TWI338512B - Display systems - Google Patents

Description

1338512 IX. Description of the invention: '[Technical field to which the invention pertains] " The present invention relates to an image display system and an image processing method for performing gamut mapping and sub-pixel rendering operations. [Prior Art]

US Patent No. 6,903,754 (the '754 patent), entitled "ARRANGEMENT OF COLOR PIXELS FOR FULL COLOR IMAGING" with simplified sizing of full color φ image devices. DEVICES WITH SIMPLIFIED ADDRESSING)); (2) U.S. Patent Application Publication No. 2003/0128225 (the '225 application) (Application No. 10/278,353), entitled "Improved Sub-Pixel Configuration of Color-Color Flat Panel Display and Improved Modulation Transfer Function Applicable sub-pixel coloring layout (IMPROVEMENTS TO COLOR FLAT PANEL DISPLAY SUB-PIXEL ARRANGEMENTS AND LAYOUTS FOR SUB-PIXEL RENDERING WITH INCREASED • MODULATION TRANSFER FUNCTION RESPONSE)", filed on October 22, 2002; (3) US Patent Application Disclosure 2003/0128179 ('179 Application) (Application No. 10/278,352), entitled "Improved sub-pixel configuration of color flat panel display and sub-pixel coloring layout with separate blue sub-pixels" (IMPROVEMENTS TO COLOR FLAT PANEL DISPLAY SUB-PIXEL ARRANGEMENTS AND LAYOUTS FOR SUB-PIXEL RENDER ING WITH SPLIT BLUE SUB-PIXELS)", filed on October 22, 2002; (4) US Patent Application Publication No. 2004/0051724 (‘724 Application) (Application No. 10/243,094), 1338512

Name "Improved four-color configuration and sub-pixel coloring illuminator (IMPROVED

^ FOUR COLOR ARRANGEMENTS AND EMITTERS FOR

% SUB-PIXEL RENDERING)", filed on September 13, 2002; (5) US Patent Application No. 2003/01 17423 ('423 Application) (Application No. 10/27 8, 3 2 8), Application "IMPROVEMENTS TO COLOR FLAT PANEL DISPLAY SUB-PIXEL ARRANGEMENTS AND LAYOUTS WITH REDUCED BLUE LUMINANCE WELL • VISIBILITY", "Application for the improvement of the sub-pixel configuration of the color flat panel display", October 22, 2002 (6) US patent application

Please disclose No. 2003/0090581 (‘581 Application) (Application No. 10/278,393), entitled “COLOR DISPLAY HAVING HORIZONTAL SUB-PIXEL” with vertical sub-pixel configuration and layout

ARRANGEMENTS AND LAYOUTS)", filed on October 22, 2002; (7) U.S. Patent Application Publication No. 2004/0080479 (the '479 application) (Application No. 10/347,001), the name of the "striped display sub-pixel configuration improvement and Sub-pixel rendering method and system (IMPROVED SUB-PIXEL • ARRANGEMENTS FOR STRIPED DISPLAYS AND METHODS AND SYSTEMS FOR SUB-PIXEL RENDERING SAME)", filed on January 16, 2003, discloses an improved image display price/performance curve New sub-pixel layout. The above-mentioned '225, '179, '724, '423, '581, and '479 published applications, and U.S. Patent No. 6,903,754 are hereby incorporated by reference herein. Repeating groups for a particular sub-pixel having an even number of sub-pixels in the horizontal direction will affect the improved system and technology, such as dot inversion mode and other improvements, as disclosed in the following U.S. patent documents: (1) U.S. Patent Application Publication 1338512

No. 2004/02462 80 ('280 Application) (Application No. 10/456,839), titled "IMAGE DEGRADATION * CORRECTION IN NOVEL LIQUID CRYSTAL DISPLAYS", filed on June 6, 2003; (2) US Patent Application Publication No. 2004/0246213 (the '213 application) (Application No. 10/455,925), entitled "DISPLAY PANEL HAVING CROSSOVER CONNECTIONS EFFECTING DOT INVERSION", 2003 Application on June 6th; (3) US Patent Application

U.S. Patent No. 2004/0246381 (the '381 application) (Application No. 10/45 5,93 1), entitled "System and Method for Performing Point Reversal with Standard Drivers and Backplanes in New Display Panel Layouts" (SYSTEM AND METHOD) OF PERFORMING DOT INVERSION WITH STANDARD DRIVERS AND BACKPLANE ON NOVEL DISPLAY PANEL LAYOUTS)", filed June 6, 2003; (4) US Patent Application Publication No. 2004/0246278 ('278 application) (Application No. 10/455,927) , "The SYSTEM AND # METHOD FOR COMPENSATING FOR VISUAL EFFECTS UPON PANELS HAVING FIXED PATTERN NOISE WITH REDUCED QUANTIZATION ERROR", June 6, 2003 (5) U.S. Patent Application Publication No. 2004/0246279 (the '279 application) (Application No. 10/456,806), the name "DOT INVERSION ON NOVEL DISPLAY PANEL LAYOUTS WITH EXTRA DRIVERS)", filed on June 6, 2003; (6) US Patent Application Publication No. 2004/024640 No. 4 (‘404 application) (application number 10/456,838), the name “LCD backplane layout and non-standard sub-image 1338512

LIQUID CRYSTAL DISPLAY BACKPLANE LAYOUTS AND ADDRESSING FOR NON-STANDARD < SUBPIXEL ARRANGEMENTS), filed on June 6, 2003; (7) US Patent Application Publication No. 2005/0083277 (the '277 application) Application No. 10/696,236), entitled "IMAGE DEGRADATION CORRECTION IN NOVEL LIQUID CRYSTAL DISPLAYS WITH SPLIT BLUE SUBPIXELS", filed on October 28, 2003; and (8) United States Patent Application No. 2005/0212741 ('741 Application) (Application No. 10/807,604), entitled "IMPROVED TRANSISTOR BACKPLANES FOR LIQUID CRYSTAL DISPLAYS COMPRISING DIFFERENT" SIZED SUBPIXELS)", filed on March 23, 2004. The above-mentioned '280, ‘213, ‘38 卜 ‘278, ‘404, ‘277 and ‘741 public applications are hereby incorporated by reference in their entirety. These improvements are particularly pronounced when combined with sub-pixel coloring (SPR) systems and methods that are further disclosed in the U.S. Patent and U.S. Patent Application Serial No.: (1) U.S. Patent Application Publication No. 2003/0034992 (Shen. No. 10/051, 612), the name "CONVERSION OF A SUB-PIXEL FORMAT DATA TO ANOTHER SUB-PIXEL DATA FORMAT", 2002 1 Application on March 16; (2) US Patent Application Publication No. 2003/0103058 ('058 Application) (Application No. 10/150,355), entitled "Gamma Adjustment Subpixel Pixeling Method and System (METHODS AND SYSTEMS FOR SUB- PIXEL RENDERING WITH GAMMA ADJUSTMENT), filed May 17, 2013, 1338512; (3) US Patent Application Publication No. 2003/0085906 (the '906 application) (Application No. 10/215,843), entitled "Using Adaptive Filtering Child image

, "METHODS AND SYSTEMS FOR SUB-PIXEL RENDERING WITH ADAPTIVE FILTERING", filed on August 8, 2002; (4) US Patent Application Publication No. 2004/0196302 ('302 Application) (Application No.) 10/379,767), entitled "SYSTEMS AND METHODS FOR TEMPORAL SUB-PIXEL RENDERING OF • IMAGE DATA", filed on March 4, 2003; (5) US Patent Application Disclosure No. 2004/01743 80 ('380 Application) (Application No. 10/379,765), entitled "SYSTEMS AND METHODS FOR MOTION ADAPTIVE FILTERING", presented on March 4, 2003 (6) U.S. Patent No. 6,917,368 (the '368 patent), entitled "SUB-PIXEL RENDERING SYSTEM AND METHOD FOR IMPROVED DISPLAY VIEWING ANGLES"; and (7) United States Patent • Application Publication No. 2004/0196297 (the '297 application) (Application No. 10/409,413), the name "Using buried pre-subpixel coloring Like the image data set (IMAGE DATA SET WITH EMBEDDED PRE-SUBPIXEL RENDERED IMAGE) ", April 7, 2003 to apply. The above '992, '058, '906, '302, '380, ' 297 applications and ' 368 patents are hereby incorporated by reference in entirety.

