TW200818926A - Transformation method of a multiple primary color signal and color transformation device using the same - Google Patents

Abstract

A transformation method of a multiple primary color signal is for transforming at least three input grayscale values I_1, I_2, and I_3 of the multiple primary color signal. First, the input grayscale values I_1, I_2, and I_3 are compared, meanwhile among the input grayscale values, I_i is set as a maximal input grayscale value, I_j is set as a medium input grayscale value, and I_k is set as a minimal input grayscale value. Then, three compensation values δI_i, δ I_j, and δI_k respectively for the maximal, medium, and minimal input grayscale values I_i, I_j, and I_k are calculated. Next, output grayscale values O_i, O_j, and O_k respectively corresponding to the maximal, medium, and minimal input grayscale values I_I, I_j, and I_k are determined, such that O_i=I_i+δI_I, O_j=I_j-δI_j, and O_k=I_k-δI_k. Finally, magnitudes of the three output grayscale values O_i, O_j, and O_k are determined, meanwhile one of the three output grayscale values O_i, O_j, and O_k, which is larger than a maximal grayscale value, is set as the maximal grayscale value, or one of the three output grayscale values O_i, O_j, and O_k, which is smaller than a minimal grayscale value, is set as the minimal grayscale value.

Description

200818926,

-* iW2879PA ' IX. Description of the Invention: [Technical Field] The present invention relates to a method for converting a multi-primary color signal and a color conversion device using the same, and more particularly to an input gray scale for converting a multi-primary signal The method of value and the color conversion device using the same. [Prior Art] Conventional displays mostly use display panels with red (R), green (Green 10 G), and blue (Blue; B) primaries. The image data to be displayed is also A digital signal representing the three primary colors, such as a grayscale value, is output to the display so that the corresponding pixel displays the correct color. However, different monitors may have different color rendering capabilities for the same three primary color signals. For example, a mobile phone or a notebook screen with a small color gamut may have distortion of image color saturation, and an gamut with an excitation light diode display. Special, and vice versa. And in different fields of use, users may have different requirements for color preferences. Therefore, the monitor will further process the original two primary color signals, for example, according to the standard sRGB color space published by Microsoft and Hewlett-Packard, and then output and output. The best color is displayed in the color gamut that the hardware can present.

Among them, the color system that is closer to the human eye actually feels HSV space, that is, the representation of hue (H), chroma (Chromaticity or Saturation; S), and brightness (Value; V). So the current technology • will represent RGB

TW2879PA 6 200818926

- Destroy the muscle, rW2879PA % The input signals of the three primary colors are converted to the corresponding HSV space for processing. For example, after the saturation S value is increased, the input signal is converted back to the color space represented by the three primary colors of RGb. However, the conversion relationship between the RGB space and the Hsv space developed according to the definition of color science is a nonlinear transformation. Therefore, the hardware that performs the conversion function inevitably imposes a certain amount of burden, so that the processing efficiency of the signal cannot be effectively improved. In addition, there is also a use of look-up table method to reduce the nonlinearity of the hardware implementation. However, this type of look-up method is a three-dimensional table lookup method. The interpolation method and the memory used are quite large and do not meet the requirements. Cost considerations. Therefore, how to achieve high-efficiency processing of the three primary color signals with less hardware resources and control the display color of the display is the current goal of the industry. ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^, SUMMARY OF THE INVENTION In view of the above, an object of the present invention is to provide a method for converting a multi-primary color signal and a color conversion device using the same. It is defined according to the degree of Hsv space. For example, in the rgb space, the rounded grayscale values of the three primary color signals of the image are directly converted and output to the display panel. In this way, in addition to achieving high-efficiency signal processing operations with more hardware resources, the cost reduction and circuit realization benefits are eliminated, and the multi-primary color signals can be arbitrarily adjusted to obtain == with different color rendering capabilities of the display panel. At least three rounds of grayscale values of the eve, and the color signal (four) yscale

TW2879PA 200818926

:W2879PA value) I__l, I-2, and 1-3. First, determine the size relationship of the input grayscale value, 1-2, and I-3, and set i-i to enter the grayscale value of one of the grayscale values, IJ for this. Enter the grayscale value—the intermediate input grayscale value, and I—k the minimum rounded grayscale value for one of the rounded grayscale values. Next, three compensation amounts of I_i, IJ, and I-k are calculated, and the three compensation amounts are 5I-i, 5IJ, and dl-k, respectively. Then 'determine the maximum rounding grayscale value Ii, the intermediate input grayscale value IJ, and the minimum input grayscale value value k corresponding output grayscale values Oj'Oj, and 〇_k, so that CUh=IJ+(5I" _ 0"=u- 5 IJ, and 〇-k=I~k_ 5 Ik. Finally, determine the size of the output grayscale values 0-i, 0J, and O-k, if 〇-i, 〇", Or 〇k is greater than the maximum grayscale value, then set the output grayscale values 〇i, 0J, or 〇k to the maximum grayscale value, if 0i, 0, or O_k is less than the minimum grayscale value, then Setting such output grayscale values 〇J, 0J, or Ο-k as the minimum grayscale value. According to another object of the present invention, a color conversion device is provided for converting at least three input grayscale values of a multi-primary color signal ( Grayseale, 馨-like) 1", I-2, and 1-3 are the corresponding three output grayscale values 〇—i, OJ, and O—k. The color conversion device includes an input unit, a control unit, a memory unit, and a calculation. The unit and the output unit are configured to receive the multi-primary color signal. The control unit board generates the memory according to the input gray scale values I丨, 1 2, — and IJ3 of the multi-primary signal. Address and selection signal. The memory unit stores at least one one-dimensional lookup table, and outputs corresponding data according to the one-dimensional lookup table, the memory address and the selection signal. The calculation unit calculates the output grayscale values Oj, 0J according to the corresponding data. And OJk. The output unit is configured to output the output grayscale values 〇", 〇", and 〇_k, wherein the input grayscale values 1_1, TW2879PA 8

