each of the possible values, the CIE XYZ coordinates of the "ideal" white color are determined. For example, Ws may have a value from 0 to 255 and the ideal white color is a specific color DD. For example, DD may have a color difference of D65, and the luminance of the brightest white primary color may be measured as described above. It is assumed that the output gamma value for the white primary color is a power function (1) with = 2.2 or a function having a close value. If the ideal white XYZ coordinates are C2i, the following is obtained.

C2i = DD * (i / 255) ^2.2, where i = 0, 1, 2, ... 255 (4)

2. Let the corresponding 256 coordinates of the measured white primary color be WXYZ _i . These coordinates can be obtained by measuring the luminance and chrominance of the white primary color mentioned above when Wg = i increases from 0 to 255. As described above, in some embodiments, the measurement may be measured at some portions Wg = 0, 8, 16 .... In this case, the remaining WXYZ _i values are obtained by interpolation (ie linear or other interpolation method).

The desired white color difference correction for each Wg = I value can be expressed in the following CIE XYZ coordinates.

SUM _i = C2 _i -WXYZ _i (5)

These colors can be transformed into the linear RGB coordinate method using the matrix R2X described above and the following can be obtained.

ACC1 _i = R2X- ¹ SUM _i (6)

Therefore, if Ws is C2i corresponding to the ideal white color, that is, according to the following expression,

Ws = const * (i / 255) ^2.2 (7)

(Since "const" is a constant determined by the bit width of Ws, the ACC1 _i is the preferred RGB calibration within the linear RGB coordinates corresponding to the primary color red, green and bluish color differences of the display. And the outputs of the RGB ACC0 table are stored in the ACC1 table 240 for the Ws values 7 rounded to the corresponding integer. For the remaining Ws values, the items of the ACC1 table can be interpolated. Or by eliminating all remaining Ws values and interpolating when displaying an image.

In some embodiments, the selection of the white-point DD affects how well the white color difference is corrected. If the DD chrominance is selected as the chrominance of the color that is generated when all the red, green, blue and white luminaries emit light at the maximum, the ACC calibration may be reduced and the ACC may be more accurate. Also, the DD chrominance may be selected as a chrominance of a color generated when the white luminous body emits light at a maximum, and the red, green and blue luminous bodies emit light at a minimum.

If the maximum ACC1 correction is small, the outputs of Rg, Gg, and Bg of the interpolation block 250 may be output to the digital processing block 210 (e.g., 210) as the possible values Rg0, Gg0 and Bg0 of the ACCO table 220 become wider The range of possible values Rs, Gs, and Bs is widened, and the possibility of overflow or underflow is reduced (if an overflow or an underflow occurs, the tone value Rg, Gg or Bg) More complex color gamut mapping can be performed using hard-calmed or well-known techniques. For example, Brown Elliott et al., Filed on December 6, 2007, Reference is made to US Patent Application Publication 2007/0279372 A1 to " MULTIPRIMARY COLOR DISPLAY WIITH DYNAMIC GAMUT MAPPING &quot;. Figure 3 shows another embodiment for implementing a "delta offset" color correction method, The ACC1 table 240 includes Stores a group ACC1 _i correction (in Fig. 3 (dRi, dRg, dBi) of the "delta offset (delta offset)" represented by). That is, the correction are within a linear color space. The ACC1 _i calibration (dRi , dRg, dBi) are added to the corresponding linear Rs, Gs and Bs coordinates generated in the digital processing block 210. The summation is performed in a summer 310. The output of the summer is shown in Figure 2 Is returned to the RGB ACC0 table 220 generated as shown in Table 2. The Rg, Gg and Bg outputs of the table 220 are provided as outputs of the output gamma 160. The Wg gray level values are shown in Figure 2 The ACC1 table 240 and the summer 310) that perform the linear color correction are shown as a separate block 320 However, the present invention is not limited to a specific circuit.

In some embodiments, the table 240 corrects the white value Ws. Alternatively, the Rg, Gs, and Bs may generate the white-value correction.

Increasing the value of the white subpixels only to increase the brightness of the image are conventional dark RGBW display devices or RGBW display devices used under poor conditions. For example, in such a display, the white subpixel value Ws is multiplied by a "white gain" factor and the resulting value is hard-calmed to the possible range of the Ws value, Quot; hard-calmping "to a range between any number A and B is to set the lower boundary A if the specific value is less than A, and to the upper boundary B if the specific value exceeds B . In Fig. 2 and Fig. 3, the white gain is adjusted before or after the white color difference correction. For example, in FIG. 2, the Ws value may be adjusted before being provided to the output gamma, or the white gain adjustment may be made to the Wg value output from the white ACCO table 230 or dither block 270 have. In Figure 3, the white gain adjustment is performed on (i) the output Ws of the digital processing block 210 or (ii) only on the input Ws of the white ACCO table 230 and It may not be performed for the input Ws.

As previously mentioned, the embodiments of Figures 2 and 3 can be used with SPR displays. If the indication is non-SPR, each pixel of the original image is associated with a quad of red subpixels, green subpixels, blue subpixels, and white subpixels. The color difference correction for the white subpixel by the ACC1 table 240 is combined with the subpixel values of the red, green, and subpixels of the same quad (either blocks 250 and 260 of FIG. 2, or summers 310).

However, for some SPR indications, there is no particular association between white subpixels and red, green and blue subpixels. One example of such a display device is a PCT application, U.S. Patent Application No. 11 / 278,675, filed November 30, 2006, WO 2006/127555, and US 2006/0244686 A1, published November 2, 2006, have. In the embodiment of FIG. 4, the input image 404 is represented by a myriad of pixels 406. Each pixel 406 is mapped to a combination of only two subpixels representing the color of the pixel.

In particular, the display screen 126 includes red subpixels 420R, blue subpixels 420B, green subpixels 420G, and white subpixels 420W. All the subpixels 420 have the same area. Each pixel 406 is mapped to a subpixel pair 424 that includes two adjacent subpixels in the same row. Each of the subpixel pairs 424 includes a red subpixel 420R and a green subpixel 420G (these pairs are hereinafter referred to as an "RG pair") or a blue subpixel 420B and a white subpixel 420W ) (These pairs are hereinafter referred to as "BW pairs"). In each RG pair, the red subpixel 420R is located to the left of the green subpixel 420G. In each BW pair, the blue subpixel 420B is positioned to the left of the white subpixel 420W. The RG and BW pairs are arranged alternately in each row and column. This is only one embodiment, and other arrangements of subpixels are possible. For example, the red sub-pixel 420R may be located on the right side of the RG pair, and the blue sub-pixel 420B may be located on the right side of each BW pair. Also, the areas of the subpixels may not be the same. For example, in some display devices, the area of the white subpixel 420W may be adjusted to produce a desirable ratio of RGB to white W, and the area of the white subpixel 420W may be different than white The area of the subpixel may be different. The red, green, and blue subpixels 420R, 420G, and 420B may also have different areas.

Each pixel 406 (hereinafter, referred to as pixel 406 _{x, y} ) arranged in the x column and the y row of the image is divided into the subpixel pair 424 (hereinafter referred to as "424 _{x , y} "). In the above screen 126, the continuous exponent x and y represent successive pairs, not continuous sub-pixels. Each pair 424 includes only two subpixels and provides a high range and resolution in luminance, not chrominance. Thus, some of the luminance of the input pixels may have to be moved to adjacent pairs 424 in the subpixel rendering process (SPR). Various subpixel rendering processes may be used. One example is: Let RGBW coordinates of the pixel 406 _{x, y be} Rw _{x, y} , Gw _{x, y} , Bw _{x, y,} and Ww _{x, y} . If the subpixel pair 424 _{x, y} is not located at the edge of the display, the value Rs of the red subpixel for the pair can be calculated as:

_{Rs = 1/2 * Rw x} , y + 1/8 * Rw x-1, y + 1/8 * Rw x + 1, y + 1/8 * Rw x, y + 1/8 * Rw x, y ₊₁ (8)

The green, blue and white subpixels 420G, 420B and 420W except for the edges can be calculated as described above. If the subpixel pair 424 _{x, y} is located at the edge of the display, pixel values outside the edge of the display are required and may be set to zero or other digits. See US Application Publication No. US 2005/0225563, entitled " IMPROVED SUBPIXEL RENDERING FILTERS FOR HIGH BRIGHTNESS SUBPIXEL LAYOUTS &quot;, filed on even date herewith, U.S. Serial No. 10 / 821,388.

2 and 3, the SPR process may be performed in the digital processing block 210, and the outputs Rs, Gs, Bs, and Ws of the SPR process may correspond to the output gamma 160 of FIG. 2, May be provided to the ACC circuit 320. The outputs Rs, Gs, Bs, and Ws of the SPR process are also applied to the output gamma 160 of Figure 2 or the delta ACC circuit 320 of Figure 3 before sharpening, gamut clamping If the values of Rs, Gs, Bs, and Ws are outside the display gamut, gamut clamping is required) or other types of processing.

In the ACC circuit of FIGS. 2 and 3, the RGB coordinates obtained in the ACC1 table 240 for the white subpixel 420W are different from those of the other adjacent subpixels 420R, 420G and 420B, (By blocks 260, 250 or summer 310) with the RGB coordinates of the modified subpixels. For example, in some embodiments, the blue subpixel 420B is obtained from the same pair 424 as the white subpixel 420W, and the red and green subpixels 420R and 420G are located immediately to the left Gt; 424 &lt; / RTI &gt; However, if the white subpixel 420W is in a pair 424 located at the left edge of the screen, then the red and green calibration is not performed, and in some embodiments the blue calibration is not performed. Other selections of calibration subpixels may be possible. In some embodiments, one calibration value (i.e., Rg1, Gg1, and Bg1 in Figure 2, or one of dRi, dGi, and dBi in Figure 3) is applied to two or more subpixels of the corresponding red, 420).

In Fig. 5, the white color difference correction is performed before the SPR process. In this embodiment, the digital processing 210 generates Rw, Gw, Bw and Ww data. That is, a data quad for each pixel 406 is generated. In the delta ACC block 320, the Ww output of the block 210 for each pixel 406 is output to the ACC1 table or to a circuit 510 for generating the appropriate RGB calibration ACC1 _i = (dRi, dGi, dBi) for the pixel ). The calibration is added to the Rw, Gw and Ww coordinates for the pixel 406 by a summer 310. [ The sum referred to as Rw, Gw and Ww is provided to the SPR block 510 along with the output of the digital processing block 210. [ The SPR block 510 performs subpixel rendering to generate Rs, Gs, Bs, and Ws subpixel values for the subpixels 420. Further, the Ww output of the SPR block 510 may be multiplied by the white gain factor WG. The multiplication is performed in a multiplier 520. The subpixel data is provided to output gamma 160, for example, as shown in FIG. The subsequent procedure is shown in FIG.