Improvements in gamut conversion and mapping are disclosed in U.S. Patent and U.S. Patent Application Serial No. 6,890,219, the entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire content ANGLE 1338512 CALCULATION SYSTEM AND METHODS); (2) US Patent Application • Publication No. 2005/0083341 ('341 Application) (Application No. 10/691,377) 'Name, “From Source Color Space Conversion to RGB W Target Color Space METHOD AND APPARATUS (METHOD AND APPARATUS FOR CONVERTING FROM SOURCE COLOR SPACE TO RGBW TARGET COLOR SPACE)", filed on October 21, 2003; (3) US Patent Application Publication No. 2005/0083352 ('352 Application) (Application) No. 10/691,396), entitled "METHOD AND APPARATUS FOR CONVERTING FROM A SOURCE COLOR SPACE TO A TARGET COLOR SPACE", , October 21, 2003 Application; and (4) US Patent Application Publication No. 2005/00 83 344 ('3 44 Application) (Application No. 1 0/690, 7 1 6), name System and method for converting color gamut (GAMUT CONVERSION SYSTEM AND METHODS), "October 21, 2003 to apply. The above '341, '352 and '344 applications and '' Other advantages are described in: (1) US Patent Application Publication No. 2005/0099540 (the '540 application) (Application No. 10/696,235), entitled "Improved plural mode for displaying displays from multiple input source format image data system

• (DISPLAY SYSTEM HAVING IMPROVED MULTIPLE MODES FOR DISPLAYING IMAGE DATA FROM MULTIPLE INPUT SOURCE FORMATS)", filed on October 28, 2003; (2) US Patent Application Publication No. 2005/0088385 ('385 Application) (Application No. 10) /696,026), the name "System and METHOD FOR PERFORMING IMAGE RECONSTRUCTION AND SUBPIXEL 1338512 RENDERING TO EFFECT SCALING FOR MULTI-MODE DISPLAY", 2003 Application was made on October 28th. The entire contents of these patent applications are incorporated herein by reference. In addition, the entire contents of the following commonly owned and co-pending patent applications are hereby incorporated by reference: (1) U.S. Patent Application Publication No. 2005/0225548 (the '548 application) (Application No. 10/82 1,3 87) , "SYSTEM AND METHOD FOR IMPROVING SUB-PIXEL RENDERING OF • IMAGE DATA IN ΝΟΝ-STRIPED DISPLAY SYSTEMS", April 9, 2004 (2) US Patent Application Publication No. 2005/0225561 (the '561 application) (Application No. 10/821,386), entitled "System and Method for Selecting a White Point in an Image Display System" (SYSTEMS AND METHODS FOR SELECTING A WHITE POINT FOR IMAGE DISPLAYS)", filed on April 9, 2004; (3) U.S. Patent Application Publication No. 2005/0225574 (the '574 application) and U.S. Patent Application Publication No. 2005/0225575 (the '575 application) Named 10/821, 353 and 10/961, 506), respectively, the new sub-pixel layout and configuration of the "High® Brightness Display" (NOVEL SUBPIXEL LAYOUTS AND ARRANGEMENTS FOR HIGH BRIGHTNESS . DISPLAYS)", filed on April 9, 2004 and filed on October 10, 2004; (4) US Patent Application Publication No. 2005/0225562 ('562 Application) (Application No. 10/821,306) ), the name "SYSTEMS AND METHODS FOR IMPROVED GAMUT MAPPING FROM ONE IMAGE DATA SET TO ANOTHER", filed on April 9, 2004; 5) U.S. Patent Application Publication No. 2005/0225563 (the '563 application) (Shen (S) 12 1338512 No. 10/821, 3 88), the name "Improved Sub-pixel-Coloring Filter for High-Brightness Sub-Pixel Layout" (IMPROVED SUBPIXEL RENDERING FILTERS, FOR HIGH BRIGHTNESS SUBPIXEL LAYOUTS)", filed on April 9, 2004; and (6) US Patent Application Publication No. 2005/0276502 ('502 Application) (Application No. 10/866,447), name "INCREASING GAMMA ACCURACY IN QUANTIZED DISPLAY SYSTEMS", filed on June 10, 2004. ® In the patent applications common to these claims, the embodiments of the display system and its method of operation and other advantages are described: (1) Patent Cooperation Treaty (PCT) Application No. PCT/US 06/12768, entitled "New Type EFFICIENT MEMORY STRUCTURE FOR DISPLAY SYSTEM WITH NOVEL SUBPIXEL STRUCTURES), filed on April 4, 2006, and as US Patent Application Publication No. [200丫/八八八八八VIII] is open in the United States; (2) Patent Cooperation Treaty (PCT) Application No. PCT/US 06/12766, entitled "Systems and Methods for Real-Time Low-Color Gamut Mapping" (SYSTEMS AND METHODS FOR IMPLEMENTING LOW- COST GAMUT . MAPPING ALGORITHMS), filed on April 4, 2006, and filed as US Patent Application Publication No. [200Y/BBBBBBB] ('BBB Application) in the US; (3) US Patent Application No. 1 1/ No. 278,675, entitled "SYSTEMS AND METHODS FOR IMPLEMENTING IMPROVED GAMUT MAPPING ALGORITHMS", April 4, 2006 No. filed and published as US Patent Application Publication No. [200Y/CCCCCCC] (‘CCC Application) in the United States 13 1338512; (4) Patent Cooperation Treaty (PCT) Application PCT/US 06/12521

- No., "PRE-SUBPIXEL RENDERED IMAGE PROCESSING IN DISPLAY SYSTEMS", filed on April 4, 2006, and as US Patent Application Publication No. [200Y/DDDDDDD] ( ' DDD application) is published in the United States; and (5) Patent Cooperation Treaty (PCT) application No. PCT/US 06/NNNNN, the name "MULTIPRIMARY COLOR SUBPIXEL RENDERING WITH # METAMERIC FILTERING" , filed May 19, 2006, and published in the United States as a U.S. Patent Application Publication No. [200Y/EEEEEEE] (hereinafter referred to as "Metamer Filtering Application"), incorporated herein by reference. all content. SUMMARY OF THE INVENTION The different embodiments of the display system described and described below have the technical effect of being able to improve the operation of a gamut mapping module that maps input image data designated in the first color | space to be designated in the second A mapped color value of the color space, the second color space being defined by the primary color of the display system "the mapped color value is input to a sub-pixel* coloring operation that produces output image material for the display. " The display system includes a display having at least four color sub-pixels, and inputting image data designated in the first color space and outputting image data designated in the second color space. The display system further includes a gamut mapping module for mapping image data designated in the first color space to image data designated in the second color space. The gamut mapping module clamps the color outside the gamut to black.