W2879PA 200818926 I-2, and 1-3 correspond to one color coordinate of the color triangle. The color triangle has several hue regions, corresponding to the input gray scale values][-!, and k3 color coordinates are located in these hue One of the hue areas of the zone. The control unit generates a memory address and a selection signal based on the color coordinates and the hue area. The above-mentioned objects, features, and advantages of the present invention will become more apparent and understood. The conversion method and the color conversion device using the same. It is mainly based on the chroma definition of the HSV space, and is directly converted to the input grayscale value of the input three primary color signals, for example, in the RGB space, and then output to the display panel. In addition, only one-dimensional look-up table method and a small number of multiplication and addition operations are used to achieve high-efficiency signal processing operations with economical hardware resources. Thereby, the chroma of the multi-primary signal can be arbitrarily adjusted, the color distortion caused by different color rendering capabilities of various display panels can be reduced, and the best color can be displayed in the color gamut that the display can present. Referring to Figure 1, there is shown a flow chart of a method for converting a multi-primary signal in accordance with the present invention. Mainly to convert the multi-primary color signal to at least three input grayscale values (8 to 75 〇 & 1 ^ ¥ & 1 \ 16) 1-1, 1_2, and 1 - 3 to achieve the effect of adjusting the color. As shown in FIG. 1 , first, proceeding to step 110, determining the magnitude relationship of one of the input gray scale values I 1 , I 2 , and 1_3, and setting I i — — , for which the gray scale values are input. A maximum input grayscale value, IJ for this input gray TW2879PA 9 200818926,

—* J.W2879PA • One of the order values is the intermediate input grayscale value, and I_k is one of the input grayscale values. The minimum input grayscale value. Next, proceeding to step 120, calculating three compensation amounts of Ij, IJ, and I_k, and the three compensation amounts are respectively (5U, (5IJ, and 5I_k. Then, proceeding to step 130, determining the maximum input grayscale value IJ, intermediate input) The grayscale value IJ and the corresponding input grayscale value Lk correspond to the output grayscale values Ο—i, OJ, and 〇—k, such that 〇—i=I—i+(5 I—i, OJ=IJ-51 ”, and 0 』=1_> (51 〇 • Finally, proceed to step 140 to determine the size of the output grayscale values 〇_i, 0”, and 0_k, if 〇J, 0”, or 0_k is greater than the maximum For the grayscale value, set the output grayscale values OJ, OJ, or ojc to the maximum grayscale value. If Oj, 0J, or 0_k is less than the minimum grayscale value, set the output grayscale values 〇j, 0J, Or 〇_k is the minimum grayscale value, and the multi-primary signal conversion method is ended. This step 140 mainly prevents the display panel from being able to process the multi-primary signal having the abnormal output grayscale value. How to calculate the three compensation amount in step 120 51_卜dj, and 占·I_k, the following series of examples are given as a detailed description, but the text and illustration are only The use of the present invention does not limit the scope of protection of the present invention. In the first embodiment, the first primary embodiment includes a red color (Red; R) and a green color (Green; G). And a blue color (Blue; B), and 1-1, 1-2, and I 3 are input gray scale values corresponding to the three primary color signals, for example, as defined by the definition of HSV space, chroma ( 〇11: 〇111&1;1(:do)€ is TW2879PA 10

TW2879PA 200818926 (I-i-I-k) ' and the degree of v is i-i, and the hue η is also represented by jj, jj, and k, which will be described below.

Referring to Figure 2, there is shown a color triangle in accordance with the present invention. For example, according to the definition of HSV space, take the color triangle plane formed by the hue_chroma of the same brightness as an example, wherein the hue is represented by an angle, and the chroma is represented by the distance from the point w representing white. Further, the color triangle has a six-color phase region p, where ρ = 1, 2 Λ 4, 5, 6, as shown in Fig. 2. Specifically: when P = 1, I - Id - 2 > 1 - 3; when ρ = 2, L2 > Ug I - 3; when p = 3, I - 2 2 I - 3 >L1; When p=4, L3 > g 〇; § p-5 day $ 'I -3 g 1-1 > 1 - 2 ; and when p=:6, I -l > i -3 ^ i 2. The hue of the color located in one of the hue regions 1 to 6 is defined as:; for the hue region U, 5, H,.吟^θ, and 0 is 0°, 120°, 240°, respectively; and for the hue area 2, 4, 6, Η = 60° χ 60. , 180. , 300, 1-1 1-1 k and 0 are respectively