Performing white color correction before SPR as in FIG. 5 is more advantageous because the combination of the Ww value and the Rw, Gw, and Bw values before the SPR is well defined by the pixels 406.

In some embodiments, the ACC1 table 240 for the embodiment of FIG. 5 is generated as follows. A specific ideal white color (DD) is selected as described in FIGS. 2 and 3. The white color measurement WXYZ _i is performed for each gray level value value Wg = i (for example, i = 0, 1, 255 for 8-bit gray level values) and Rg = Gg = Bg = For each value Ww, ACC1 output of the table 240 is determined by the difference between the ideal color DD * Ww / max {Ww} and the measured color WXYZ _i. The measured color WXYZ _i is selected to have the luminance (e.g., CIE Y coordinate) closest to the luminance of Ww. The vector difference is expressed in linear RGB coordinates using the above-mentioned matrix R2X- ¹ .

In some embodiments, the ACC1 table 240 is made as follows. CIE xyy coordinates are obtained through a large number of measurements. In each measurement, one of the tonal values of the primary colors is a positive number, and the remaining tonal values are 0 (as in step 1 mentioned above for the embodiment of Fig. 2). For each of the x, y and Y coordinates of each primary color, the measurement is made flat by the box filter. Also, for each primary color, the Y measurement subtracts the smallest Y value, normalizes the result by dividing it by the appropriate factor, and adds 1 to the normalized Y value.

Let mx be the number of measurements performed for each primary color. For the sake of discussion, it is assumed that the tone value has an 8-bit width, and a measurement of mx = 256 is performed for each of the primary colors. If i is a tone value, the flattened and normalized values for the red primary color are referred to as Rx _i , Ry _i , RY _i , and the same notation is applied to the other primary colors.

The flat and normalized values for the brightest primary color (i.e., the gray-scale value of 255) can be obtained by a conversion matrix between the RGBW color space limited to the brightest measured primary color and the CIE YXZ color space. In particular, the following matrix is defined.

Each element of p is multiplied by the corresponding Y / y value (where Y is normalized and y is flattened). The resulting matrix (hereinafter M2X) provides the conversion from RGBW to XYZ. The first three columns of the M2X form a matrix of R2X and convert RGB to XYZ. The RGB is a linear color space defined by the brightest red, green, and blue primary colors of the display. Both R2X and M2X can be said to be "normalized" in that the brightest primary colors are represented by normalized values of Y values.

Ideally, the x and y coordinates for each of the primary colors should be independent of the gray value i of the primary colors. However, this is not the case for some display devices discussed above.

The measured values of Rx _i , Ry _i , RY _i, etc. are "re-sampled " to be more evenly distributed within the linear color space. In particular, since the tone values i are not linear with respect to the luminance, the tone values are not evenly distributed in the linear space. The resampling process is assumed to be 2.2 of the power function (1). It is assumed that the variable j has a value of linear space. The j value is equal to the corresponding j = (i / 255) ^2.2 * MAXCOL for each i value. MAXCOL is the maximum j value (in some embodiments MAXCOL is the maximum value for each of Rw, Gw, Bw, and Ww output from the digital processing circuit 210 and the conversion circuit 130. Alternatively, MAXCOL is 2047). The above-mentioned measurements provide xyY coordinates for the j value of about (i / 255) ^2.2 * MAXCOL, and these values are not evenly distributed. Thus, the xyY coordinates for the evenly distributed j = 0, 1, 2, ... MAXCOL are obtained by cubic spline interpolation by the planarized and normalized xyY measurement for the unevenly distributed j values ). The resampled values are converted into the XYZ color space. And by displaying the red, green, blue, and the XYZ values of white primary color to each RXYZd _{j, j} GXYZd, BXYZd _j, and _j WXYZd.

The ideal white color DD can be selected through various methods. One possible way is to choose a color that is the same as the chromaticity for the lightest displayed color when all of the tonal values are the maximum of mx (i.e., 255). The color DD is as follows

Another possible method is to select the same color as the chromaticity for the brightest displayed color when the white tonal values are Wg = mx and all the remaining tonal values are zero. The color DD is as follows.

In any case, it is possible to obtain ideal white colors (gradation colors) for different J values by multiplying the DD color by j / MAXCOL. MAXCOL is the maximum value of j.

The resulting color DD * j / MAXCOL is transformed into the RGB space by applying the matrix R2X- ¹ . The result "ramp" is represented by IRGB _j (ideal RGB).

If the DD = DRGBW, the re-sampling the RGBW data deohaejyeo for each j are calculated by RGBWXYZ RXYZd _j = _j + _j + GXYZd BXYZd WXYZd _j + _j, the sum is RGB using the R2X matrix ^-1 . The resultant values are denoted by RGBRGBW _j . The ACC table 240 is created with the difference IRGB _j -RGBRGBW _j . Each difference includes the red, green, and blue components provided by table 240 as a result of Ww = j.

If DD = DW, the resampled W data is converted to RGB using the matrix R2X- ¹ . The resultant value is denoted by WRGBj. The ACC1 table 240 is created by the difference of IRGB _j -WRGB _j . Each difference includes the red, green, and blue components provided by the ACC1 table 240 as a result of Ww = j.

If DD = DRGBW or DD = DW In some embodiments, the blue components of table 240 are hard-clamped to a maximum value of zero. That is, the blue calibration does not become positive.

Metamer adjustment

The white color difference correction includes adding the positive or negative values output by the ACC1 table 240 to the red, green, and blue values. The resulting colors may be located outside the gamut (e.g., if the addition causes overflow or underflow). The hue outside the gamut can be shifted into the gamut range by an appropriate gamut clamping process. However, gamut clamping can cause image distortion. The image distortion may be reduced by selecting an appropriate RGBW representation at block 130 or 320, since the color representation by RGBW coordinates is not generally unique. For example, if the Rw, Gw, or Bw coordinates overflow, the Rw, Bw, and Gw coordinates may be reduced by increasing the Ww coordinate. In the case of underflow, the Rw, Bw and Gw coordinates can be increased by decreasing the Ww coordinate.

Differences in RGBW representations for color are referred to in the present application and in some documents as "metamers" (other references use "metamers" as the meaning of electromagnetic waves in other spectral power distributions However, other RGBW representations do not necessarily mean different spectral power distributions).

6 is the same as FIG. 5 except block 320, which includes a metamer calibration circuit 604 ("post-color calibration" or PCC) at the output of adder 310 Fig. The ACC1 table 240 is made as in FIG. The SPR block 510 may or may not be included (i.e., the embodiment is suitable for both SPR and non-SPR displays).

FIG. 6 represents the RGBW transform block 130 as two blocks 130.1 and 130.2. The block 130.1 converts the sRGB data into a linear RGB color space. The color space may be independent of the display device, or may include a primary color having the same chrominance as the red, green, and blue illuminants of the screen unit 120. The 130.1 block is sometimes referred to as "input gamma ". In some embodiments, all processing between the input gamma 130.1 and the output gamma 160 is performed in linear coordinates (Rw, Gw, Bw and Ww and Rs, Gs, Bs and Ws are linear expressions of color) do.

The 130.2 block transforms the linear RGB data (denoted r, g, and b) into RGBW data (denoted as Rw, Gw, Bw, and Ww) using well known methods. In some embodiments,

r = M ₀ * Rw + M ₁ * Ww (9)

_{g = M 0 * Gw + M} 1 * Ww

b = M ₀ * Bw + M ₁ * Ww

The M ₀ and M ₁ are constants corresponding to the luminance characteristics of the pixels 420 as follows.

(10): M ₀ = (Y _r + Y _g + Y _b ) / (Y _g + Y _b + Y _w )

M ₁ = Y _w / (Y _g + Y _b + Y _w )

Y _r , Y _g , Y _b, and Y _w are defined as follows. Y _r is the luminance of the display 110 when the red subpixels 420R emit light at the maximum and all the remaining subpixels emit light at the minimum. The Y _g, Y _b and Y _w are defined in a similar way with respect to the green, blue and white sub-pixels.

If the Ww coordinates are known, Rw, Gw, and Bw are calculated by (9).

In some embodiments, each of r, g, b, Rw, Gw, Bw, and Ww is an allowed integer ranging from zero to a specific maximum number MAXCOL. For example, in some embodiments, MAXCOL = 2047.

In the above equation (9), if r, g, or b is 0, Ww must be zero. If r = g = b = MAXCOL then Ww = MAXCOL. However, for most colors, Ww can be selected in a variety of ways (selecting one or more metamers). In order for each of R w, G w, B w and W w to be in the range of 0 to MAXCOL, Ww must satisfy the following range.

minW? W? maxW (11)

Lt;

minW = [max (r, g , b) -M 0 * MAXCOL] / M 1

maxW = min (r, g, b) / M ₁

Table 1 below shows the pseudo code used as the simulation code for one embodiment of the ACC / PCC block 320 of FIG. The simulation code is written in a well-known programming language, LUA. The LUA is similar to the C language.

In line ACC7 of Table 1, the Rw, Gw, Bw and Ww are taken from the processed pixel 406. In line ACC9 of Table 1, the Ww value is multiplied by the white gain factor. The white-gain-corrected result is expressed as Wg, which should not be confused with the gradation value Wg mentioned elsewhere. In the line ACC10-12 of Table 1, the white-gain-corrected result Wg is supplied to the ACC1 table 240. [ The result values dRi, dGi and dBi are expressed as Dr, Dg and Db, respectively. In the lines ACC13-16 of Table 1, the summer 310 generates summation values R_pcc, G_pcc and B_pcc through summation. The white value W_pcc of the line ACC16 is simply the original value Ww. The R_pcc, G_pcc, B_pcc and W_pcc values are metamer adjusted by the circuit 604 described in the line ACC24-36.

Specifically, it is first determined which of R_pcc, G_pcc, and B_pcc is negative. The minimum value of the three values is indicated as minRGB_pcc in the line ACC24. If the minimum value is negative, a positive number adjustment is required. That is, positive pos_adj_pcc must be added to R_pcc, G_pcc, and B_pcc. If the minimum value is not negative, pos_adj_pcc is set to zero. See line ACC24-28.

The white coordinate W_pcc should be reduced by W_adj_pcc corresponding to pos_adj_pcc. If M ₀ = M ₁ = 1/2, W_adj_pcc = pos_adj_pcc. Generally, W_adj_pcc = M ₀ / M ₁ * pos_adj_pcc as shown in the above equation (9). The M ₁ / M ₀ value is often denoted M ₂ .

However, the white coordinate W_pcc (i.e., Ww) should not be reduced to a number less than or equal to zero. Thus, in line ACC31 the decreasing value W_adj_pcc is hard-clamped to its possible maximum value, Ww. In line ACC32 of Table 1, the adjusted increment RGB_adj_pcc for the red, green and blue coordinates is calculated from W_adj_pcc. In line ACC 33-36 of Table 1, the metamer adjustment is made by increasing R_pcc, G_pcc and B_pcc and decreasing W_pcc with RGB_adj_pcc. In line ACC 39-43 of Table 1, the new values R_pcc, G_pcc, B_pcc and W_pcc are hard-clamped to a maximum value MAXCOL.