14 1338512 The display system further includes a bucket „ _ , ^ β Τ different element for calculating the color value of at least the first primary color of 5 , β γ in the second color space according to the brightness of the image data outside the color gamut. In an embodiment, a display spear crying system includes a display having at least four color sub-pixels, and inputting image data specifying the goods in the first color space and outputting the shadows specified in the second color space.杳虹κ*J^象贝振原色. The display system further includes a gamut mapping module for mapping the image data of the specified A~t meaning in the first color space to the image color specified in the second color space. The gamut mapping module clamps the out-of-gamut color using at least the first and second clamping systems. The first and second clamping systems generate first and second clamp values. The display system further includes A weighting module 'generates from the first value - a wire position value from which a final rounded image value is obtained. In another embodiment, the display system includes a background image having at least four colors And the input of the image data in the first color space and the output of the image data in the second color space. The display system further includes a pre-reduction module for reducing the value of the input image data and Receiving a reduced color gamut mapping module for inputting image data values. In another embodiment, a display system includes a display having a sub-pixel repeating group that includes at least four color sub-pixels and is input at a The image data of a color space and the output of the image data specified in the second color space. The display system further includes a tonable domain mapping module, which repeats the group input parameters according to the sub-pixels of the display. [Embodiment] Here The accompanying drawings constitute a part of this specification, which exemplarily exemplifies an exemplary embodiment and embodiment of the invention* and together with the description of the description « (5 15 1338512 principles of the invention. Figure 1 is an integrated color gamut Image processing system 1 with improved features of the mapping and sub-pixel rendering system. The system circle includes an RGB input. Group 1 〇2, which receives input image data in a variety of possible formats, including but not limited to rgb striped image data and other common digital data formats such as YCbCr. Input image data into input gamma unit 丨04, thereby system i 〇〇 Providing image data in a linear color space. The selective pre-reduction module 106, which will be described in more detail below, can be used to reduce the number of out-of-gamut colors that may need to be clamped to another point in the system.

The appropriate red, green, blue, and white (RGBW) color values for use by the system 1 are calculated by inputting the image data at this point to the calculation W module 1 〇 8 and calculating the RwGwBw module 11 〇 '. For any RGB image data point outside the gamut of the RGBW system, you can use the color gamut clamp 丨丨 2 to select the appropriate gamma gamma value in the color gamut to render the e through a number of clamp possibilities. The choice may represent an optimization of a particular and desired effect [eg, brightness, hue, saturation, etc.]. From these appropriate RGBW image values, the sub-pixel rendering (spR) module 114 can process the image data further to complete any number of "votes. For example, if in the first-display format (eg, RGB stripes, tri-color groups, etc.) 〕Specified the input image (4) 1G2, and the output image data is to be colored in the second display format ^ as many (4) people's (four) towel please (4) multi-pixel projection. The first and second The display format will inevitably occur: the pixel color test (SPR) single A 114 includes hardware and/or software suitable for implementing the sub-pixel coloring techniques described in the above-mentioned plurality of cited patent applications.

16 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 -PIXEL RENDERING WITH GAMMA ADJUSTMENT)], 2005/0225562 (SYSTEMS AND METHODS FOR IMPROVED GAMUT MAPPING FROM ONE IMAGE.DATA SET TO ANOTHER) ] and 2005/022 5 5 63 [IMPROVED SUBPIXEL RENDERING FILTERS FOR HIGH BRIGHTNESS SUBPIXEL. LAYOUTS]. Therefore, the image data can be processed by the output gamma module 161 and the output image data can be transferred to the display panel 来 8 for coloring. The display panel 118 substantially includes a plurality of novel sub-pixel layouts and configurations (NOVEL SUBPIXEL LAYOUTS AND ARRANGEMENTS _ FOR HIGH BRIGHTNESS DISPLAYS) disclosed in U.S. Patent Application Publication Nos. 2005/0225574 and 20050225 5 75. The sub-pixel repeating group of the type. - Gamut Clamping System and Algorithms A variety of color gamut clamping techniques are disclosed in various patent applications cited above. For example, the “Black Clamp” technique reduces brightness; however, it preserves hue and saturation. Figure 9 depicts an embodiment of the black clamp module 900 in the prior invention. The high order bits of the RwGwBw signal are subjected to a 〇R operation [" or " operation] to produce an out-of-gamut (OOG) signal. Such as

17 1338512 The color shirt has no gamut outside the gamut, and the RwGwBw signal bypasses the gamut clamp logic by a set of multiplexers (MUX). When the color has a gamut signal, the maximum value in the RwGwBw value is used to get the reciprocal value from the lookup table ([υτ). The reciprocal value can be pre-computed so that when multiplied by the RwgwBw and w signals, the color is returned to the color gamut. This may create undesirable dark areas in some images, and the method of the present invention focuses on this problem. In summary, there are many possible clamp algorithms and systems for implementing them. Figure 2 is a cross-sectional representation of a color space 200, where line 2〇2 is a color-limited color demarcation allowed in the color space of the display panel of the display system. For points outside these lines (eg, point P), these points are "out of gamut, (〇〇G), and it may be desirable to specify a color point within a color gamut to color instead of the color point p. From point P' A plurality of lines can be drawn toward the color gamut surface of the display system. For example, in a clamP to black, a line can be drawn along the line b toward "black from point P, and b and line 2 〇 2 intersect. The point at which the point is set to the color point of the finger of p. One advantage to the black clamp may be that although in some cases at the expense of some brightness, the newly designated point retains the hue of the color outside the gamut and Saturation, as described in the referenced patent, there are several: Counting cheap methods to calculate the clamp to black, these methods can be built with less gate-level circuits than the previous month's method. Other (four) bit algorithms can also be used. For example, "C-to-Whlte" can be located from p towards "white, leading lines [eg along line w] and line 202). To pay tribute to the za 匕Point to 点 to point P. "Clamp to white, one of::: The point specified by point p will increase the brightness of the color point, which is in the sunlight of expectation = Γ or in other bright environments The condition T is in the viewing image is 18 1338512. Additionally, the intervening clamping system may include "clamp t〇luminance"; where the line L is perpendicular to the gray line from point P. The clamp system may have the advantage that it preserves the brightness of the color points outside the gamut (possibly at the expense of a slight saturation sacrifice). Another intermediate clamp system may include "clamp t〇 diagonal"; it is located at an intermediate point between the clamp toward the luminance and toward the black clamp [along line d and intersect with line 202]. One advantage of diagonal clamping is that it is closer to the minimum distance from P to the gamut housing or edge. The diagonal clamping technique may tend to have better suitability for the hue and saturation of P than to the luminance clamp; however, it produces a brighter color than the color produced by the black clamp technique. Figure 3 is a cross-sectional view depicting the CIEL*a*b* color space. In this figure, the dot is displayed as a color point outside the color gamut, and the white brightness and the black clamp are respectively displayed as lines (more generally, curves) w, l and b and color gamut outer lines/curves 302. Convergence. The line labeled "a" is the specific line/curve that is the theoretical minimum distance between point p and gamut outer casing 302. Although it is possible to calculate the intersection from the line, to the outer boundary 3 () 2, the cost may be large. As an approximate solution to find the low cost of the minimum point, as the inward 2 bit and the brightness clamp The weighted average between them can be more easily calculated as ':angle...the point. The intersection of the line "d" and the outer shell curve 3〇2; this will be described in further detail below. The algorithm is similar to "to black r" g; but the final calculation for w is different. In addition to clamping the w value while the ^ position, it can be bright according to the known input.