Therefore, it is apparent that the corresponding color change in the HSV space can be controlled by adjusting I-i, and IJc. For example, in order to achieve the effect of merely increasing the chroma c and maintaining the hue and the brightness V, it can be achieved by maintaining u and reducing u and I-k. At this time, the [[i], IJ, and IJc three compensation amounts are set to 5I_i=0, and 51"· and 51-k can also be obtained by one-dimensional look-up table method. Therefore, the first query table is provided, and the first query table includes a plurality of pairs (x, f(x)) 'where the domain of the f function is one of the minimum gray scale values to a maximum that the system can display Grayscale value, and the range of values is the minimum grayscale value to the maximum grayscale value. For example, if the system is 8-bit grayscale display capability, then the f function TW2879PA 11 '

fW2879PA 200818926; domain,: [: '255] and the range of the range is [called Ming Cangzhi, brother 3 map, its function of thousands of thousands - lookup table function diagram: the first real, the example of the function 'and definition The domain scope is [. , 255] and value: heart: for explanation. The first lookup table is a graph representing the f function; ; Qu, ϋ is a set of pairs of numbers (x, f(x)). At this time, for example, to take a cup = Figure 3 is shown. (4) If the table is entered into the checklist, if = set the three compensation amount 5IJ...", it is: δ IJ: Ji-k occupies I_k=f(I J-IJc) xf(UI_k); and (formula 1) In this way, for example, the input gray scale value U of the multi-primary color signal; [_2 and 1_3 are located in the hue area 1 (IJ=I_IJ=I-2, and ^k=&) and 0-3 The hue Ho is 0 ~ -1 - (/1 - 3 - 3). The output gray scale value determined by the compensation amount of Equation 1 is input to the gray scale value I 1 ν I 2 -. The hue Hi of the two's and 1_3 is e〇xLj-^, and is represented by aqj b=I_2^_3, and C=U7 (=a_b), and H〇Hi is · 60x Φ An X (5?_3 + δΙ_3)(ά)- (a + ^_3)(6) (α+没有-3)(α), that is, the output grayscale values 〇_1, 〇_2, and 〇_ 3 Compared with the input gray scales I 1 , I 2 , and I - 3 , the chroma is higher, and the brightness and hue are both TW2879PA 12 200818926 Erlianfuhang·W2879PA unchanged. Or, if the input gray scale values of the multi-primary color signals are 1-1, I-2, and 1~3 in the hue area 2 (^4-2, and, the output gray scale determined by the compensation amount of the formula 1) The values 〇j, 0-2, and 〇~3 have hue Ho of [2 - / 60x 2-/ 3 read -3) - +6 〇, and the input gray scale value ij has the hue Hi system Is 60x'-2 JJ+60, and is _2, and, -2 - / -3 a, = I-2-1-3, b'HU, and C' = I_2 - I_l (= a, -b ,) for the representation, you can get Ho-Hi as: -y } (four)-3) gas 60x seat. (α, + condition one 3) (heart - υ 60 χ (α' + 〇 (α,)) is also available, the output gray scale values 〇 1, 〇 2, and 〇 3 are maintained better than the rounds. For the hue area 3 In the case of ~6, the same effect can be obtained. On the other hand, if only the hue needs to be adjusted to maintain the brightness and chroma, the three compensation amounts - i, 51", and 5I-k can be set as: δ l_i = 0 ; IJ=uxf(I—iI Jj); and δϊ±=0〇 where ιι is equal to 1 or 〇(Equation 2). For example, the input grayscale value 2 and I-3 of the multi-primary signal are located. The hue area ι(ι-i I-i, and j-, and if you want to increase the hue, you can select it. At this time, use the formula 2 TW2879PA 13 work, 200818926

2. Depending on m · iW2879PA, 0-2, and the input grayscale value I 1 , I 2 , the output gray scale value determined by the compensation amount 〇 i hue Ho system is 60 χ ^ 2 + δ Ι ^ = 2 ϊζΙ _ 3 , / 1 — / ο and the hue 卬芍 6 ϋχ 60 χ δ ΐ 2 1 / / Ho-Hi is 3, that is, the output gray scale values 〇j, 0-2, and 〇_3 are more than the input gray scale value I ", 1-2, and 1-3 have a larger hue angle, and the brightness and hue remain unchanged. 1-/

-60, and input the grayscale value j_丨, 〖2, and or 'when the multi-primary color signal input grayscale values u, ^2, and 3 are located in the hue area 2 (L2, IJ=Ij, and 1>;χ 3 ): When the same color increases the hue, u=a can be selected. At this time, the output gray scale values 〇—i, 〇_2, and 〇_3 hue are determined by the compensation amount of Equation 2. / 2-/ 3

The Ho system is 60xm—Μ 1-3 has a hue of 60x|^5^|+60′, and Ho-Hi is 6〇X factory frb °, which also outputs gray scale values 〇_1, 〇—2 And 〇-3 have a larger hue angle than the input grayscale values 1-1, 1_2, and 1_3, and the brightness and hue remain unchanged. Furthermore, using Equation 1 and Equation 2 together can also achieve the effect of maintaining the brightness and adjusting the chroma and hue. At this time, a second lookup table is further provided, which includes a plurality of pairs (x, g(8)), wherein the domain range of the g function is also the minimum grayscale value to the maximum grayscale value that the system can display, and the value The domain range is from the minimum grayscale value to the maximum grayscale value. For example, when the system is 8-bit grayscale display capability, the domain range is defined as [〇, 255] and the range of values is [〇, 255]. To obtain the value of gG-i-I-k), it can be obtained by looking up the table according to the second lookup table.