The embodiment of Table 1 calibrates the overflow using hard-clamping (in lines ACC39-42). However, in other embodiments, the overflow is corrected similarly to the underflow. The maximum overflow of the coordinates (Rw, Gw, and Gw) output from the ACC1 table 240 is determined, and the corresponding Ww increase amount is determined, and the maximum overflow Reduce or eliminate.

The above-mentioned embodiments may or may not be used with SPR. Also, the embodiments may or may not be used with DBLC. Now, some embodiments useful for indication using DBLC will be described.

FIG. 7 shows an LCD DBLC embodiment in which the SPR 510 is performed before the ACC. The ACC1 calibration is performed in the output gamma block 160 as shown in FIG. However, it may be performed before the output gamma block 160 as in FIGS. 3, 5 and 6 (if the ACC is performed as in FIGS. 5 and 6, the ACC is performed before the SPR 510) .). The DBLC block 140 and the scaling block 150 may operate as in FIG. The SPR 510 is performed on the output of the scaling block 150.

As mentioned above, since the parameters BL and INVy are calculated based on the previous ACC data, the DBLC can interfere with the ACC calibration. The embodiment of Figure 8 allocates this problem as described below. This embodiment is suitable for both SPR or non-SPR indications. The SPR may be performed before or after the ACC / PCC block 320. The ACC / PCC block 320 may skip the metamer adjustment as in FIG. 6 or in FIG. 3 or FIG.

Referring to FIG. 1, the BL and INVy are delayed by one frame, as in the prior art. Specifically, the INV _y value determined by the sRGB for one frame ("current frame") is used in the scaling block 150 for scaling of the next frame. The BL value determined for the current frame is used to control the light source 110 when the screen unit 120 displays the next frame. The current frame is scaled and displayed using the BL and INVy values determined by the previous frame of data. The "k" to denote the frame number, and let to be indicative of the above-BL and BL INVy value calculating _k and _k INV from the data of the k-th frame. The BL _k and INV _k are used to scale and display the next k + 1 frames. This one frame delay makes it possible to display the current k frame before the BL _k and INV _k values are determined. In fact, displaying the current frame may be started before all sRGB data for the current frame is input. To reduce image errors, the BL values may be "decayed &quot;. That is, the BL _k value may be generated in the block 140 as a weighted sum of the BL value determined by the current frame and the previous BL value BL _k-1 . In this case, the reciprocal of the INVy may also decline. For some display devices displaying 30 frames per second, if the brightness of the image suddenly changes, it may take 36 frames to capture the brightness of the image for the BL and INVy values. This delay can be tolerated in many applications. In fact, in the absence of a sudden change in the screen brightness, the BL and INVy values generally do not change significantly between frames, and a delay of about one frame does not cause significant degradation of the image. If the brightness suddenly changes, it takes time for the observer to adjust the image, so that the error due to the delay of the BL and INVy values is not noticeable. Which is incorporated herein by reference and is incorporated herein by reference in its entirety, Hwang et al. U.S. Application No. US 2009/0102783 A1, filed April 23, 2009, which is hereby incorporated by reference.

In Fig. 8, the above-mentioned frame delay is shown in block 804. Fig.

In some embodiments of the invention, the BL and INVy values are determined from the ACC corrected data for the previous frame, rather than the uncorrected data using the delay.

In the example of Fig. 8, the suffix "c " means the current frame. Rw, Gw, Bw, and Ww data for the current frame is represented by Rw, Gw, Bw, and Ww _c.

The scaler 150 is scaled using a parameter determined by the INVy _c-1 of Rw, Gw, Bw, and Ww _c the current frame to the previous frame. ACC block 320 is ACC calibration and PCC meth Murray (metamer) of Fig. 3, Fig. 5 or in connection with 6 are mentioned above as possible from Table 1, the output of the scaler for the current frame Rw, Gw, Bw, and Ww _c Adjustment is performed. Wherein the Rw, Gw, Bw, and Ww _c output of the ACC block 320 may be processed as discussed above with respect to displaying on the screen 126 (e.g., SPR block (510 in SPR display) , Output gamma 160, and possibly other methods).

Rw, Gw, Bw, and Ww _c output of the ACC / PCC block 320 may also be provided to scaling block 810 - "un-scaled (un-scaling)" for unloading. Can be scaled (un-scaling) (or the equivalent, determined in the previous frame, said Rw, Gw, Bw, and Ww _c the language by the data division by the parameter INVy _c-1 determined in the previous frame The parameter BL _c-1 may be multiplied by the Rw, Gw, Bw and Ww _c data).

DBLC block 140 the language available in the prior art using the Rw, Gw, Bw, and Ww _c data supplied from the scaling block 810 to determine the parameters of the INVy BL _c and _c. These parameters are provided to the scaling block 150 and the light source 110 for use in the next "c + 1" frame.

In some embodiments, the un-scaling is performed for BL and INVy rather than Rw, Gw, Bw, and Ww. In particular, the un-scaling block 810 is omitted. The DBLC block 140 scales the thresholds (such as MAXCOL) by multiplying the threshold by INVy _c-1 and scales the thresholds (such as MAXCOL) for the current frame using the scaled threshold un-scaled from the (un-scaling) the output Rw, Gw, Bw, and Ww _c determines the INVy BL _c and _c. These BL _c and INV y _c values are "un-scaled &quot;. That is, it is divided by INVy _c-1 (or equivalently, multiplied by BL _c-1 ).

9 shows that the unscaled block 810 outputs the Rw, Gw, Bw and Ww data outputs of the ACC 320 and Rw, Gw, Bw and Ww of the circuit 130.2, 8 is the same as the embodiment of Fig. 8 except that it is performed for the difference of data output dRw, dGw, dBw and dWw. The data differences dRw, dGw, dBw and dWw are small and smaller bits and logic gates may be needed for un-scaling. The summer 910 adds the un-scaling data differences dRw, dGw, dBw, and dWw to the Rw, Gw, Bw, and Ww data outputs of block 130.2, 140. The rest of the procedure is the same as described above with reference to FIG.

In particular, if the ACC block 320 is the same as in FIG. 5, (dRw, dGw, dBw) is equal to (dRi, dGi, dBi). In some embodiments, ACC block 320 is the same as in FIG. 6 (including metamer adjustment 604), but (dRw, dGw, dBw) is still set to (dRi, dGi, dBi) Thereby reducing the number of gates.

The embodiments of Figures 8 and 9 are suitable for RGB display devices. For example, the ACC 320 performs luminance correction on the RGB data. The brightness correction is better, for example, by tables similar to the ACCO table 220 (FIG. 2). In some embodiments, the calibrated RGB luminance is placed in a linear color space. That is, the output of the ACC 320 may be linear RGB coordinates Rw, Gw, and Bw. The invention is not limited to RGB or other types of primary color combinations except as defined in the appended claims below.

We will now describe possible metamer adjustments before scaling. In the LCD, the Rw, Gw, Bw and Ww coordinates for each pixel 406 should be perfectly close together to provide a high image quality with a minimum output power BL. In some embodiments, at block 130.2, Ww is set by an average value of max (r, g, b), r, g, and b or other methods. The above-mentioned Higgins et al. , Which is incorporated herein by reference in its entirety. However, the ACC adjustment may not achieve this, even if the output power BL is minimized or pursued for other purposes by the metamer selection in the transform block 130.2. Thus, in some embodiments, the RGBW transformation 130.2 follows the ACC-associated metamer adjustment ACCMET 1010 (eh 10) to adjust the metamers based on the expected ACC operation. For example, the metamer may be adjusted such that the Rw, Gw, Bw and Ww coordinates after the ACC may be close to each other.

In particular, equation (9) requires that the Rw, Gw, and Bw values exceed MAXCOL or MAXCOL / M0. For example, if b = 0, Ww = 0. If r = g = MAXCOL and Rw = Gw = MAXCOL / M0. For purposes of illustration, assume M0 = M1 = 1/2. That is, the white subpixels are as bright as the red, green, and blue subpixels. In this case, the Rw, Gw, and Bw values may be as large as 2 * MAXCOL. The screen unit 120 can only accept colors whose linear Rw, Gw, Bw, and Ww coordinates do not exceed MAXCOL. To display different colors, the power of the backlight unit may be increased up to 1 / M _o (e.g., doubled) by the multiplication factor BL. In this case, the Rw, Gw, Bw and Ww coordinates must be multiplied by INVy = 1 / BL. This multiplication causes the Rw, Gw, Bw and Ww coordinates to be in the range of 0 to MAXCOL, which is a possible range. The maximum INVy value INVymax, which makes the Rw, Gw, Bw and Ww coordinates within the possible range, determines the minimum BL value BLmin. However, for power saving, some embodiments increase the power of the backlight unit by increasing or multiplying the power of the backlight unit by a number less than BLmin. Thus, the Rw, Gw, Bw and Ww data multiplied by INVy = 1 / BL may include coordinates outside the possible range, and thus the scaler 150 performs a specific gamut clamping to determine Rw, Gw, Ww coordinates within the above-described possible range. Suitable gamut clamping techniques include hard-clamping and other well-known techniques. Which is incorporated herein by reference and is described in Brown Elliott et al. No. 2007/0279372 A1, filed December 6, 2007, entitled &quot; MULTIPRIMARY COLOR DISPLAY WIITH DYNAMIC GAMUT MAPPING &quot;.

10 shows a non-SPR embodiment in which the output of block 130.2 is provided to a metamer adjustment circuit 1010 that adjusts the Rw, Gw, Bw and Ww values as appropriate for the next ACC operation of block 320 . In some embodiments, ACCMET 1010 reduces image distortion in gamut clamping. The values of Rw, Gw, Bw and Ww adjusted by the ACCMET 1010 are provided to the scaling block 150. [ The outputs Rw, Gw, Bw and Ww of the scaling block 150 are provided to the ACC block 320. [ The ACC block 320 may be the same as in FIG. 3, FIG. 5, or FIG. A multiplier 520 may be used to multiply the output Ww of the ACC 320 by the white gain WG. Since the SPR is not used, the pixel values Rw, Gw, Bw and Ww behave like the subpixel values Rs, Gs, Bs and Ws and are provided to the output gamma 160.