19 1338512 Back to w i. In __ implementation, the calculation of the brightness can be adjusted according to the actual sub-pixel layout that can constitute the display honor screen. For example, Figure 19 depicts a number of different sub-pixel layouts that may constitute display 191 of display system 1900. The display system 1 shown here has roughly included an interpolation/sub-pixel coloring (sp_group just 2, a certain time controller 1904' and row and column drivers, respectively 1906, 1908. Display 1910, group sub-pixels' It can further include - sub-pixel repeating group • I 912. This repeating group can be varied, as in the other possible repeating groups 1912 to 1936 shown in the figure. Different hatching indicates different colors, for example, Vertical hatch 1914 represents red, horizontal hatch 1916 represents blue diagonal hatch • 1918 represents green. Dotted diagonal hatch, tighter horizontal hatch or unshaded line. Nothing else - fourth color ' For example, cyan, magenta, white (or no chopper) or yellow. Other sub-pixel repeating groups and other display system architectures are of course also possible and are within the scope of the invention. Although a plurality of embodiments of the invention are Involving GMA (Gamut Mapping Alg〇rhhm), which converts RGB to RGBW, will realize that the technique is applicable to more than three primary colors (eg 4 or more primary colors) For example R, G, B,

Cyan (cyan) and W] multi-primary color system. Moreover, the techniques of the present invention are applicable to display systems manufactured by a variety of manufacturing techniques, including but not limited to lcd, OLED, PDP, with similar efficiencies. For a sub-pixel layout, for example, 1930 which can constitute a repeating group of displays, the following equation for brightness can be established:

Lrgb = (2*Rw + 5*Gw + Bw + 8*W)/16 This equation indicates that the calculated RGBW output brightness must be equal to the source coffee value 20 (S ) 1338512. The equation can be solved in the case where w is known: W1 = (16* τ λ Λ·

Lrgb - ( Rw + 5*Gw + Bw)) / 8 The formula is usually obtained by comparing the value of w with the formula, in the Mard *, p. 1 November. The “field will be brighter” that people see is too dark for those who see the black clamps in these areas. The above equations are easy to implement in hardware or soft (4), = based only on relative brightness. For each Different layouts: = 疋 For example, the formula for the sub-pixel repeating pattern 丨936 can be: W1 - (12* Lrgb - (*RW + 5*Gw + Bw))/4 The difference between the above two formulas is that The result of the relative brightness of w compared to the sum of r and enthalpy in each layout. Additional parameters can be added to the formula to take this relative brightness into account: W1 = (L*M1 J„v.(2*R +5*GH-B)*M2_inv/8)/32 where M1"nv^ M2Jnv may be a constant, as described below in the discussion of the tunable domain mapping (GMA). This may indicate that these constants allow for the formation of a single-GMA hardware module, which can be correctly calculated by pre-calculating and manned values for both the ^, M2_i and other registers. Wi of a variety of different sub-pixel layouts. Diagonal Clamp When performing a clamp to the brightness, some observers may think that the resulting color is too lacking in saturation. The effect of stretching is opposite to the effect of clamping to black, and some observers may think that the result of clamping to black is too dark. A solution to both of these problems may be to calculate the W value to the black clamp and the Wi value to the luminance clamp.

(S 21 1338512 Figure 15 depicts the gamut location module for calculating the intermediate value _ extra-domain (00G) module 502 detects the input rGbw" color if τ a < w 疋 疋 在 in the color gamut: : No, the value itself meets the requirements, and can be bypassed [not shown in the circle = other parts of the color gamut clamp module β to the black clamp module (10), the usual black clamp described in the patent Bit algorithm [page 9 in the back page:] 'Get R, (m value can be the value of the ^ output from the color gamut clamp module. The coffee and value from the module test is also used in the brightness Bit/and 1506' can be used to combine the black clamp w value and the luminance clamp w value into the diagonal clamp module (4) using one of the above-described ones to the luminance clamp algorithm. 'The intermediate value can be calculated by generating the final diagonal character module 1 508 output from the module 1 as a weighted average of the W&W1 values as an example and embodiment. For example, two values can be calculated. In another embodiment as shown in Fig. 10, there is a clamp register 1〇2 in which the values of W and W1 are respectively set. The weighting amount is performed. In this embodiment, the w and W1 values are shifted to the right by different amounts (for example, i, 2 or 3 bits shown in the figure), and then these values are added in different combinations. Up to 1%, 5〇%, 25%, and 12.5% of the combination of 〇%, 50%, 5%, 75%, and 87.5%, respectively. The tool ()^1^ again)1〇〇4 selects one of these options as the clamp diagonal combination w output. This embodiment can be designed to select several useful weighted combinations with a minimum of gates. In other embodiments, there may be more different percentage combinations to select from. In another embodiment, a multiplier module may be used on W and W1 instead of a simple shift and adder' to generate more The continuous percentage range. In Figure 3, a different brightness to the black line L and to the black clamp line b 13 1338512 ^ lies in the w primary of the RGBW display. In an embodiment, I expect this calculation %, and the m B value can be maintained in these steps. ° or more] other primary colors are outside the gamut And has returned pro rata

These calculations can be made to the color and surface. In these cases, only the W of the ifL of the two W values is changed, and the value of the value or any other weighted average gives the color on the color surface. > There are many ways to calculate the clamp register value. Setting or selecting views will effectively produce an average of the two W values. Set or select the value of 75% of the chest, which is closer to the line in Figure 3. One way to calculate the weight ratio of the weights is to use the computer program to calculate the out-of-gamut values of all fertile b offline. The program will take 6 ― 弋 疋 疋 母 母 母 母 找到 找到 找到 找到 找到 找到 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且 并且All of these weighted averages..., into the internal hardware clamp register latch. Medium load (clamP_resister latch)* 〇 As described above, use the d-j ^ ^ ^ ^ in the figure 3, J to the white clamp Line W may be beneficial

In the case of W1 and *J, W can be calculated in a similar manner to the white clamp in the 〇, color clamp, and the value of W is 信 ρ, "V + nr and ww Calculate the weighted average. Between values or in an alternative embodiment, a separate weighted average can be used to replace one of the independent input parameters of the "j[beta] function that varies between different positions. For example, weighting the number of flats. The change, as a function of the input color, the function of the R, G, and B values outside the gamut, or the change of the function of 'changes as the number of inputs. According to the characteristics of the display, for some of the above The function of the combination can be used for different monitors.