TW2879PA 14 200818926

TW2879PA So far m compensation amount..., η", and k are: (Π 〇 〇; ~ i - I k xg(I - iI - k) + uxf (I - i_l - k); and 3 I - k = g ( I—iI—k) where u is equal to i or q. (Formula 3) From the above, it can be seen that Equation 3 can be regarded as maintaining the hue and brightness, first inputting the grayscale value, and then maintaining the color. Degree and brightness - color adjustment; when _ sequence or vice versa. And - and the second query table = function graph curve ' can also arbitrarily adjust the position of the maximum value or the nature of the curve 2 so - to 'simple The -vicha table method performs the calculation of the compensation amount to achieve the purpose of _-color, and greatly improves the execution efficiency of the hardware. The younger--the embodiment is in the first-real _ in the three primary color signals with different hues, _ The same look-up table to calculate the compensation amount. However, different display color of various display panels, for example, for the three primary color signals with adjacent 'Hi phase, if the color rendering ability is less accurate, the color of the hue will be completed. Degree, that is, for the different hue areas to provide the corresponding 'to calculate the amount of compensation separately. Therefore, in the second implementation The figure illustrates how the multi-color phase region is compensated for the amount of β 2' in step 12G to achieve the (four) continuous variation and color optimization process of the full color gamut, but the technique of the present invention is not limited thereto. (Green;

Also includes a red color in multiple primary colors (Han ed; TW2879PA 15 200818926

rW2879PA G), and a color (Blue; B) for the sake of explanation, and for convenience of description, I - Bu 1_2, and 1-3 represent the input gray of the three primary color signals of red, green, and blue respectively. Order value. Further, in the second embodiment, color (R), green (G), blue (B), cyan (Cyan; c), magenta (Magenta; Μ), and yellow (Yellow; γ) are The main hue is exemplified as an example, and the six-phase phase is distinguished accordingly. That is, the six-color phase region 丨~6 as shown in Fig. 2. Referring to Figure 4, there is shown a schematic diagram of a hue region and its corresponding look-up table in accordance with a second embodiment of the present invention. Further, in Equation 3, a lookup table is provided for each of the ¥ degree adjustment and the hue system. Therefore, two look-up tables are provided for each of the six main hue for chroma and hue adjustment. That is, mentioning six query tables ΜΗ-3, L_4, L-5, and U, and six lookup tables M", M_2, M_3, M_4, M5, and M_6, and query tables L-i and 屹1 1^ and 12, 1^ and heart 3, 1^- and 1^4: L-5 and M-5, and L-6 and corresponding to R, γΓο, C ΓΒ, and Μ, respectively, as shown in Figure 4 Shown. Where 'query table L-1~L 6 德公拓丨 a (x,f_l(x)) ^ (x,f_2(x)); (x,〇W) $ 6 (=and lookup table μ-i~ M_6 is divided into pairs (x, gl(x)) ^ (x, g_2(x)) > (x, g^3(x)) . (x, g^4(x)) . (χ g 5(χ)). ~ slave 6 function heart two gray scale value to the maximum gray scale value 'and the range of values from the minimum gray scale value to the six primary color phase - the hue of the three primary color signals

Equation 3 in the H embodiment of the compensation amount can be made according to the corresponding lookup table. Of course, the fl~f6 function = for example, using the brother one - the mountain number and the g~l~g_6 function are the same as 16

TW2879PA 200818926

In the case of two consecutive hops, I can easily adjust the color of the graph curve properties of each function to the color of each hue region. For example, how to set a color I in the hue region 1 between the R and gamma phases. For the amount of compensation, the following practices are taken as an example. That is, the color I is decomposed into a color combination on the ruler and the gamma phase, and is calculated by the interpolation method corresponding to the checksum L-Bu M" and L-2, M_2 as the compensation amount" and in the second embodiment In the case of the R and γ hue and the color I with the color I, the color is taken as an example. However, the technology of the present invention is not limited thereto.

Please refer to Fig. 5, which is a color map of the hue area in Fig. 4. Color I has the same chroma color on the adjacent R and Y hue as color S and color E, respectively, and color j is represented by its included u, 匕 2, and 3 components, color I, color S, and The relationship of color E is as follows: 1 = I 2 : / 1 - / 2 T 1-1 3

X / 2- factory 3

X

Where 'I-121-2>1 to 3. The color s and the color E are, for example, the formula 3 of the example of the first example, respectively, according to the corresponding query table for color adjustment, and converted into the color S, and the color E, and then the color I is combined with the color § and E The weight ratio is combined with the color ^,. That is, the set color j is converted to Γ after the color adjustment, as shown in Fig. 5. In addition, for convenience of explanation, let (4)-1-1-3, 41=1-14-2, and d2=l-24-3, that is: Γ=4χ a where / a 3+(ι-"ι ) χ / - receipt) L - L3 + Wlx / Lg a 1 (4), dl + TX / a 1 - w2x / a 2 (4) s · -] a 3 - g a 2 ((