Since the color correction performed by the ACC 320 can change the peak sub-pixel values Rw, Gw, Bw, and Ww to affect the proper setting of the output power of the backlight unit (as specified by the parameter BL) The DBLC 140 is adjusted by white color difference correction. In particular, the output of ACCMET 1010 is provided to ACC_DBLC circuit 1020. The ACC_DBLC circuit 1020 performs a similar function to the ACC function of the block 320 to estimate the ACC white balance correction. In generating the BL and INVy values, the DBLC 140 considers the estimation. One frame delay may or may not be used. That is, the BL and INVy values generated from the current frame may be used for the current frame, the next frame, or a specific subsequent frame.

11 is the same as FIG. 10 except for the sum of the SPR block 510 at the output of the ACC 320. FIG. The white gain adjustment by the multiplier 520 is performed on the Ws output of the SPR 510. The output gamma 160 may be the same as the example of FIG. 3 or FIG. The ACC 320 can be described with reference to FIG. 3, FIG. 5, or FIG.

In some embodiments, ACCMET 1010 is omitted, but ACC_DBLC 1020 is maintained. In some other embodiments, ACC_DBLC 1020 is omitted, but ACCMET 1010 is maintained. In some embodiments, the functions of ACCMET 1010 and ACC_DBLC 1020 are not clearly distinguishable in the circuit.

12 illustrates one embodiment of the ACCMET 1010. Table 2 below is a numeric code used as the simulation code for one embodiment of the ACC block 320 of FIG. The simulation code was written in LUA.

The RGBW output of the conversion block 130.2 is provided to the delta ACC predicting unit 1210. [ The delta ACC prediction unit 1210 (lines ACM1-27 in Table 2) predicts the white color difference correction performed in the ACC 320. [ The prediction may not always be accurate but may be an estimate of an approximation. For example, the predictor is similar to the ACC1 table 240 of FIG. 5 or FIG. 6, but may use fewer bits for each item to reduce the size of the circuit. Further, since the ACCMET 1010 can change the white coordinate Ww, the corresponding color difference correction calculated on the basis of the changed coordinates is obtained by performing the predicted correction calculated based on the Ww value input to the ACCMET 1010, can be different.

In the above embodiment of the line ACM3-14, the predicted value may be 7-bit wide, but the RwGwBwWw coordinates may be 12 or 14-bit wide. The predicted white value W_predict is a value obtained by multiplying Ww by the white gain factor (line ACM5) and scaling the obtained value by multiplying Ww by INVy _c-1, which is the INVy value for the previous frame (line ACM6). The INVy _{c -1} is used to scale the data of the current frame in the scaler 150 (in particular, since the white gain factor and the predicted values are represented by the same integer as 256 times their actual value, The result of the W_predict value is clamped to a value of up to 7-bits of 127 (line ACM7) and provided to an ACC table similar to the ACC1 table 240 but with fewer bits. In Table 2, the ACC1 table 240 appears as three tables accRmini, accGmini and accBmin (line ACM8-10). These tables provide the red, green and blue calibration values Dr, Dg and oil. Since the metamer adjustment is performed on the outputs Rw, Gw, Bw and Ww values of the unscaled block 130.2, these values are divided by INVy (line ACM11-13) to not perform the scaling.

Line ACN 16-26 represents another option for calculating Dr, Dg and Db. In this option, the predicted values have the same width as the Rw, Gw, Bw and Ww values (i.e., 12 or 14-bits). The W-predict value is clamped to MAXCOL rather than being truncated to 7 bits. The accRdel, accGdel, and accBdel tables are used instead of the tkdrl accRmini, accGmini, and accBmini tables. Otherwise, the option is similar to the option in line ACM3-14.

After calculating Dr, Dg, and Dr by block 1210, the metamer adjustment is made consistent with the purpose of the metamer selection. In the example of Table 2, it can be assumed that the above object is to reduce the output power of the backlight source 110. To achieve this goal, the maximum value of Ws for each white subpixel should be approximately equal to the maximum of the Rs, Gs and Bs for adjacent red, green and blue subpixels. In one embodiment of Table 2, an attempt is made to process such that the ACC-adjusted Ww value for the pixel is approximately equal to the Rw, Gw, and Bw values for the pixel. In the transform block 130.2, it is assumed that the Ww value is determined as a maximum value of Rw, Gw, and Bw for the pixel. In order to obtain some intuitive understanding, if there is no ACCMET 1010 and M ₀ = M ₁ = 1/2, Rw = Gw = Bw = Ww and Dr = Dg = Db, Suppose that Rw, Gw and Bw are increased by D values which are common values of Dg and Db. Then, in order to make Rw = Gw = Bw = Ww at the output of the ACC block 320, the ACCMET 1010 must reduce Rw, Gw and Bw by D / 2 and increase Ww by D / 2 do.

The D / 2 value is determined in the "Negative Adjust" block 1220 appearing on line ACM 34-45 (D / 2 may be negative or positive, despite the name "Negative Adjustment"). In Table 2, the D / 2 value of a negative number is named "neg_adj ". In particular, as shown in line SCM 35-37, the predicted calibration values R_corr, G_corr, and B_corr are determined in line ACM39. If the maximum value is equal to the red corrected value, neg_adj is set to Dr / 2. If the maximum value is not equal to the red corrected value and equal to the green corrected value, the neg_adj is set to -Dg / 2. If the maximum value is not equal to the red corrected value and the green corrected value and is equal to the blue corrected value, neg_adj is set to -Db / 2. The negative adjusted neg_adj is then added to Rw, Gw and Bw (line ACM 55-58). The Tkdrl sum is denoted as R_low, G_low and B_low. These sums may be the RGB coordinates of the new metamer if the sum does not include a negative component or the Ww adjustment does not cause an overflow or an underflow.

The total value added to the outputs Rw, Gw and Bw of block 130.2 is denoted RGB_adj (line ACM83) and is hard-clamped to 0 at line ACM 97-99 in addition to Rw, Gw and Bw in line ACM 92-94. The result will be the RGB coordinates of the new metamer.

RGB_adj is the sum of neg_adj, pos_adj1 and pos_adj2 calculated in the circuit 1220 mentioned above. The pos_adj1 and pos_adj2 values may be calculated in the " positive-adjustment "circuit 1230 to avoid underflow.

Specifically, pos_adj1 is introduced to correct R_low, G_low and B_low for the underflow. Overflow can be calibrated by scaling 150, so avoiding underflow (i.e., Rw, Gw, Bw or Ww being a negative number) is more important than avoiding overflow. If the minimum value of R_low, G_low and B_low is negative, pos_adj1 is set to the minimum value of the negative value (line ACM 59-60). Otherwise, pos_adj1 is set to zero. That is, no calibration is required (line ACM62). Then, R_low, G_low and B_low are increased by pos_adj1 (line ACM 64-66). The results are shown as R_low_pos, G_low_pos, and B_low_pos.

The pos_adj2 value is introduced to avoid a possible underflow in the ACC block 320. [ Specifically, in line ACM 69-71, the ACC-adjusted RGB output of block 320 is estimated as the sum of (R_low_pos, G_low_pos, B_low_pos) and (Dr, Dg, Db). The sum is represented by (R_low_cc, G_low_cc, B_low_cc). If such a minimum value of the three is negative, pos_adj2 is set to the minimum value of the negative value (line ACM 72-74). Otherwise, pos_adj2 is set to zero (line ACM76).

The calculation of RGB_adj (line ACM 83) and subsequent processing is performed in adjustment block 1240. In line ACM 84-91, the RGB_adj value is adjusted to reduce the probability of underflow of the white value Ww when the metamer adjustment is performed on line ACM95. The white reduction W_adj is initially calculated as RGB_adj / M ₂ (line ACM 84), hard-clamped to the maximum value of Ww (line ACM 85) and the RGB_adj is again calculated as W_adj * M ₂ (line ACM 87).

In particular, line ACM 89-90 represents another embodiment without calibration to prevent underflying. In any case, the new metamer values R, G, B and W are calculated by adding RGB_adj to the RwGwBw output of block 130.2 from line ACM 86 and ACM 92-99 and subtracting W_adj from the Ww output. Then, the coordinates of the new metamer are hard-clamped to a minimum value of zero (line ACM 97-100).

Table 3 shows LUA codes for different ACCMET embodiments. This embodiment calibrates the Ww value taking into account the actual, measured relationship between the white chrominance and the chrominance of the red, green and blue subpixels. Particularly, in the line ACM84 of Table 2, when the white adjustment value W_adj is calculated as RGB_adj / M ₂ , the Ww is reduced by W_adj (line ACM95), and Rw, Gw and Bw are increased by RGB_adj respectively (Line ACM 92-94). This calculation assumes that the chrominance of the white subpixel is the same as the chrominance difference provided by Rw = Gw = Bw, with red subpixels, green subpixels and blue subpixels all having the same values. Actual display is different from this process. In the embodiment of Table 3, this assumption is not applied. In line AMT 34-36 of Table 3, the white reduction amount is represented by Wm, and the value of one white subpixel corresponds to values experimentally determined for one red, one green, and one blue subpixel in the color difference do. These values appear as follows.

RWR_REG / 128, GWR_REG / 128, BWR_REG / 128

Therefore, when the Ww coordinate decreases by Wm (line AMT37), the red, green and blue coordinates Rw, Gw and Bw are Wm * (RWR + 1) / 128 and Wm * (GWR + , Wm * (BWR + 1) / 128 (line AMT 34-36). (Adding 1 is for proper rounding in integer arithmetic.)

This technique of considering the actual, measured relationship between the white chrominance and the chrominance of the red, green and blue subpixels can be applied to the embodiment of Table 2 and other embodiments. In particular, the embodiment of Table 2 is designed to reduce power consumption. The embodiment of Table 3 above does not relate to power consumption. The embodiment of Table 3 above is an embodiment for preventing the ACC 320 from making negative Rw, Gw, Bw and Ww coordinates. In many displays the blue coordinates are selected because the white subpixels (including the various LCDs) emit bluish light and, consequently, the blue coordinates are likely to be negative relative to the red and green coordinates in ACC. (Any negative coordinate outputs by the ACC 320 can be hard-clamped to zero)

Specifically, the embodiment of Table 3 operates as follows. Line AMT2-10 is similar to line ACM3-13 in Table 2 above. The W value in Table 3 corresponds to W_predict in Table 2 above. The W value is not truncated by the upper value, but is hard-clamped to the MAXCOL maximum value in line AMT4 (as opposed to line ACM34). In lines AMT11-13, the Rw, Gw and Bw coordinates are increased by Dr, Dg and Db, respectively, and the results are denoted by Rn, Gn and Bn. The Rn, Gn, and Bn values are thus the expected outputs of the red, green, and blue of ACC 320. The minimum value of Rn, Gn and Bn is calculated as Mn (at line AMT15) and the negative value of Mn is denoted by RGBm (at line AMT18). The remaining steps of Table 3 are performed only if Mn is negative (line AMT16).