23 1338512 So it is different. Features that affect which function is used may include: measured primary chromaticity, original 'color brightness, output gamma curve, backlight brightness or color. One step in finding the desired weighted averaging function would be to measure the display characteristics and compare the independent variables to the results of the above program, which calculates all candidate ideal a-point values for all gamut colors. There may be a direct correlation between an input function and a weighted average, or there may be some approximate mappings that can be used. Once you have selected a function, you can encode it into software or built into hardware. The function can be implemented as an algorithm or pre-computed and stored in a lookup table. Pre-shrinking and subtraction algorithm One possible side effect of RGB to RGBW GMA is to reduce the total number of color states that the display system can produce. There may be up to half of the total number of input colors being mapped to other colors. This is more likely to occur for colors that have both high brightness and high saturation, so that combinations with high brightness and high saturation are not common for natural and unmodified images. ^, it may be desirable to have a pattern where the total number of colors and arrivals of the GMA output match the number of colors. Pre-reduction is a way to achieve this. Figure 4 depicts the prior art in which the fine color gamut is scaled, straight: the wheeled RGB white point maps onto the RGBW white point. The high-brightness + high-saturation color in the shaded area (4) is outside the gamut and must be mapped to the allowed GBW value using the above-described black-bit, free-to-free clamp, diagonal clamp, or other algorithm. Figure 5 shows the pre-reduction through the round-robin value, eventually the entire RGB gamut will fit into the j RGBW gamut. In this case, the step of mapping outside the gamut can be omitted. Although the resulting image may not be as bright as before, in fact more total output H is used instead of all possible 24 1338512 w values, but all possible RGB output values can be used, while some colors are colored This may not be the case outside the domain. In some layouts, such as the 1930, pre-shrinking the input value by half will make the RGB gamut fit inside the RGBW gamut. When the brightness of the w sub-pixel and the brightness of the other three sub-pixels are different, it is necessary to reduce the layout by other amounts. Similarly, pre-shrinking in less than half of the amount will increase the total number of output states to some extent, although some colors will still appear outside the gamut, as shown in Figure 6. Even in displays where the brightness of the W sub-pixels and the brightness of the other three primary colors are equal, the process increases the brightness of the image and is desirable. The figure depicts the effect of an alternative embodiment. In this case, the input RGB values can be pre-scaled until there are no out-of-gamut values. You can then use the normal RGBW GMA to convert to RGBW. Finally, the W value is scaled up by the amount that maximizes the value of w (usually from a bright saturated color close to white). Figure 6 depicts the layout 1930 where the input rgB value has been pre-scaled by half and then the W value has been scaled up by a factor of two. This gives a maximum brightness of about 75% of the maximum possible value. Other combinations produce a brighter maximum, such as reducing the input by less than half. However, these combinations produce more out-of-gamut colors and reduce the total number of output states. Figure 7 depicts an embodiment 7 of the pre-reduction module 106A between the input gamma module and the gamma mapping algorithm (GMa) module. In this embodiment, the percentage is stored as a fixed-point binary number in the pre-reduction register. In one embodiment, the pre-reduction register can be an 8-bit size and can store a number between 〇 and 255 representing a fixed-point number between 〇 and approximately 0.996. Enter each input R after the gamma (the jB value can be multiplied by the pre-shrink value and then divided by 256 (eg >>8) in the right 25 1338512 shift module. Implementation in Figure 7 The example is easy to implement in software, but the multiplier may require a large number of gate stages. The pre-reduction module 106B of Figure 8 is an alternative embodiment that requires fewer levels of circuitry. The figure only describes the path of the R channel; A similar module can be implemented in G and Β β instead of using a multiplier to shift the input value to the right by a different amount. The result is added in different combinations to produce an input of 100% [no reduction], 75 %[25% reduction], 62.5%, 50%, 37.5%, 25%, and 2.5%. In addition to storing a certain number of binary digits in the pre-reduction register storage, you can store the pre-calculated percentage using the MUX selection. The index of one is stored in it. This setting of percentage is just an example. By adding more shifters, adders, and more input multiplexers, you can generate any number of optional reduction percentages. Pre-shrinkage as another pre-reduction Instead of using a fixed amount of reduced input RGB values, the amount used as a function of saturation is not necessarily used. The first graph depicts several different embodiments that can be a suitable saturation function. When the saturation is close to 〇 A function with a value close to 1.0 has the advantage of roughly mapping the input RGB white value to the output RGB W white value, as shown in the is map. This is better than the above pre-shrinking algorithm [may not reach the maximum possible white value] More advantageous. In another embodiment, the 'maximum value may be less than 1. 〇 to reduce the simultaneous brightness contrast error. The saturation function of Fig. 17 will decrease to a certain percentage Pmax when the saturation is maximum. If the Pmax value is greater than the ratio of the brightness of W2 to the sum of the brightness of the R+G+B sub-pixels in the display, some out-of-gamut color will appear as shown in shaded area i 802 of Figure 18. Use 26 1338512 as described above for the gamut-based module. One possible curve for this saturation function is a Gaussian curve such as line 17 ' 1 'but this is computationally difficult to implement in hardware. As line i 70 The line shown in 2 may be suitable 'for example, a piecewise linear function as shown by line 1703 will also produce a satisfactory image. Multiply the value from the function by the input RGB value. Multiplying by 1 · 0 does not cause Reduction with a low saturation input value, multiplied by Pm" 其他 other scores less than 1 will reduce the input value with high saturation. All of these multiplications can be multiplied by a fixed-point binary number followed by Appropriate shifting to the right is implemented in the hardware. Other methods of multiplying by shifting and adding are also included in the scope of the invention. It can be considered that the saturation in Figure 18 is the vertical distance from the gray line. , • The surface of the gamut is often scaled from 0 to 1.0. Although any number of saturation calculations can be used, there are well known approximations in the industry to calculate this value, for example, saturation = (max(r, g, b) - min(r, g, b)) / max(r, g , b) The resulting saturation value can then be used to generate one of the # in the curve shown in Figure 17. For example, a piecewise linear line 1703 having a Pmax value of 0.75 can be generated via the following equation. Pre-shrinking = min(l,1-((saturation_〇·25)/(ΐ-〇.25)))). Then multiply the input red, green, and blue values by any of the above examples. Pre-shrinkage: Pre-reduction G=G* pre-reduction and B=B* pre-reduction

S 27 1338512 Finally, these R, G, and B values can be converted to RGBW by the gamut mapping algorithm. • In another embodiment, the pre-reduction function can also be a function of hue. In the above-referenced patent application, a device for calculating a tone value that can be used for this purpose is disclosed. For example, facial and other skin tones have a very narrow range of tonality, and images with these features preferably have the same pre-reduction function. • In another embodiment, the pre-saturation reduction function can also be a function of luminance. So for a given saturation value, it scales based on the proximity to black instead of using a constant scaling value. This works like a gamma function that allows the output pixel distribution to be shifted closer to [or farther away from] R (JBw & extra-domain. It should also be realized that the pre-reduction function can also be hue, saturation, and brightness. A function of some combination. In the above discussion, an embodiment may have only one pre-reduction function for all primitives and colors. However, it is desirable to have each [or subset] of the input R, G々B primary colors. Separate pre-reduction function. This can increase the ability to perform color correction or adjust the white point of the display. By the public 2|丨 Λ ab knives change the upper left end of the curve 1 703 by the red, green and blue knives [ Decreasing the γ 舳 • 、 、 、 、 、 、 、 、 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第The ancient gossip 1 primary color has separate controls or adjustments that allow adjustment of the color of the mixed color (eg yellow, ...^ Λ ^ Ciba Moonlight, magenta, etc.). For example, 'If red and green have a branch , Pmax with knife open And the green Pmax control is lower than the red pmax control value (10), then the page color point will be shifted toward the red primary color t. Further, if the slope of the curve 1 703 is near pmax, the foot is 28 丄 illusion 12 It is enough to be steep and so can not affect the change. The white point of the display is realized in the yellow. In the first picture, the 'pre-reduction module 1〇6 is located in the input gamma module and the RwGwBw module 11 is calculated. You can also put the pre-reduction module 106 in the other positions of the image processing system. The position is {'J before the input gamma module 1 〇4诵堂 a 逋* has fewer bits, and based on this design, it can advantageously reduce the number of electrical gates in the hardware gate circuit. In addition, the pre-shrinking function and the numerator function phase, .σ can be Perform gamma correction and pre-reduction in the same step. The heart-transfer function is usually implemented as a pre-computed look-up table' so that a higher-level algorithm can be used, such as the sigma curve 1701 of the 17th panel. Don't have to pay the price of more complicated hardware.