Ul is equal to ❻ or 卜' is equal to ❶ or 卜 in the ^ picture, the color S' is increased by the hue angle of the color S, that is, the 〇 color phase is added TW2879PA ^ 200818926 - the amount of dirty hunger JW2879PA, so ul = 0 . The color E' reduces the hue angle compared to the color E, that is, the component of the G hue is eliminated, so U2 = l. Then, according to the color〗, and the component difference of the color I, the three compensation amounts are obtained (Π J5 1_2, and d 1_3 are: 51-1 = 竺 x (l-u2) xf - 2 (d); dd δ 12 = (5 I__3=yxg_2(d)-|x[ulxf_l(d)-g_i(d)] 〇 Using the above interpolation method, the other hue regions can be treated the same way to the relationship between the lookup tables. , and wait p=l ', then L_ps=L_ls=L ! and L_pe=L_le=L_2 ' M_Ps=M_ls=M_l and M_pe=M”e=:M 2, ie f_ls=f_l and f−le=f_2 , g_is=g丄 and g—le=g—2 ; when P=2 ' L_pe=L_2e=L_2, M_PS=M_2S=M_3 and M_pe=M_2e=^ 2, ie f_2s=f_3 and f_2e=f_2 , g_2s=g__3 and g_2e=g-2; - ' When ^ L3 > L1, iL_ps=L> L3 and L_Pe=L_3e=L_4 ' M_Ps=M_3s=M_3 and M_pe, a 3e==M 4, ie f A 3s=f_3 and f_3e=f-4, g_3s=g-3 and g_3e=g-4; ~ When P=4, L3 >1_2 ...' then L』;=L-4; l $ and L_pe =L - 4e = L - 4 ' M - ps = M - 4s = M - 5 and M_pe = M_4e, 4, that is, f_4s = f - 5 and f_4e = f - 4 ' g_4s = g_5 and g_4e = g - 4 ;~ When P=5 ' L3 g L1 > L2, then 5;^ & and L 』e=L"e=L_6 ' M_ Ps==M_5s=M_5 and M_pe,_5e, 6, ie f_5s=f_5 and f_5e=f_6, g-5s=g-5 and g_5e=g 6 〜’

TW2879PA when P=6 ’ I_1 > I_3 g l_2 ’ then L_ps=L—6s= 18 200818926

—mmm · rw2879PA L_pe=L−6e=L—6, M—ps=M−6s=M—l and Mjpe=M−6e=M−6, ie f-6s=f-1 and f-6e= F-6, g-6s=g-1 and g-. In this way, according to the lookup table L_ps, the lookup table L_pe, the lookup table M_ps, and the lookup table M">e, f_ps(I ii-1 - k) and fjpe (I iiI - k) can be obtained. ), the values of g-pS(ij-i-k), and gjpe(I-iI-k), and set the three compensation amounts ά IJ, 1 IJ, and (5 i_k are: x(l-w2)x /1; • 5 I"—xw2x'—ΜΔ)-hand X[(l-wl)x/—;λί(Δ)υ(Δ)]; and 5 I—k=specific xg je. Wix / one (four) Δ) υ (Δ)]; (Formula 4) wherein ul is equal to 〇 or 1, u2 is equal to 〇 Δ 1 = 1 > 1", and Δ 2 = I" - IJc. In addition, if it is considered to maintain the three primary color signals of a solid color, the display system is an 8-bit gray scale display capability, that is, any of the input gray scale values I-1, I-2, and 1-3 of the three primary color signals is 〇 or 255, the color compensation of the three compensation amounts (H-i, 51) and k at the same time in Equation 4 may cause a solid color distortion. A third look-up table is provided, and the third query table is respectively Including a plurality of pairs of numbers (x, h(x)), wherein the domain of the 11 function ranges from a minimum grayscale value to a maximum grayscale value, such as [〇, 255] in an 8-bit grayscale display system, and The value range is [0, η. - Please refer to Fig. 6, which is a functional representation of the second lookup table in accordance with the second embodiment of the present invention. As shown in Fig. 6, the h function is a nonlinear function. ^, and the domain range is [〇, 255] and the range of values is 丨]. The input value of this TW2879PA h function is greater than ^ equal to Μ and less than or equal to 192 200818926

When the two numbers are .i, W2879PA, the resulting output value is 1 ' and h(25 5)=11(0)=0. So far, according to the third lookup table, obtain the value of 11 (111^1[255-:〇,1一幻), min[255-I-i, I-k] is (255-1—i) and IJk The minimum value, the brigade setting three compensation amount, 5IJ, and 5I_k are: 5 I - i = h (min [255 - I - i, I - k]) x ^ > x (i - "2) χ / - %(△); Hand X[(l-ta)x厂ΜΔ)υ(Δ)]}; and Δ • (5 I^k=h(min[N-IJJ^k])x{ Δ |x[ Wlx/_MA)U^0]} ·, Δ (Equation 5) where ul is equal to 0 or 1, 112 is equal to 〇 or 1, eight = 1 less 1] <;, Δ 1 = 1" -1", And Δ 2 = I" - I - k. Then 'if it is the three primary color signals of pure color, that is, I~i==255 or I' with the three compensation amount in formula 5 (51 Ii, (5 IJ, and 5 corresponding to φ output grayscale value 〇一i , 〇", and 0 ak, obviously also a solid color. And the main, taking the graphical curve of the h function shown in Figure 6 as an example, when j } is very high or I k is very low, h (min[ 255-I-i, I-k]) are sharply sharply reduced, that is, for the high-color three primary color signals, the compensation amount i of the maximum input grayscale value I-i is suppressed without causing chroma oversaturation. The graph curve of the lun function is the horizontal line system corresponding to 1 in most of the release fields. It means that the compensation for the non-pure three primary color signals does not compensate for IJ. Of course, the range of ^ to Μ can also be seen by the color rendering capability of the display or Use the requirements for adjustment.' Below is a description of the application of the second embodiment of the shield TW2879PA 20 200818926