In lines AMT24-26, the WBR_REG, GWR_REG and BWR_REG values are calculated from the data values stored in the internal registers. In line AMT29, the white reduction amount Wm is calculated as RGBm / BWR_REG (or RGBm * WBR_REG, where WBR_REG is the reciprocal of BWR_REG.) In line AMT30-32, the Wm reduction amount is a hard- (In line AMT37, when the reduction is applied, Ww can not be negative). In lines AMT 34-36, the Rw, Gw and Bw coordinates are increased by the value calculated from Wm as described above. At line AMT37, the Ww is reduced.

Lines AMT41-52 represent a simpler embodiment in which the red, green and blue coordinates are increased by RGBm. Specifically, the white reduction amount Wm is set to RGBm / M2 (at line AMT42) and hard-clamped to the maximum value of Ww, and when hard-clamping changes the Wm value, RGBm is recalculated as Wm * M2 (lines AMT46) (At line AMT46), the red, green and blue coordinates are increased by RGBm (in lines AMT48-50). Ww is reduced by Wm in line AMT51. This embodiment does not take into account the chrominance relationship between the white subpixels and the red, green and blue subpixels.

Figure 13 shows a display circuit according to another embodiment of the vector method. The data path according to the dither block 270 may be the same as in Fig. 2 or Fig. The output gamma 160 of FIG. 13 may be the same as the output gamma of FIG. 3 for ACC0 tables 220 and 230 only. Table 230 above corrects the white luminance but does not correct the white color difference. The ACC0 tables 220 and 230 can be made by well known methods. For a fully saturated color, the ACC0 table 220 can reliably correct color representation when Wg = 0.

The white color difference can be corrected by an input ACC block 1310 which is designed by the following description. The input ACC block 1310 receives sRGB data R _in , G _in , B _in and outputs sRGB data "RGB" to provide a corrected white luminance for R _in = G _in = B _in . The RGB output of the ACC block 1310 is converted to linear RGB data "rgb" by the input gamma block 130.1 and the output rgb of the input gamma block 130.1 is converted by the block 130.2 into linear RGBW data Rw1, Gw1, Bw1 and Ww1. For example, blocks 130.1 and 130.2 may be the same as in FIG. Also, the input ACC 1310 may be combined with the input gamma 130.1.

Also, the input sRGB data R _in , G _in , and B _in are provided to the input gamma block 130.1M, which may be the same as block 130.1 of Fig. The rgb output of block 130.1 may be converted to sRGB data Rw0, Gw0, Bw0 and Ww0 by block 130.2M. The block 130.2M may be the same as block 130.2 except as described below. The blocks 130.1M and 130.2M are designed to provide accurate color representation for fully saturated colors. Because the corrected color representation for the fully saturated color can be achieved by the ACCO table 220 mentioned above, the blocks 130.1M and 130.2M do not perform any color correction.

The RGBW outputs of blocks 130.2 and 130.2M are provided to interpolator 250. [ The selection input of the interpolator 250 receives the Sat output of block 260. The block 260 may be the same as that of FIG. For example, in some embodiments of FIG.

Sat = Ww0 / max (Rw0, Gw0, Bw0,1) (12)

Alternatively, Sat may be generated as follows.

Sat = Ww1 / max (Rw1, Gw1, Bw1,1)

The interpolator 250 may be the same as in FIG. 2 and may perform linear or non-linear interpolation, among others. In the case of linear interpolation, if Sat is not less than 0 but not more than 1, and in some embodiments, the outputs Rw, Gw, Bw, and Ww of the interpolator 250 are as follows.

Rw = Sat * Rw1 + (1-Sat) * Rw0 (13)

Gw = Sat * Gw1 + (1-Sat) * Gw0

Bw = Sat * Bw1 + (1-Sat) * Bw0

Ww = Sat * Bw1 + (1-Sat) * Ww0

Therefore, if the input sRGB data R _in G _in B _in defines a color that is close to white or white, the value of Ww 0 will be high (e.g., equal to the maximum of R _in , G _in , B _in The Sat will be close to 1 and therefore the Rw, Gw, Bw and Ww values will be close to the white-chrominance-corrected outputs Rw1, Gw1, Bw1 and Ww1 of block 130.2. For saturated colors, Sat is close to zero, so the Rw, Gw, Bw and Ww values will be close to the uncorrected outputs Rw0, Gw0, Bw0 and Ww0 of the block 130.2M.

The digital processing block 210 may include scaling and DBLC operations 150 and 140, for example, as shown in FIG.

The input ACC tables 1310 for the red, green, and blue data provide sRGB data R _in = G _in = B _in to different levels and output gamma values (160). The above table 1310 provides an RGB output such as a white color of ideal or non-ideal color for each sRGB input level R _in = G _in = B _in .

The RGBW generation of block 130.2M may be the same as block 130.2. However, in some embodiments, the block 130.2M generates a low Ww0 (e.g., Ww0 = 0) if the input color is dark, even if it is close to "gray" Here, the "gray " indicates that the color difference is the same as white. This is because dark color quantitative errors are particularly noticeable, as one can better perceive the difference in brightness of dark colors. If Ww0 is set to 0, the RGB coordinates (Rw0, Gw0 and Bw0) will increase and the corresponding subpixels will be brighter. Thus, the quantitative error will not be easy to perceive. The output of the interpolation block 250 will approach the non-ACC-adjusted values Rw0, Gw0, Bw0, and Ww0, as it selects a lower Ww0 for dark colors.

Thus, the Ww0 value is a function of the brightness defined by the output rgb of block 130.1. If the rgb color is a dark color, the Ww0 function will be close to zero or zero. If the rgb color is a light color, the Ww0 function will be approximately equal to the Ww1 function defined by block 130.2. As the rgb color changes from a darker color to a lighter color, the Ww0 function of the block 130.2M will gradually increase from 0 to Ww1 rather than abruptly avoiding artifacts.

FIG. 14 illustrates a less gate-intensive embodiment that performs the interpolation 250 prior to conversion to RGBW. Input ACC 1310, input gamma 130.1 and 130.1M, output gamma 160 and block 210 may be the same as in FIG. The output of the input gamma 130.1 is shown as r1g1b1 and the output of the input gamma 130.1M is shown as r0, g0 and b0. Block 260 calculates the parameter Sat as follows.

Sat = min (r0, g0, b0) / max (r0, g0, b0,1)

Alternatively, the parameter Sat may be:

Sat = min (r1, g1, b1) / max (r1, g1, b1,1)

The output rgb of the interpolator 250 may be:

r = Sat * r1 + (1-Sat) * r0 (15)

g = Sat * g1 + (1-Sat) * g0

b = Sat * b1 + (1-Sat) * b0

Non-linear interpolation may be used. The output rgb of block 250 is converted to RGBW by block 130.2, which may be the same as in Fig. The output of block 130.2 is provided to digital processing block 210. [

The present invention is not limited to the embodiments described above. For example, the present invention may be applied to display devices other than RGB and RGBW. An RGBCW (Red, Green, Blue, Cyan, and White) display device, and the like. In such display devices, the white color difference may be corrected by red, green, blue, and cyan sub-pixels. Further, the cyan color difference can be corrected by the green and blue subpixels. Other primary colors are possible.

In the present invention, the term "color coordinate" sometimes refers to a coordinate axis (e.g., R in RGB or X in XYZ) in the color space and a value of the color coordinate (e.g., &Lt; / RTI &gt; of Rw) can be used ambiguously.

Some embodiments provide a method of processing image data to display an image using a plurality of primary colors (e.g., RGBW, RGBCW, or other primary colors). The primary colors include a plurality of first primary colors (e.g., RGB) and one or more second primary colors (e.g., W).

The method includes obtaining a color signal (e.g., Rs, Gs, Bs, and Ws shown in Figure 2) representing one or more colors in the image. The color signal represents the amount of each of the primary colors in one or more colors. For example, the amount may be Rs, Gs, Bs, and Ws.

The method further includes obtaining a first signal (e.g., Rg0, Gg0 and Bg0) combined with a first calibration of the color signal. The first signal is dependent on each one or more quantities of the first primary color of the representation. For example, the first signal may be dependent on Rg, Gs and Bs or simply Rs or Bs. Alternatively, the first signal may be dependent on other additional parameters such as Ws.

The method further includes obtaining a second signal (e.g., Rg1, Gg1, and Bg1) combined with a second calibration of the color signal. The second signal is independent of the amount of the first primary color of the representation. However, the second signal is dependent on one or more of the second primary colors of the representation. For example, as shown in FIG. 2, Rg1, Gg1, and Bg1 are not dependent on Rs, Gs, Bs, but are dependent on Ws.

The method includes obtaining a relative-amount signal (e.g., Sat). The relative-amount signal represents an indicator indicative of a relative amount of at least one second primary color associated with at least one or more first primary colors. For example, the relative-amount may be a reciprocal of Sat. The reciprocal of Sat is Rs, Gs. Since the relative amount of Ws related to the maximum value of Bs can be plotted, the reciprocal of Sat can also be the relative-amount.

The method includes obtaining a combined calibration signal (e.g., Rg, Gg, and Bg) from the first and second signals and the relative-amount signal. The combined calibration signal represents a combined calibration dependent on the first and second signals and the relative-amount signal.

In some embodiments, the first signal is dependent on the amount of the one or more second primary colors in the representation. For example, in Fig. 2, Rg0, Gg0 and Bg0 are independent of Ws.

In some embodiments, the combined calibration is closer to the first calibration as the relative-amount is smaller. For example, in Fig. 2, Rg, Gg, and Bg become closer to Rg0, Gg0, and Bg0 as Sat becomes smaller.

In some embodiments, the relative-amount is an increasing function for one or more values of one or more second primary colors, and is a decreasing function for one or more values of the first primary colors. For example, in Fig. 2, Sat is an increasing function for Ws and a decreasing function for each of Rs, Gs, Bs.

In some embodiments, the combined calibration signal defines one or more tone values of the non-linear color space for the combined calibration.

In some embodiments, the value of each color is the color coordinate values of the linear color space.

Some embodiments provide an apparatus for displaying an image. The apparatus includes circuitry for calibrating a color signal representing one or more colors within the image. The color signal includes color coordinate values (e.g., Rs, Gs, Bs, and Ws) of each color. The color coordinates are associated with a plurality of primary colors. The primary colors include a plurality of second colors (e.g., Rs, Gs, and Bs) and one or more second colors (e.g., Ws). The color coordinates for each color include first coordinates associated with the first primary color and one or more second coordinates associated with the one or more second primary colors.

The circuit includes a first circuit (e.g., table 220 of FIG. 2) for generating a first signal (e.g., Rg0, Gg0, and Bg0) coupled with a first calibration of the color signal. Wherein the first signal is dependent on one or more of the first coordinate values in the color signal.

The circuit includes a second circuit (e.g., 240) for generating a second signal that is coupled to a second calibration of the hue signal. Wherein the second signal is independent of the first coordinate values in the hue signal and is dependent on one or more of the second coordinate values in the hue signal.