Adjustable RGBW GMA

In the above-referenced patent application, some embodiments disclose RGB w GMA with pre-calculated fixed values in the algorithm. These pre-calculated fixed values are based on the measurement or simulation of the primary color and the relative brightness of different sub-pixels. . Embodiments for calculating such values and embodiments of rgb to RGBW GMA with greater flexibility are now disclosed. In one embodiment, the fixed value may be a variable of the temporary (four) towel in the GMA algorithm and/or the hardware design. $ enables a wide range of displays to be used with a single GMA algorithm by changing these values during the initialization phase. These values, specified M0, Μ1, and M2, can be pre-computed and used to simplify the hardware design. A set of possible equations is as follows:

29 1338512 Y + Y + Y. Μη - rgb Y w \Λ 1 · Y ivpw Yr + Y + Yk + γ rgbw L--- IV11 — Y + Y + Yw + Y rgbw Y + Y + YL Γ gb More— Typically, 'MO is the ratio of the sum of the luminances of R, G, and B divided by the sum of the r, G, b, and W exemptions; 1^ is the ratio of the luminance divided by the sum of the luminances of r, g, b, and w; M2 Is the ratio of W brightness divided by the sum of R, G, and B brightness. In a particular layout (e.g., layout 1930), the luminance of the W sub-pixel is approximately equal to the sum of the luminances of R'G and B, so the M2 constant has a value of approximately 1 。. In this case, 'M0 and Ml can be approximated as:

More generally, M0 is the ratio of the sum of the luminances of R, G & b divided by the sum of the luminances of r, g, b &w; Ml is the ratio of the luminance of w divided by the sum of the luminances of R, G, B and W; W brightness divided by the ratio of the sum of the luminances of R, G, and b. The above "M" value is usually a floating point value that varies from slightly larger than 〇 to slightly larger than one. In gamut mapping algorithms that use these values, these values can be used in multiplication and division. In hardware, it is advantageous to store these values as fixed-point binary digits so that integer multipliers and shifting modules can perform these operations. To simplify the hardware design, you can pre-calculate the possible fixed-point binary digits and store them in several registers. These registers can be loaded as part of the display initialization or retrieved from the lookup table (LUT) based on the M2_reg index. Table 1 below describes one embodiment of such a LUT that can be calculated from the useful range of M2 values.

30 1338512 Table 1 M2_REG M2 M0_inv MO—sub Ml reg Ml_inv M2—inv 0000 0.5 48 681 85 96 64 0001 0.625 52 629 98 83 51 0010 0.75 56 584 109 74 42 0011 0.875 60 545 119 68 36 0100 1 64 511 128 64 32 0101 1.125 68 480 135 60 28 0110 1.25 72 454 142 57 25 0111 1.375 76 430 148 55 23 1000 1.5 80 408 153 53 21 1001 1.625 84 389 158 51 19 1010 1.75 88 371 162 50 18 1011 2 96 340 170 48 In this embodiment, M2_REG is the LUT index (binary format) of the fixed-point binary digital value. M2 is the initial floating point value that is not stored in the LUT. MO_INV is the reciprocal of M0, which is calculated by multiplying 1/M0 by 32 to obtain the fixed-point φ binary digital value. MO_sub is the maximum color (this is 1023) multiplied by M0. Ml_reg is the M1 value multiplied by 256 to get the fixed-point binary digital value. Ml_inv is the reciprocal of Ml calculated as a fixed-point binary digital value from 1/Ml multiplied by 32. M2_inv is the reciprocal of M2 calculated as the fixed-point binary digit value multiplied by 1/M2 by '32. It will be appreciated that such a table can be constructed in a variety of ways, and the scope of the invention includes these other embodiments. The pre-multiplied by 256 or 32 can be selected based on the expected value of the constant and the bit width of the internal multiplier. Using this combination of premultiplied values allows all values to fit into the 8-bit register (except for the MO sub that is not a multiplier). 31 1338512 Multiplying by 32 makes it possible to store values greater than or less than 1, but other powers of 2 are also possible. For example, multiplying by 64 adds extra precision but reduces the maximum value that can be stored. Note that not all combinations of fixed-point binary values and their reciprocals are listed above. For example, only M2_inv is calculated, and the value of M2 non-reciprocal is not calculated. This is because the equation in the following GMA algorithm is not multiplied by M2, but is simply divided by M2, so the "non-reciprocal" M2 value is not calculated or stored in a lookup table or scratchpad. Once a set of these fixed-point binary values are obtained from the above table, or a set of these fixed-point binary carry values are calculated and stored in the scratchpad during system initialization, the following virtual code is displayed. Then how can you use the value ___calculate the range of possible V/ values minW=math.floor((math.max(r,g,b)-M0_sub)*Mlinv/32) —minimum possible W minW=math. Max(minW,0) —clamp to zero maxW=math.floor(math.min(r,g5b)*Ml_inv/32) —maximum possible W maxW=math.min(maxW,MAXCOL) —clamp to maximum possible value minW =math.min(minW,maxW) -min must be <= max —calc W from Luminosity, clamped to the possible range L = math.floor((2*r + 4*g + g + b)/8) -luminance approximation W = math.min(L,maxW) --start by setting W to luminosity W = math.max(W,minW) —but clamp it to the max and min allowed -calc RwGwBw from input and WR = math .floor((r-math.floor((W*Ml_reg+128)/256))*M0_inv/32) G = math.floor((g-math.floor((W*Ml a reg+128)/256 ))*M0-inv/32) B = math.floor((b-math.floo r((W*Ml_reg+128)/256))*M0_inv/32) 32 1338512

—check for out-of-gamut (OOG) if (math.max(R,(iB)>MAXCOL) then oog= spr.band(math.max(R,G5B)5MAXCOL) —lower bits of OOG primary inv = math.floor((256*(MAXCOL+l))/(math.max(R,QB)+l)) —inverse value —clamp to black calculation R = math.floor((R * inv + 128)/ 256) G = math.floor((G * inv + 128)/256) B = math.floor((B * inv + 128)/256) W = math.floor((W * inv)/256) -clamp To black value for W —clamp to luminance calculation —calculate the W that reproduces the input luminance W1 = math.floor((L*Ml inv-math.floor((2*R+5*G+B)*M2_inv/8 ))/32) W1 = math.min(Wl,MAXCOL) —do not exceed the max! —clamp diagonal calculation W = math.floor((Wl*clamp_diagH-W3|t(128-cIamp_diag))/128) end —OOG In the virtual code of Table 2, r and b are the input colors after the input gamma correction, and math.floor(_) represents the integer result of any calculated omission of the fractional part, 鲁 math.min(_) and math. Max(_) returns the minimum or maximum of their arguments, spr.band(_) returns the byte-level logic of both arguments and (and ). The term inv is The reciprocal of the gamut (OOG) distance; this value is usually pre-computed and stored in the lookup table indexed by the color 4 field. When the color gamut is clamped, the black clamp value is first calculated, and then the luminance clamp W1 value is calculated. Finally, the clamp_diag value (0 to 128) can be used to calculate the weighted average of the two clamped W values. An illustration of one embodiment of the tonal map mapping algorithm is shown in Figure 14 and Figure 11 shows how Figure 14 calculates L [bright 33 1338512 degrees] and the W value based on luminance in the gamut mapping algorithm. The L value can be approximated as l = (2*r+5*g+b)/8, which is easy to encode on hardware. The MO-sub and Ml_inv values can be used to calculate the minimum and maximum possible W values that do not result in out-of-gamut values. The W value can be set to a value of 1 and then clamped between the MINW and MAXW values. As described in the above-referenced patent application, as long as it is clamped between the MINW and MAXW values, there are many ways to initialize the W value in addition to starting with L. Figure 11 depicts how the M0_inv register 11〇4 and the M1-Kg register 1102 are used to calculate the RwGwBw value from the input RGB value and the w value. In this embodiment: since M0"nv is multiplied by 32 to make it a fixed-point binary: number, the result of the multiplication is then shifted to the right by 5 bits (i.e., 32 = Μ). The resulting RW value may be outside the gamut 'will need to be tested and selectively clamped via a color gamut clamp module such as the η diagram: shown. Post-SPR Filtering In several patent applications referenced above, several methods of sub-pixel rendering (SPR) for images for different display configurations may be described, optionally including sharpening filters. As a step of its order. For example, there may be cross-lighting (cross]uminance_sharpening) and conditions such as metamer_sharping. In the tonable domain mapping algorithm, when the value of (10) is close to 1.0, that is, the ratio of the luminance of w and the ratio of R to the sum of the lengths of 13 are required to be used, it is necessary to use the conditional color illuminating. However, the value of t M2 is not close to the first time '彳 can be used in the embodiment of the cross-brightness, so that the bins can be found (for example, the conditional color and the cross-over) and the result of the use of the result is Weighted average. Figure Q depicts the calculation of the weighted average of the results of applying the sharpening chopper.