The color conversion device of the conversion method of the Zhao Mahang-.i'W2879PA number, but the technique of the present invention is not limited thereto. Referring to Figure 7, there is shown a color conversion device in accordance with the present invention. The color conversion device 700 includes an input unit 71A, a control unit 720, a memory unit 730, a calculation unit 74A, and an output unit 75A. The input sheet 70 71 is used to receive the multi-primary signal s. The control unit 720 generates the memory address S1 and the selection signal S2 based on the input grayscale values l丨, JJ2, and 3 of the multi-primary signal S. The control unit 72 includes, for example, an address generator 720a coupled to the input_unit 71 and a classification control controller 720b. The memory unit 730 includes a plurality of memories 730(1) to 730(6) and a memory interface 731, and each of the memories 73〇(i) stores a one-dimensional lookup table L_i and M_i, i=i~6. As described above, according to the definition of HS V space, the input gray scale values I_1, I-2, and I-3 are one color coordinate corresponding to one of the color triangles. The classification controller 720b is set with a definition rule of the hue area, such as the six-color phase area distinguished by the above-mentioned RGBCMY six φ main hue. Thereby, the classification controller 720b generates a selection signal based on the hue region corresponding to the color coordinates, and the memory interface 731 accesses the portion of the memory 730(1) to 730(6) according to the selection signal S2. For example, for the color belonging to the hue region in Fig. 4, the memory interface 731 is accessed by selecting the memory 730(1) and 730(2) according to the selection signal S2. In addition, the address generator 720a generates the memory address S1 mainly according to the color coordinates of the HSV space corresponding to the grayscale values 1_1, I-2, and 1-3. The memory interface 731 is based on the memory. Body address TW2879PA ... 200818926

Erda lion muscle.jfW2879PA • si searches for the corresponding data S3 in the look-up tables L-1, Μ-::ι, L-2, and Μ2 stored in the memory 730(1) and 730(2). The corresponding data S3 includes, for example, fj (l-iI-k), g-l (I-iI-k), f-2 (Ij-7k), and g-2 (I-k), and each of the formulas 4 The parameters ul, u2, Δ = ι - Δ 1 = 1 i - IJ, and Δ2 = Ι" - Ι - k, and the like. At this time, the calculation unit 740 calculates the output grayscale values Οi, OJ, and Ojc by Equation 4 based on the corresponding data S3. The output unit 750 is configured to output a plurality of primary color _ signals S' converted to the output gray scale values Οi, 〇j, and 〇k. However, those skilled in the art to which the present invention pertains can clarify that although an embodiment of the present invention is exemplified by an RGB CMY six-color phase region which is easily recognizable to the human eye, it is obvious that it is possible to reselect different six main hue or distinguish as needed. More hue areas. For example, the sixth hue of the two main hue phases is further increased by the sixth main hue in Fig. 4 to be divided into 12 hue regions, and the lookup table is added correspondingly. At this time, the color in each hue region can also be decomposed into the color mixture on the adjacent two main hue, and the corresponding amount can be calculated as the compensation amount. In addition, as shown in Equations 4 and 5, the adjustment of the interpolation method can be used to verify that the brightness of the primary color signal is reduced. At this time, the value of U can be changed to enhance the balance of the luminance. The color conversion device can also be simply matched with the (4) conversion formula to change the value of each parameter of the memory and the lookup table in the memory unit. As described above, the term "controlling the color" is not necessary for the purpose of controlling the color as long as the original color is called the direct conversion of the color of the person. TW2879PA 〇〇 200818926,

^.rmw/u * TW2879PA The multi-primary color signal conversion method and the color conversion device using the same according to the above embodiments of the present invention are mainly based on the chroma definition of the HSV space, for example, directly converting the three primary color signals of the image in the RGB space. The grayscale value is input and output to the display panel, and only a simple one-to-one look-up table method and a small number of multiplication and addition operations are used, and the evaluation is still within the appropriate complexity. In this way, high-efficiency signal processing is achieved with more economical hardware resources, which reduces costs and meets circuit implementation benefits. In addition, the color presented by any of the multiple primary color signals can be reduced to reduce the color distortion caused by different color rendering capabilities of various display panels, and the best color can be displayed in the color gamut that the display can present. Of course, the display system using other multi-primary signals can also perform the conversion processing on the round grayscale values of the multi-primary color signals according to the conversion relationship with the HSV space as in the above two embodiments. The invention has been described above by way of a preferred embodiment, and is not intended to limit the invention. Those having ordinary skill in the art to which the invention pertains may make various changes and refinements without departing from the spirit and scope of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims.