Wherein the circuit is adapted to generate a combined calibration from the first and second signals based on an index of a relative-amount of one or more second primary colors associated with the at least one first primary colors by interpolation And an interpolator (e.g., 250).

Some embodiments provide an apparatus for displaying an image using a plurality of primary colors (e.g., RGBW). The primary colors include a plurality of first primary colors (e.g., RGB) and one or more second primary colors (e.g., W). The apparatus includes circuitry for calibrating a color signal representing one or more colors of the image. The color signals include one or more color coordinate values for each color (e.g., Rin, Gin, Bin in Figures 13 and 14).

The circuit includes a first data path for processing the hue signal and a second data path for processing the hue signal. The first and second data paths comprise a shared circuit. For example, in FIG. 13, the first data path may include an entire circuit except blocks 1310, 130.1, and 130.2, and the second data path includes an entire circuit except blocks 130.1M, 130.2M, and 260 can do. In FIG. 14, the first data path may include the entire circuit except for blocks 1310 and 130.1, and the second data path may include the entire circuit except for blocks 130.1M and 260. FIG.

The first data path may include a first color correction circuit (e.g., a first color correction circuit) including a separate color correction circuit (e.g., a separate table within input gamma 130.1M) for calibrating each value of the plurality of first sets of color coordinates For example 130.1 M or a combination of 130.1M and 160). For example, the first set may be a set {R _in }, a set {G _in }, and a set {B _in }. In some embodiments, each first set may be one or more, less than all color coordinates of each color. In some embodiments, the separate color correction circuits of the first data path may suitably correct the color signals representing the at least one first primary colors to a predefined image quality. For example, in FIG. 13, the table 130.1M may suitably correct the color signals representing at least one of the red, green and blue primary colors to a predefined image quality.

The second data path includes a second color correction circuit (e.g., 1310) that includes separate color correction circuitry for correcting each value of the plurality of second sets of color coordinates. The second set may be one or more, less than all color coordinates of each color. The second data path may suitably correct the color signal representing the at least one second primary colors to a predefined image quality. For example, in Figures 13 and 14, the input ACC 1310 may suitably calibrate the color signal representing white to a desired image quality.

The sharing circuit includes an interpolator circuit (e.g., 250) that interpolates the signal obtained through the first color correction circuit and the signal obtained through the second color correction circuit to produce a combined calibration signal. The interpolation is performed based on an index of a relative amount of at least one second primary colors associated with at least one of the first primary colors within the one or more hues.

In some embodiments, the sharing circuit may include a third color correction circuit (e.g., output gamma 160) coupled to the bottom of the interpolation circuit. The sharing circuit is a circuit for correcting the values of the primary colors (for example, Rs, Gs, Bs).

Some embodiments provide a method of processing image data for displaying images in a screen unit of a display device including a plurality of sub-pixels. Each of the subpixels has a luminance dependent on the state of the subpixel defined by the corresponding subpixel value. The images are displayed using a light source that provides light to the screen unit. The method includes (1) obtaining image signals representing each image using color coordinates, and (2) generating the sub-pixel values from the image signals. Step (2) includes the following steps for each image IM1 of one or more images (e.g., the image "c" in FIG. 8 or 9)

(2) generating scaling-information signals (e.g., INVy _c-1 and / or BL _c-1 ), the scaling-information signal being generated by multiplying the output power Wherein the scaling information comprises information about a plurality of images and information about the image and the method for how image signals representing the image IM1 are scaled corresponding to the output power of the light source, the scaling information being subordinate to one or more images

(2B) scaling the color coordinate values of the image IM1 in accordance with the scaling information to generate a scaled image signal;

(2C) above expressing the color correction for the scaled video signal color of the image IM1 - obtaining a correction signal (for example, Rw, Gw, Bw, and Ww _c), the color-correction signal is at least one sub- Used to provide pixel values

In some embodiments, the scaling information of the scaling-information signal of the image IM1 is obtained from at least a partial representation of the color correction for at least one image (e.g., a preceding image that is image "c-1" have. "At least a portion of the color correction" is output (Rw, Gw, Bw, and Ww _c), or is output (dRw, dGw, dBw and dWw) of ACC (320) of Fig. 9 of the ACC (320) for eight days . Or another expression.

For reference, scaling includes multiplying the values of the color coordinates by a fixed value such as INVy and output power for a particular type of non-uniform scaling, i.e., adjustment of the color coordinate values. For example, the color coordinates may be located in a non-linear color space.

In some embodiments, the step of generating the scaling information signal for the image IM1 comprises generating an un-scale signal (e. G., An un-scaled signal) representative of at least a portion of the adjusted color correction for the at least one image. The input of the circuit 140 of Figure 8 or Figure 9). The adjusted color correction is a color correction that adjusts the output power level independent of the scaling information for one image (e.g., the output power level is the un-scale output of circuit 130.2 The reference level may correspond to BL = 1). The un-scale signal may be generated from the scaling information of at least one image and a partial representation of the color correction for the at least one image.

Some embodiments provide an image data processing apparatus for representing an image on a screen unit of a display device including a plurality of sub-pixels. Each of the subpixels has a luminance dependent on the state of the subpixel defined by the corresponding subpixel value. The image is displayed using a light source for providing light to the screen unit. The apparatus includes circuitry for obtaining an image signal representative of images using color coordinates and generating the subpixel values from the image signals. The circuit includes the following.

(A) a dynamic-light-source-control DLSC circuit (e.g., circuits 810, 140, 804 of FIG. 8 or circuits 810, 910, 140, and 804), the scaling information includes information about output power generated by the light source in displaying the images, and information for scaling the image signals representing the image to match the output power of the light source

(B) a scaling circuit (e.g., 150) that calibrates the color coordinates of the image according to the scaling information provided by the DLSC and generates a scaled value of the color coordinates.

(C) a color-correcting circuit (e.g., 320) that calibrates the scaled value and generates the calibrated scaled value of the color coordinates, the calibrated scaled value is used to provide the subpixel values do.

In some embodiments, the DLSC circuit generates the scaling information in response to the color correction circuit based on at least a portion of the information on the calibrated scaled value.

The display may be the SPR scheme or the SPR scheme.

Some embodiments provide a method of processing image data for displaying the image. The method includes obtaining a color signal comprising values of at least one color of the color coordinates (e.g., Rw, Gw, Bw, and Ww in Figure 6) included in the image. For each of the colors (i. E., One or more of the colors mentioned above, each possible color included in the image, or each subset of colors included in the image), the following processes are performed. From the color signal, a calibrated color signal containing the calibrated values of the color coordinates can be obtained (e.g., Rw, Gw, and Bw outputs of summer 310 and Ww output of circuit 210). The calibrated values incorporate the calibration of one or more color coordinate values of the above-mentioned color according to the transition characteristics of the display device. That is, the calibration includes calibration of Rw, Gw, Bw and other calibrations, but does not include calibration of Ww. From the calibrated color signal, an adjustment signal (e.g., RGB_adj_pcc in Table 1) is generated that includes integrated information regarding the adjustment performed on the calibrated value of the hue. The calibrated color signal is modified with the calibration signal to produce a calibrated calibration color signal (e.g. Rw, Gw, Bw and Ww output of PCC metamer adjustment 604). The calibrated calibrated color signal comprises calibrated and calibrated values of the color coordinates of the hue. Wherein the color coordinates include color coordinates in which at least one adjusted and calibrated value comprises a color coordinate that is greater than the calibrated value and at least one calibrated and calibrated value is less than the calibrated value (e.g., Rw, Gw and Bw can be increased while Ww is decreasing, or vice versa, Rw, Gw and Bw can be decreased while Ww is increasing).

In some embodiments, each color coordinate is associated with a corresponding primary color. The primary colors include a plurality of first colors (e.g., RGB) and one or more second colors (e.g., W). Each second color may be represented by a mixed color of two or more first primary colors. The color coordinates include first color coordinates related to the second colors and one or more second color coordinates relating to one or more second primary colors.

Wherein the color coordinates include a second color coordinate wherein at least one adjusted and calibrated value comprises a first color coordinate than the calibrated value and at least one calibrated and calibrated value is less than the calibrated value. Alternatively, the color coordinates include a first color coordinate, wherein at least one adjusted and calibrated value comprises a second color coordinate greater than the calibrated value, and at least one calibrated and calibrated value is less than the calibrated value .

Some embodiments provide an apparatus for processing image data for displaying an image. The apparatus includes circuitry for processing a color signal comprising color coordinate values of one or more hues included in the image. The circuit includes the following.

(E. G., 240 and 310 in FIG. 6) to obtain a calibrated value of the color coordinates that is interleaved with the color coordinates for each of the mentioned colors, the calibrated value Incorporating at least one correction of the color coordinate values).

(E. G., 604) for determining whether to adjust for the calibrated value and for modifying the calibrated value after adjustment to generate adjusted calibrated values of the color coordinates of each color, Increasing the calibrated value of one of the color coordinates and decreasing the calibrated value of the at least one other color coordinate.

Some embodiments provide a method of lowering image data for displaying one or more images in a screen unit of a display device including a plurality of sub-pixels. Each of the subpixels has a luminance dependent on the state of the subpixel defined by the corresponding subpixel value. The method includes obtaining, for each image IM1 of at least one image, an image signal comprising color coordinate values (e.g., the output of block 130.2) of one or more hues included in the image IM1. The method further comprises generating a scaling-information signal (e.g., INVy and BL). The scaling-information signal includes information about output power generated by the light source in displaying the image IM1 and information about a method of scaling the image signal representing the image IM1 to match the output power of the light source As shown in FIG. The scaling information is dependent on at least one or more images. The method further comprises adjusting the video signal to obtain an adjusted video signal (e.g., by ACCMET 1010). The adjusted image signal expresses the image IM1 using an adjusted value of the color coordinates. The method further includes scaling the adjusted value of the color coordinates of the image IM1 according to the scaling information to generate a scaled image signal (e.g., the output of the scaler 150). The scaled image signal includes scaled values of the color coordinates. The method further includes obtaining a color correction signal (e.g., an output of the ACC 320) from the scaled image signal. The color correction signal expresses color correction for the scaled image signal of the image IM1. The color calibration is a calibration according to the transmission specification of the indication. The color correction signal is used to provide at least one of the subpixel values.

Further, for at least one color of the image IM1, the color coordinates may include at least one color coordinate whose adjusted value is greater than the color corrected by the first value (e.g., RGB_adj in Table 2) At least one color coordinate smaller than the color calibrated by the second value corresponding to the first value. The first and second values are dependent on the scaling information for at least one image.