34 <S 1338512 One implementation of the method ^ In this figure, Rmeta is a conditional color sharpening result of red 'Rcross is a red cross-brightness sharpening result. The calculation for green and blue will be similar. However, calculations for white may require different weighting coefficients than those used for color; thus separately displaying Wmeta& Wcr〇SSe in order to perform a weighted average of red, M2_c〇L register 1202 may be in M2 MULT unit 1204 Multiply by the Rmeta value. Typically, the M2-c〇L register includes a fixed-point binary digit representing the percentage between 〇 and _%. D M2 - c〇L value can be generated by S at inverter 1208 to produce an "inverse percentage". The inverse percentage of the definition means that the sum of a percentage and its inverse is 1〇〇%. For example, if m2—c〇l - contains 75%, the result of inverting $12〇8 is 25%. In one embodiment Ί* is achieved by subtracting the M2-COL value from 100%, or by a simpler method as follows. The Rcross value can be multiplied by the inverse M2 in another M2-MULT DM. a COL value, then sum the results of the two multipliers to produce the final R value. The calculation for green and blue will be similar. The calculation for white is similar; however, different percentage values are used and the percentage value is used. It can be stored in the M2_WHT register 1206. As just one example, Table 3 provides an embodiment of a virtual code for computing a weighted average of one of the sharpening rate optics results implemented in a software. , 100% fixed-point binary representation is a value of 128, so the operation of the inverter 1208 128 is achieved by subtracting from The virtual code is also described relates crossing dimmer result Gmeta, Gcross, Bmeta Bcross green and blue and calculation, it is not particularly indicated in Fig. 12 TABLE 3

35 1338512 R = R + math.floor(Rmeta*M2_col/128) + math.floor(Rcros*(128-M2_col)/128) G = G + math. floor(Gmeta*M2_coI/128) + math.floor( Gcros*(128-M2_col)/128) B = B + math.floor(Bmeta*M2_col/128) + math,floor(Bcros*(128-M2 a col)/128) W = W + math.floor(Wmeta +M2_wht/128) + math.floor(Wcros*(128-M2_wht)/128) In the above description, the M2_MULT module appears to be a multiplier. Since there are two such multipliers for each color, for a total of eight, such an embodiment would require many gate-level circuits in the hard body. So it would be better if these multipliers could be replaced by simpler circuits. Figure 12 depicts such an embodiment. Instead of acting as a multiplier, the M2_MULT module 1 204 shifts the input filter values to the right by a few different amounts and adds them by the adder into several different combinations. These combinations are selected by the multiplexer 1 3 04. In addition to passing the fixed-point binary percentage value to the M2_INDEX input 1302, the index value can be transmitted to select a percentage value. Note that the index value passed to the MUX can be selected so that the inversion of the indexed φ bit produces the "reverse probability" as described above. This allows the inverter 1208 module of Figure 12 to simply invert the bits to produce an inverse probability; to obtain a simpler operation than subtraction and to reduce the total number of gate stages required to implement. Note that all input states of MUX 1 304 may not be used. In this case, additional states can be introduced by adding more shifters and adders to generate different percentages of state selections that are not used by the MUX. A MUX with more or less states can be used. The shifter can alternatively be shifted to the left by a different amount rather than to the right. This results in a M2_MULT module with higher precision, but a larger 36 1338512 adder and a right shifter are required after the adder of Figure 12. In the tunable domain mapping algorithm based on the above M2 value, when the value of M2 is increased to 1.0, the value of M2_COL should be increased to 100% so that the value of M2__C0L is approximately equal to M2. Within this range of M2 values, the M2_WHT percentage will remain at approximately 100%. When the M2 value is greater than 1,0, the M2_COL value is maintained at approximately 100%, and when the M2 value is increased from 1.0 to other values (eg, 2.0), the M2_WHT percentage will decrease from 100% to 0» for M2 values greater than 1.0, A formula for calculating M2_WHT can be: M2_WHT = 2-M2. These percentages may be stored in hardware in fixed-point binary numbers or may be stored as M2_INDEX 1302 values as depicted in Figure 13. Table 4 below describes the possible values of M2 and the possible M2_COL and M2_WHT values. Table 4

M2_REG M2 M2COL M2—WHT 0000 0.500 11 (50%) 15(100%) 0001 0.625 12 (62.5) 15(100) 0010 0.750 13 (75) 15 (100) 0011 0.875 14(87.5) 15(100) 0100 1.000 15(100) 15(100) 0101 1.125 15(100) 14 (87.5) 0110 1.250 15(100) 13 (75) 0111 1.375 15 (100) 12 (62.5) 1000 1.500 15 (100) 4(50) 1001 1.625 15(100) 3 (37.5) 1010 1.750 15(100) 2(25) 1011 2.000 15 (100) 〇(〇) 37 1338512 In Table 4, for M2 c can be used in M2-MULT module 12〇4 〇L and M2-WHT show that the M2-index value from Fig. 13 is 13〇2. The corresponding percentage obtained is displayed in the fox behind the index value.

Although the above embodiments are described in terms of specific functional modules, operations, or (four) (four) I, the description is not intended to be limited to a particular specific f. 筱 筱 yoke, and those skilled in the art will appreciate that these functional modules and operations can be μ 』 is implemented in any combination of hardware and / or software. For example, you can configure programmable between <inter-level arrays or similar circuits

To achieve such a functional module. In other examples; in the examples, a microprocessor that operates a program in a memory can also implement such a function, and thus, the techniques and specific resources have been described with reference to exemplary embodiments. Those skilled in the art will understand that various changes may be made without departing from the scope of the appended claims, and that equivalents may be substituted for them. The specific situation or material is adopted in the present teaching, and in the case where the Qing can be beyond the basic scope, the second: Xu Xi modified. Therefore, the specific embodiments disclosed herein are intended to be limited to the scope of the appended claims.卜疋38 [Simple Description of the Drawings] The first solid image 1 depicts a block circle of one possible implementation of the image processing system of the present invention. Figure 2: depicts a schematic diagram of a color space in which color points outside the gamut of the special display system can be clamped back into the color gamut along multiple paths. Figure 3. depicts a cross-sectional view of the CIE L*a*b* color space, where color points outside the gamut of a particular display system can be clamped across the stomach back to the color gamut along multiple paths. Figure 4: depicts an area of the color space that clamps the red, green, and blue (RGB) color values back to the area for these colors on displays with RGB and w [white] primary colors. The value is colored. Figure 5: A diagram depicting a color space that pre-scales the RGB color primary colors to reduce the number of input color values outside the color gamut according to the techniques described herein. • Figure 6: depicts a schematic representation of the color space. It scales the effect of the W primary color after pre-reducing the color saturation value. Figure 7. A block diagram depicting one possible embodiment of a pre-reduction system. • Figure 8: A block diagram depicting another possible embodiment of a pre-reduction system. Figure 9 depicts a block diagram of one embodiment of a color gamut clamping system that performs a black clamp. Figure 10 depicts a block diagram of one embodiment of a color gamut clamping system that performs diagonal diagonal clamping.