TW2879PA 23 200818926

一顾顾肌·rW2879PA '[Simple Description of the Drawing] Fig. 1 is a flow chart showing a method for converting a multi-primary signal according to the present invention. Figure 2 is a schematic illustration of a color triangle in accordance with the present invention. Figure 3 is a diagram showing the function of the first look-up table in accordance with the first embodiment of the present invention. Figure 4 is a diagram showing the hue area and its corresponding look-up table in accordance with the second embodiment of the present invention. • Figure 5 shows a color exploded view of the hue area 1 in Figure 4. Figure 6 is a diagram showing the function of the third look-up table in accordance with the second embodiment of the present invention. Figure 7 illustrates a color conversion device in accordance with the present invention. [Main component symbol description] (none) TW2879PA 24

Claims (1)

200818926 Sanda number: TW2879PA X. Patent application scope: 1. A multi-primary color signal conversion method for converting at least two gray-scale values (graySCale value) I j, 1-2, and I 3 of the multi-primary color signal The conversion method includes: (a) determining the input gray scale value I1; [, 〗 1-2, and I 1-3, and setting IJ as one of the input gray scale values, the maximum input gray scale The value, IJ is the grayscale value of one of the rounded grayscale values, and I_k is the minimum input grayscale value of one of the input grayscale values; (b) Calculating I-i, I", and I -k three compensation amount, the three compensation amounts are j5 IJ, and 5 I_k; (c) determining the maximum input gray level value Ij, the intermediate input gray level value IJ, and the minimum input gray level value I_k Corresponding output grayscale values 〇j, 〇j, and Οk, such that 〇i=I—i-(Π—i, 〇J=I”-5 I”, and 〇—k=I—k-( 51 a k ; and (4) determine the size of the output gray scale values OJ, 〇 J, and 〇 _k, _ if 0 - i, 0 J, or 〇 - k is greater than a maximum gray scale value, then set the outputs Gray scale value〇 0", or 0-k is the maximum grayscale value. If 〇", 〇", or OJc is less than a minimum grayscale value, the output grayscale values 〇j, 0J, or Ojc are set to the minimum grayscale 2. The conversion method of claim 1, wherein the multi-primary color system comprises a red color (Red; R), a green color (Green), and a blue color (Blue; B) ° TW2879PA. 25 200818926 达达编号♦ TW2879PA 3. The conversion method of claim 1, wherein the step (b) comprises: providing a first lookup table comprising a plurality of pairs ( x, f(x)), wherein the domain of the f function ranges from a minimum grayscale value to a maximum grayscale value, and the range of values ranges from the minimum grayscale value to the maximum grayscale value. The conversion method of the third aspect, wherein the f-function _ number is a non-linear function. 5. The conversion method of claim 3, wherein the step (b) further comprises: according to the first query The table obtains the value of f(U-I_k). 6. The conversion method according to claim 5, wherein the step (b) is further These include: • Setting the compensation amounts, 51”, and 5I_k are: 5 IJ=0; 5 I”=uxf(I—iI_k); and δI_k=0° where u is equal to 1 or -1. The conversion method according to claim 5, wherein the step (b) further comprises: setting the compensation amounts 5 IJI, 5IJ, and (5I_k is: TW2879PA 26 200818926 Sanda number: TW287卯A δ I_i= 〇; ; and 5 I-Bu f (I ii-k). 8. The conversion method of claim 5, wherein the step (b) further comprises: providing a second lookup table comprising a plurality of pairs of numbers 10 (x, g(x) Wherein the range of the g function is a minimum gray scale value to a maximum gray scale value and the range of values ranges from the minimum gray scale value to the maximum gray scale value; according to the second lookup table, g(I_i) is obtained -IJc); and setting the compensation amounts (Hj, 5IJ, and 5I_k are: 5 I - k = g (l - iik). Where II is equal to 1 or _1 〇 9 · as claimed in claim 8 The conversion method of the item, wherein the g function is a non-linear function. The conversion method of claim 1, wherein the conversion method further comprises: providing six lookup tables L-1, L2 , L 3, L 4, L 5, jL6, —~ — — — — — — and six query tables—1, M-2, M-3, M-4, M-5, and M-6, Some of the lookup tables L-1 to L-6 include a plurality of pairs (x, f_i(x)), TW2879PA _ 200818926, three numbers: TW2879PA (x, f-2(x)), (x, f- 3(x)), (x, f-4(x)) , (x, f - 5 (x)), and (x, f - 6 (χ)), and the query tables M1 to M-6 respectively include a plurality of pairs (x, g - l (8)), ( x, g - 2 (8)), (x, g - 3 (8)), (x, g - 4 (x)), (x, g - 5 (χ)), and (x, g - 6 (x)), wherein The domain of the f-1 to f-6 and g-1 to g-6 functions ranges from a minimum grayscale value to a maximum grayscale value, and the range of values ranges from the minimum grayscale value to the maximum grayscale value. 