In some embodiments, the color corrected signal includes calibrated and scaled values of the color coordinates. The adjustment process includes obtaining an approximation (e.g., Dr, Dg, and Db) indicative of at least one calibrated and scaled value associated with the corresponding scaled value. Also, the adjusting step may include adjusting a corresponding value of the video signal such that if the calibrated and scaled value is greater than the corresponding scaled value, the corresponding adjusted value corresponds to the corresponding And if the calibrated and scaled value is less than the corresponding scaled value then the corresponding adjusted value is greater than the corresponding value of the video signal.

Appendix: Additional Methods for Creating ACC Tables

Output gamma measurement

It is common practice to measure the gamma when characterizing a display device. The gamma is the luminance response curve of the linear slope of the linear ramp or digital input values of the voltage. In order to best suit the eyes of the observer in the room, the output gamma response curve should be approximated by a power curve of the 2.2 index. The brightness of the display device is continuously measured with respect to the input point so as to know whether or not it follows the output gamma response curve. The data is used to adjust resistive ladders in the display to make it more consistent with the calibrated curve. Alternatively, the data is used to generate a digital table to calibrate the value to calibrate the display.

Input and output gamma

The 2.2 gamma response of the indication means that all data is sent to the display assuming non-linearity. Within this range, about half of the input values are not mapped to about half the radiance measurement between the brightest and darkest values. It is difficult to calculate these nonlinear values and to obtain calibrated values. For this reason, it usually begins with the conversion of values of all colors into a "gamma pipeline ", which ends up processing these linear values and then converting them back to non-linear values. These transformations are performed using a LUT (Look-Up-Tables), a calculator such as a computer, or a combination thereof. The linear transformation that is the beginning of the gamma root usually uses the "sRGB hypothesis" and the reference designation of a non-linear quantitative assumption in most small computer image files. Finally, converting non-linear values back to sRGB inverse gamma can be done simply. However, if the output conversion to non-linear is to inverse the inverse of the measured display gamma, then the output transform can also correct any abnormal values of the representation. Due to the complete deletion of the abnormal output gamma, the remaining gammas may be perfect sRGB curves.

Gamma of each channel

In the display device, since the gamma response curves of each channel (R, G, B and W in the present invention) are measured separately, there are four different output gamma tables. This is done in ACC, for example, as output gamma 160 of FIG. In the remaining cases, the same tables as ACC0 luminance calibration tables 220 of FIGS. 2 and 3 are used. For other multi-primary displays, different and additional output tables may be used. This is the assumption for the input data, and since it is not an assumption for the output representation, the typical same input gamma curve is used for all of the input channels. In the present invention, conditions for using R, G, B and W luminance measurement and a general program called gamma curve are used. The program corrects missing values using a cubic interpolation algorithm when the entire data set is unavailable. This translates the floating-point measurements into integer values in the normalized range of hardware values that can be stored in the LUT. If necessary, the program may generate an input gamma curve that is an inverse of the limited output gamma curve (which is used in the discussion below).

Chromaticity measurement

The luminescence measurement is also related to a chromaticity meter capable of measuring the chromaticity of the display. It is very common to measure only the chromaticity of each channel under the brightest conditions. From these chroma values and the brightest luminance values for each channel, a matrix can be generated that converts from display values to CIE XYZ triangular stimulus values. It is possible to convert most of the color spaces from XYZ to, for example, sRGB, Lab, Luv, YCbCr and the like.

In the case of a display using three primary colors, such as RGB display devices, the conversion matrix is inversely transformed and used to convert XYZ colors to values suitable for use in a particular display device. This makes it possible in the present invention to display more accurate color representations for the display device or colors obtained from other sources than RGB files. For RGBW displays or other multi-primary displays using three different primary colors, the RGBW to XYZ conversion matrix is not squared and can not be inversely transformed. This is the case with four unknowns but only three equations. One solution is not always possible, and the conversion to RGBW is performed by the same algorithm as in equation (9).

Chromatic tilt measurement

As with the impression of linear algebra and matrix operations, the transformation matrices discussed above have very large assumptions. It is assumed that the matrix is made by only a few brightness measurements and that the chromaticity remains a constant decreasing through the dark values up to black. If the chromaticity measurements are performed on all bright values, you will find that the assumption is not poultry-accurate. The movement of the color difference value as the brightness of the display device changes. Light leakage is a problem that can occur in the dark color. The chromaticity for all the primary colors may show a gray-to-white variation approaching black. This feature may appear and can not be avoided. However, the change of the chromaticity in the bright color is not preferable. This can lead to visual defects that can be seen as intermediate or even gray levels seen as yellow or white against the overall white, even for RGB display. In the case of DBLC (Dynamic Back Light Control), the change in chromaticity to the input level causes full on white, so that while the LCD is calibrated by turning-on and turning-on, The color can be converted by turn-down. The purpose of the ACC of the present invention is to correct for such changes in the chromaticity of the input values. An ACC table for this can be made by measuring the luminance and chrominance over a range of input values for all primary colors.

Measurement conditions

A measurement of luminance and chromaticity is performed on the slopes of the input values of R, G, B, and W, respectively. Typically, the display device can accommodate 256 different input values. It is not necessary to measure all input values, such as the measured brightest, darkest, and required values. Normally, in the present invention, a uniformly distributed measurement is performed every 32 for each primary color, and the 33th brightest possible value is measured. There is always a certain amount of noise in these measurements. Thus, the data is filtered to output an average for such noise. A simple box filter is usually used. After filtering, the present invention interpolates the values to produce a data combination with 256 values. By this, when simulating, it is possible to use values in the normal range as devices and the regular tables such as the input gamma table. Presently, the present invention utilizes cubic interpolation to produce smooth points between the measured values.

The colorimetry of the present invention records the measured values as CIE xyY values. The x and y values are chrominance and the capital letter Y is luminance. The CIE xyY values are designed with linear values, but the input samples of the present invention are effectively non-linearly distributed. For a complete display device, the luminance curves would appear as a 2.2 exponential curve on a simple graph paper. If on an accurate exponential-axis graph paper the luminance curves will look straight. In the gamma route of the present invention, the present invention linearizes the input values using the sRGB input gamma LUT. These are done using the measured data and the samples are placed in a linear space. Since the result is a sparse data combination, it is also interpolated to the entire bit width inside the gamma wiring. Typically, this extends the input values to a width of 12 bits. In this regard, the measurements are still stored in floating point.

However, using the same sRGB input gamma used in the route may not be a suitable method to use. Instead, the present invention obtains the output gamma from each channel and inversely transforms it into input gamma tables (using the same program to generate the output gamma tables mentioned above). Since the output gamma tables are generated from the luminance data, the process must convert the luminance data to a straight line for all channels. Since the present invention uses input and output gamma tables that are approximated by integers, this can be achieved almost exactly. This more closely stimulates the digital hardware of the present invention, but can introduce some quantitative error.

When such a linear process is performed on the color difference data, the position of the sample points is changed so that the position of the sample points is more concentrated at the darker point than the bright end of the tilt. In both non-linear and linear spaces, the present invention can be of constant chromaticity for any input value and horizontal line. Also, the light leakage through the LCD display may change from gray to white at a certain point, as described above. One of the data conditioning steps of the present invention attempts to detect and begin to linearly interpolate the chrominance values from the region that this light emitter starts. This attempt prevents the ACC calibration of the present invention from attempting to correct light gaps in dark colors.

Converts any input color to CIEXYZ

The result of the data condition steps is a combination of tables of given linear color values. The combination of the above tables can find the CIE xyY value. The hardware may generate the CIE xyY value. The measured values are values from separate slopes of R, G, B, and W, but in the present invention xyY values can be converted into XYZ values that can be added together. This makes it possible to predict any color value of 17 million colors that the display device can make. The present invention also measures several other slopes. One of the slopes increases with R, G, and B and the remainder increases with R, G, B, Comparing the combined measurements with the result of adding together the separated R, G, B, and W values ensures that the present invention correctly simulates the sum of the optical colors of the display device with the theoretical sum of the X, Y, and Z values .

Generating the ACC Table

The CIE XYZ graph of the chromaticity of the slope from black to white (all R, G, B and W increasing from 0 to the maximum value) must be a straight line and a line of gradation. This is rarely true, due to the chromaticity shift in accordance with the voltage change of the LCD cell. The slope of all the white slope values is generated by the method of the table mentioned above and converted into CIE XYZ. This produces an uncorrected gradation slope. The purpose of the ACC process of the present invention is to slightly change the RGBW values to return the hue to the desired straight line of gradations. It is possible to generate starting from the desired white point of the CIE XYZ space and a series of tonal values linearly increasing from black to white. 256 values are simply interpolated between black (all zeros) and the white point. Using the measured 100% white points of the display (R, G, B and W for the whole), generate ACC tables containing the smallest possible values. The purpose of the ACC is to linearize the lines of gradation and can also be used to change the color temperature of the display device by selecting a preferred white point, such as D65. If this is done, the magnitude of the values of the final ACC table will either increase above what is required to store or be very large for actual use.

At each uncorrected tone slope value, the present invention can obtain desirable tone values. The uncorrected values are RGBW and CIE XYZ. The preferred tone values are only CIE XYZ. The present invention includes a simple method for converting RGBW to CIE XYZ as described above. However, there is no simple way to reverse it. Therefore, the present invention can not easily convert the desired tone values into RGBW. However, since the present invention can perform inverse transform, if the present invention randomly obtains a good combination of R, G, B, and W, the present invention can confirm whether or not it is close to a desired tone value. Together, the present invention can use the numerical solution techniques to examine the RGBW combination closest to generating the preferred gradation XYZ values. Numerous numerical solution techniques are disclosed in mathematical literature. Most of these numerical solution techniques are designed for solving continuous functions, and the continuous functions widely require the derivation conditions of the functions. The second purpose of many of these methods is to find the minimum values in the smallest possible steps. The present invention does not include a function for the ratio changing from XYZ to RGBW. However, since the present invention is used to perform such a process once and then to make a table for quick access, it does not matter how many steps are taken to obtain the solution. Thus, the brute force method is perfectly suited, and for modern computers this takes only a few seconds.

To numerically find the RGBW combination closest to generating the desired CIE XYZ value, the present invention begins at the initial guess of the RGBW values. One reasonable initial guess would be the uncorrected RGBW value corresponding to the desired gradation slope value. A better initial guess would be the year from close proximity. In the present code, the above two combinations are used. An uncorrected value in the middle of the range is used for the initial guess. From this, attempts to find the best solution are repeated. Thus, this first solution is used as the initial guess to find the next element that goes up or down in the table. The number of iterations is reduced to find the minimum value from this. Each successive solution is used to guess the next solution up or down until the table is complete.