Figure 1 : A block diagram depicting a portion of one embodiment of a tonable domain mapping system. Circle 12: A block diagram depicting one embodiment of a portion of a sub-pixel rendering system. Figure 13 is a block diagram depicting one embodiment of an element of the sub-pixel rendering system shown in Figure 12. Figure 14: A block diagram depicting a portion of one embodiment of a tonable domain mapping system. Figure 15: A block diagram depicting one embodiment of a color gamut clamp unit. Figure: A schematic diagram depicting a color space that changes the pre-reduction factor to restore the luminance-unsaturated color according to any of the embodiments of Figures 7 and 8. Figure 17: depicts a graphical representation of several different function curves that can be used with saturation-based pre-reduction in accordance with any of the embodiments of Figures 7 and 8. φ Fig. 18 depicts a schematic diagram of a color space that restores the brightest non-saturated color based on the pre-reduction of saturation according to any of the embodiments of Fig. 7 or Fig. 8 t. Figure 19: A block diagram depicting a portion of a display system, including a display panel that is substantially "included as one of several illustrative sub-pixel repeating groups. [Main component symbol description] 40 1338512 102 106 110 114 118 100 image processing system 1〇4 input gamma module 108 calculation W module 112 color gamut clamp module 11 6 output gamma module RGB input module pre-reduction Module s ten nose RwGwBw module sub-pixel coloring module and display panel

Rmeta red conditional color sharpening result Rcross red cross-brightness sharpening result Wmeta white conditional equi-color sharpening result Wcross white cross-brightness sharpening result

41

Claims (1)

1338512 X. Patent application scope A display system for receiving the specified image data and outputting the macro macro. The soil system further known in the second color shirt space further includes: Ί - the display panel includes at least four The original pixel repeating group; '' $ a gamut mapping module for mapping the specified image data to the specified color in the second color hollow mapping module to the color outside the color gamut to the black clamp-computing unit' for at least Partially based on the "places to calculate at least the _--primary color value, the color contains at least three colors $ θ # & contains at least one color image data brightness calculation according to the scope of the patent application! RGB color _ &amp ; " *Menba shirt space' the second fantasy color and the second color space 2 3 According to the application of the 5th patent scope & Λ 2, the color value of the Li color is W, ^, τ' according to the input image of the application of the patent scope - ☆ one of the calculations w, The W 5 group includes Wi = Π 6* ί Lrgb-d5*Gw + £ Wj = ί 1 r rSb ' (*Kw + 5*Gw + R The input image data of the color space, the child of the color sub-pixel of the display ~ The color image of the color space; the color gamut and the color outside the color 包 the first color space of the package includes at least four colors, and the space is based on the input shadow, wherein the first color space I is RGBW color $ to make the at least The brightness of the first original is calculated. • Use the following townships and -, • and ίί 42 1338512 00:-6, 11-η Year and month repair (more) replacement page wi = (L*M1 Jnv_(2 *R+5*G+B)*M2_inv/8)/32. An awarding system' for receiving input image material specified in the first color space and outputting image data designated in the second color space, the display The system further includes: a display panel comprising a sub-pixel repeating group of sub-pixels having at least four primary color colors; The domain mapping module is configured to map the input image color value of the input shirt image specified in the first color space to the shirt color value specified between the second color | the gamut mapping module uses at least a first and a first sweet spot system clamps the out-of-gamut color, the first and second clamp systems generate a first and a second clamp value; and a weighting module 'from the first The second clamp value produces a result clamp value; wherein a final output image value is further derived from the result clamp value. The display system of claim 5, wherein the first clamp is a clamP-to-black system, and the second clamp is a clamp to a redundancy clamp (clamp) -to-luminance) system. ^Please refer to the display system of the fifth scope of the patent, in which the first clamp system. Firstly, it is a clamp-to-black system, and the second clamp system is fully clamped to the white clamp (clamp) -to.-white) system. The display system of claim 5, wherein the weighting module produces the first and a fixed percentage of the second clamp value. According to the display system of claim 5, wherein the weighting module produces 43 8 1338512 ^-----) the replacement page is the average of the first and second clamp values. Ίο, according to the towel, the patent Fan Fan 5 items of the display... Generate a weighted average containing the input shadow #, the middle weighting module ~ the color value function. 11 According to the scope of the patent application, the display system is the color of the image. Child f β where the input shadow = Γ = 一个 one of the groups, the group includes the input image color seven-value-saturation function, the brightness function, and the tone function. 12. The display m is configured to receive input image data specified in the first color space and output image data designated in the second color space, the display system comprising: a display panel substantially comprising at least four primary colors a sub-pixel sub-pixel repeating group, the sub-pixels of the four primary color colors define the second color space; a pre-reduction module to reduce the value of the input image data; and a gamut mapping module to receive the reduced input image Data value. 13 'A display system according to claim 12, wherein the pre-reduction module reduces the input image data value by a fixed percentage. 14 The display system of claim 12, wherein the pre-reduction module reduces the input image data value by a function of input image data tones. 15. The display system of claim 12, wherein the pre-reduction module reduces the input image data value by a function of brightness of the input image data. 16' The display system of claim 12, wherein the pre-reduction module reduces the input image data value by a function of a combination of saturation, hue and brightness of the input image data. 44 171338512 W~S7TT Year Eil (Ai) is replacing the page, according to the scope of the patent application No. 12 of the display system' where: the module reduces the input image as a function of the saturation of the input image data. The material of the display system according to claim 17 of the patent application, wherein the function has a value close to 1 对于 for the input image data saturation value close to 〇, and the ', middle shoulder function increases the saturation value of the input image data. Has a value of ', 1.0. , less than 19, according to the scope of application of the patent scope of the 17th system, including a Gaussian curve. ‘10', the display system according to item 17 of the patent application scope, which includes a straight line. Function package 21, the display system according to claim 17 of the patent application, wherein the segmentation linear function is included. L 22 . A display system according to the scope of the patent application, wherein the letter includes a separate saturation function for each of R, G and B. 23 - a display system 'for receiving specified in the first color space, image data and output specified in the second color space ~ the system further comprises: ""枓" the display a display panel, including sub-pixel repeating groups; and a tunable domain mapping module, based on the configuration of the pixel in the sub-pixel repeating group, wherein the sub-pixel repeating group includes an anti-sub-pixel and the parameter is based on a maximum brightness of the w sub-pixel and the The ratio of the sum of the maximum output luminances of the r, G, and B sub-pixels. 45 1338512 39;~8. 11-year-old repair (more) replacement page 24. The display system according to the scope of patent application No. 23, wherein the parameter is loaded into the scratchpad in the hardware. 25. A display system according to claim 23, wherein the parameter is selected from a look up table (LUT). 46