11. The conversion method according to the first aspect of the patent application, wherein the step (a) further comprises: determining the hue regions of the input gray scale values I-1, 1-2, and 1-3, wherein ρ =1, 2, 3, 4, 5, ό, and when P=1, I~12 1~2; 乂3, then L_ps=L-1S=L~1 and [一卩6吐一16= 1^一2']^,=]\4一13=:]^1一1 and]^1』6=^116=]^2 ie f-ls=f-l and f_ie=f- 2, g - Is = g" and g - ie == 2 · When ρ = 2, I - 2 > I - 1 2 I - 3, then L - ps = L - 2s = = I 〇 and • L_pe = Lj2e = L - 2, M - ps = M - 2s = M - 3 and ] VI - pe = Mj2e = M 2, that is, f - 2s = f - 3 and f - 2e = f - 2, g - 2s = g - 3 and g - 2e = g 2 ; when p = 3, 1-2 2 I -3 > I -1, then L - ps = L - 3s = = l ~ 3 and ! ^) 河一3柯一4,]^1』3=]^1_35=^" and &gt ;1,=]^一36=^4, that is, fL3S=f^3 and g-3s=g-3 and g-3e=g-4; when p=4, I-3 > I 2 2 I -1, then L - ps = L - 4s = = L ~ 5 and L - pe = L - 4e = L j, Mjps = M - 4s = M - 5 and M - pe = M 4e = lV [ 4 that is, f - 4s - f - 5 and f - 4e = f - 4 ' g - 4s = g - 5 and g - 4 ; when P = 5, 1 - 3 - I - 1 > I - 2, then And TW2879PA 28 200818926 Sanda number: TW2879PA L-pe-L-5e=L-6' Μ ps^M-5s=M-5 and M-pe=M 6, ie f-5s=f-5 F-5e=f-6, g-5s=g-5 and g-5e=g-6; and when P=6, I-1 > I-3 g I-2, then L_ps=L- 6s=Lj and L-Pe-L-'M-ps=M-6s=M-1 and M-pe^M6e=M6, ie f-6s=f-1 and f-6e=f-6 , g-6s=g" and g-6e=g-6. 12. The conversion method as described in claim n, wherein the step (b) further comprises: φ obtaining f according to the lookup table L_ps, the lookup table Ljpe, the lookup table M_ps, and the lookup table M-Pe The values of js(l-iI-k), f-pe(I>I-k), g-ps(i-H~k), and g-Pe(I-iI-k). 13. The conversion method according to item i2 of the patent application scope, wherein the step (b) further comprises: setting the compensation amounts 5IJ, 5IJ, and (51) to: • occupying the divination x (l- "2) χ / 〆 Δ); 5 ΐ" = ^ ΧΜ 2 χ / - Μ Δ) - ^ χ [(1ι1) χ / - π (Δ) ι - Μ Δ)]; and I - k - xg - pe ( A)·Special [[ΐχ/—^^(4)-容一(四)厶)]; where ul is equal to 〇 or 1, U2 is equal to 〇 or 1, k, △ 1 = 1 i-IJ, and △ 2= =Ij4 - k 〇 14 · The conversion method according to claim 12, wherein the step (b) further comprises: TW2879f for the third lookup table 'the third query table respectively includes a plurality of 29 200818926 Sanda number: TW2879PA number pair (x, h(x)), where the domain of the h function is the minimum gray scale value to the maximum gray scale value, and the range is [〇, 1]; Look up the table and obtain the value of (1), where N1 is the minimum grayscale value and N2 is the maximum grayscale value, and min[N2-I_i, I_k-Nl] is the minimum of (N2-U) and (Ι>Ν1) And setting the compensation amounts d l_i, 5IJ, and “sand: 5 U=h(min[N2-IJ5I_k-Nl])x^x(1_M2)x/_MA); • 5IJ=h(min[N2-I_i,I_.k-Nl])x{A2xu2x/_MA). △1 △ TX[(1_wl)x/_MA)-g_(four)Δ)]}; and <5 I_k=h (min[N2-U,I_k-Nl])x{ Δ2Χ^ ΜΔ). △ Δ1 r χΝχ/—μδ) ι—μδ)]}; where: ul is equal to 〇 or equal to 〇 or △ △=“ · 〇, △ 1 = 1 i-Ij, 15 · The conversion method as described in claim 14, wherein the h function is a nonlinear function, and when the h function has a + + enemy rounding value is the minimum When the gray scale value or the large input gray value is taken, the resulting round out value is zero. 16. The conversion method described in claim 15 of the patent application, m=o and /2=255/when the function of h The round-up value is greater than or equal to 63 and less than or 4 is maintained at 192 days. The resulting output value is: 17·—A color conversion device, TW2879PA is used to convert a multi-primary signal to 30 200818926 Sanda number: TW2879PA The grayscale value (grayScale valuell, L2, and j3 are the output grayscale values 0-i, 〇j, and 〇-k of the pair, the color conversion device includes: an input unit for receiving the multi-primary signal a control unit, according to the multi-primary signal The round grayscale values 1-1, , and ^ generate a memory address and a selection signal; a memory unit stores a plurality of one-dimensional lookup tables, and according to the one-dimensional lookup table, the memory address And the selection signal outputs a corresponding data; a calculation unit calculates the output gray scale values 〇-i, 〇", and 0-k according to the corresponding data; and an output unit for outputting the output gray scales Value 〇", 〇卜 and Ο Jc ; one of which, the input grayscale values L1, Ϊ-2, and! Corresponding to the color coordinate of a color triangle, the color triangle has a plurality of hue regions, and the color coordinates φ corresponding to the input gray scale values U, L2, and L3 are located in one of the hue regions. The control unit generates the memory address and the selection signal according to the color coordinate and the hue region. The color conversion device of claim 17, wherein the control unit further comprises: an address generator coupled to the input unit and generating the memory address according to the color coordinates And a classification controller, which is connected to the input unit, and generates the selection signal according to the hue region. 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 The memory interface accesses a portion of the memory according to the selection signal, and searches for the corresponding data according to the memory address in the one-dimensional lookup table of the memory of each portion to output the Corresponding TW2879PA 32