Starting from the initial guess, the present invention converts R, G, B and W to CIE XYZ by simply adding or subtracting 1 and compares it to the desired CIE XYZ value. If the distance from the preferred CIE XYZ value decreases, it is closer to the solution in the present invention. Converting the CIE XYZ values to CIE Lab coordinates and calculating the distances in the color space is one way to generate meaningful distances generally considered (CIE Lab was designed so that a Pythagorean distance of 1.0 is defined for a basic observer "Simply recognizable distance"). In the course of performing a brute force approach, the present invention simply calculates the distance using Pythagoras distance in CIE XYZ space. It is a reasonable and practical way to obtain a close approximate estimate of the solution of the present invention, even though distances in the X, Y, and Z space do not generally take semantics into account. When the distance decreases, simply changing the R, G, B, and W by 1 or adding and shifting in that direction is a way of finding the answer of the present invention as a "walk down hill" G &quot;, &quot; B &quot; and &quot; W &quot; matching the minimum distance and the desired gradation slope value. At present, the ACC methods of the present invention are all indexed by the W value, so the present invention typically keeps W values and causes the solution finding method to go down on the R, G, and B axes.

In conventional displays, the chromaticity of the white slopes typically contained too much blue. Thus, the corrected RGBW values reduce the blue values and slightly increase the red and green values. Since the green has a higher luminance than blue, the result is a slight increase in the luminance curve of the gradation slope. One solution to this problem is to change the W value in the method of finding the solution of the above invention to return the brightness to the original brightness level. Since the table of the present invention is indexed by W, the preferred embodiment of the present invention avoids this. Another solution to the increased brightness after ACC is to find the new luminance curve and change the output gamma table for W to correct.

Converting linear RGBW to XYZ using the table through measurement can only convert values within the gamut range. If the method of finding the solution disappears to one or more end values of the above tables, it means that there is no suitable solution. This situation typically occurs at the bright or dark end of the slope, where there is little room for movement in the lines of gradations. In these cases, the present invention only locates in the gamut or looks for another possible optimal solution.

The result of finding the solution for the slope of all gray values is a table of calibrated RGBW values. The calibrated RGBW values produce a more desirable gradation line when displayed on the hardware. This table is an intermediate step and is used to generate other tables used in some other ACC methods.

Vector ACC Method

In the vector ACC method, the present invention interpolates between the grayscale-corrected RGBW value and the uncorrected RGBW value based on saturation of the hue. The interpolation may be performed after the output gamma, and a table of the calibrated values discussed above may be used if the results are immediately activated through the output gamma tables. However, the two steps are connected to one another and appear as a single table for R, G, and B. This is the ACC1 saturation calibration table 240 of FIG. There may be three tables, but they are all indexed by the same Ws value, so they can be combined into a single table by single address decoding. If it appears as a separate table, it will look like R, G, and B ACC0 luminance calibration tables.

Delta ACC Method

In the delta ACC method, the present invention calculates the RGB distance between the calibrated and uncorrected RGBW gradation slope values. These are stored in the delta ACC chromaticity calibration table 240 of FIG. Several conditions are performed on the delta table before using it. Typically, the delta values approach zero at the bright and dark boundaries of the slope. However, since the error of finding the solution increases near the boundaries, the delta table is often noisy in this area. We often eliminate these noises by clamping the values to zero or lowering the slope to zero near the edges. The remainder of the delta ACC table may also be noise, and the present invention may filter it to soften the discontinuity to create a visual effect on the images. Upon driving, the delta ACC table is indexed with the Ws value and added to all R, G and B values in the route.

Simulation of effect of delta ACC tables

After making the above tables, the present invention drives the EEPROMs in the display hardware and measures the effect on gradation scale gradients among the multiple images. It is possible to simulate the expected effect, at least in black and white, using the tables before the hardware is ready, and it can be verified whether the ACC tables are linearized with the color difference. The first method used is an easy method. The slope of the input values operates on the input gamma table. For the gradation scale slope, it is necessary to simulate the transformation (9) since the R, G, B and W results together are meant for all slopes. The W value is used to index the delta ACC table and the deltas are added to the R, G, and B values. The tables prepared above will be used to convert these values to CIE XYZ and to be converted to the CIE xyY values to be produced by the colorimeter. These predictions are visualized and compared against the measured values. The results are very well matched between theory and practice.

The method produces CIE XYZ directly with linear RGBW values. Since the output gamma tables are not considered, this is a simple method. This is a very reasonable consideration since the output gamma and the display measurements must be inversely transformed each other. Such inverse conversion may be performed with the luminance value and may have no exact effect on chromaticity. To investigate this, a more complete simulation was performed in the present invention. The slope of the input gamma is actuated through the input table to linearize it. The W value is used to index the delta ACC tables and adds the resulting values to the linear R, G, and B values. This summation that adds W is performed for the four output gamma tables R, G, B, and W. [ The results are located in the same non-linear space as the original measurements. Therefore, these values are used to index conditional measurements (before linearization). These measurements are in CIE xyY. The CIE xyY is converted to CIE XYZ and the values can be added together. Thus, the sum can be converted back to xy by the easy method and the comparison and design of the measured data. This can also result in a good match between theory and actual implementation.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the present invention as defined by the following claims. You will understand.

According to the present invention, the hue and chrominance characteristics of a display device can be accurately displayed by adjusting and correcting hues included in an image in the color space.

Claims (57)

delete delete delete delete delete delete delete delete delete A video display device for displaying an image, the video display device including circuitry for calibrating a color signal representing one or more colors included in an image and including color coordinate values for each color,
Wherein the color coordinate values are associated with a plurality of primary colors, wherein the primary colors comprise a plurality of first primary colors and one or more second primary colors, and wherein the color coordinate values of each color are associated with a first Coordinates and second coordinates associated with the one or more second primary colors,
Said circuit comprising:
A first circuit for merging a first calibration of the color signal to be calibrated for the transition characteristics of the display device and for generating a first signal dependent on values of one or more of the first coordinates in the color signal;
Wherein the first color coordinates of the first color coordinate are merged with a second calibration of the color signal to be calibrated for the transition characteristics of the display device and are independent of values of one or more of the first coordinates in the color signal, A second circuit for generating a dependent second signal; And
And interpolating circuitry for interpolating the first and second signals based on at least one of the at least one of the at least one second primary colors associated with the at least one primary colors to generate a combined calibration signal . 11. The display device of claim 10, wherein the first signal is independent of a value of one or more second coordinates in the color signal. 11. The video display device of claim 10, wherein the interpolation is linear. 11. The video display of claim 10, wherein the relative amount is an increasing function for one or more of the one or more of the second coordinates and a decreasing function for one or more values of the first coordinates. 11. The video display of claim 10, wherein the combined calibration signal defines the combined calibration as one or more grayscale values in a non-linear color space. 11. The image display apparatus of claim 10, wherein the color coordinate values for each color are located in a linear color space. 11. The image display apparatus of claim 10, wherein each of the second primary colors can be expressed by a combination of two or more of the first primary colors. delete delete delete delete delete delete delete A method of processing image data for displaying images on a screen unit of a display device including a plurality of subpixels,
Each subpixel has a luminance dependent on the state of a subpixel defined using a corresponding subpixel value and the images are displayed using a light source that provides light to the screen unit
The image data processing method includes:
(1) obtaining initial image signals representative of an initial image using initial color coordinates; And
(2) generating the subpixel values from the initial image signals,
For each one or more images IM1 of the images, the step (2)
(2A) information on the output power generated by the light source and information on how image signals representing the image IM1 are scaled according to the output power of the light source in displaying the image IM1 Generating a scaling-information signal that includes scaling information that is dependent on one or more of the images;
(2B) scaling the color coordinate values of the image IM1 in accordance with the scaling information to generate a scaled image signal; And
(2C) is used to provide at least one sub-pixel value and obtaining a color-correcting signal representing a color correction of the scaled image signal of the image (IM1) . 25. The method of claim 24, wherein the scaling information included in the scaling-information signal for the image IM1 is obtained from at least a partial representation of the color correction for at least one image. 26. The method of claim 25, wherein the scaling information for each image IM1 is dependent on the values of the color coordinates of one or more images prior to the image IM1. 27. The method of claim 26, wherein the scaling information for each image IM1 is independent of the values of the color coordinates of the image IM1. The image processing method according to claim 25, wherein the scaling information for each image IM1 is dependent on the color correction of one or more images before the image IM1 27. The method of claim 26, wherein the scaling information for each image IM1 is independent of color correction for the image IM1. 26. The method of claim 25, wherein generating the scaling-information signal comprises generating an unscaled signal of at least a portion of the adjusted color correction for the at least one image,
Wherein the adjusted color calibration is a calibration for a transition characteristic of the display device, the color calibration adjusted for an output power level independent of the scaling information for the at least one image,
Wherein the unscaled signal is generated from at least a portion of the scaling information of the at least one image and the color correction of the at least one image. 26. The method of claim 25, further comprising, after step (2C), generating a subpixel-value signal representing at least one of the subpixel values for the image IM1 from the color correction signal,
Wherein the step of generating the subpixel values comprises a subpixel rendering process performed using the color correction signal. delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete A method of processing image data to display one or more images on a screen unit of a display device comprising a plurality of subpixels,
Wherein each subpixel has a luminance value dependent on a state of the subpixel defined using a corresponding subpixel value and the images are displayed using a light source that provides light to the screen unit,
For each image IM1 of one or more of the images, the method comprises:
Obtaining an image signal including values of color coordinates of one or more colors included in the image IM1;
Adjusting the image signal to obtain an adjusted image signal representing the image IM1 using adjusted values of the color coordinates;
And a method of scaling the image signals representing the images and the output power generated by the light source in displaying the image IM1 for the images so as to match the output power of the light source Generating a scaling information signal including information that includes scaling information that is dependent on one or more of the images;
Adjusting a color coordinate of the image (IM1) according to the scaling information to generate a scaled image signal including a scaled value of the color coordinates; And
Obtaining a color-correcting signal indicating a color correction for the scaled video signal with respect to the image IM1 from the scaled video signal and correcting according to a transition characteristic of the display device,
The color-correcting signal is used to provide at least one of the sub-pixel values,
For at least one color of the image IM1, the color coordinates are determined by at least one color coordinate whose adjusted value is greater than the calibrated value by a first amount, 2 &lt; / RTI &gt; quantity,
Wherein the first and second quantities are dependent on the scaling information for at least one image. 51. The method of claim 50, wherein for at least one color contained in the image IM1, the adjusted values represent a metamer of the color, metamer) of the image data. 51. The apparatus of claim 50, wherein the color-correcting signal comprises a calibrated and scaled value of the color coordinates,
The step of obtaining the adjusted video signal
Obtaining an approximate indicator of the amount of change in at least one calibrated and scaled value associated with a corresponding scaled value,
And adjusting corresponding values in the video signal. If the calibrated and scaled value is greater than the corresponding scaled value then the corresponding calibrated and scaled value is less than a corresponding value in the video signal and if the calibrated and scaled value corresponds to the corresponding scaling value , The corresponding corrected and scaled value is greater than a corresponding value in the video signal.
delete delete delete delete delete

Images