RU2259602C1 - Pixel color forming method - Google Patents

Abstract

FIELD: matrix indication devices.

SUBSTANCE: method presumes mounting an arbitrary number of sub-pixels in one pixel, estimation of properties of each sub-pixel, forming of control signal for each sub-pixel, calculation of difference between produced and required color signals in selected linear-independent basis and serial transfer to next sub-pixel with consideration of priorities.

EFFECT: higher efficiency, higher quality, lower costs.

4 cl, 4 dwg

Description

The invention relates to circuits and control devices for visual matrix indicators in which the color of a pixel is formed from multi-colored subpixels made of LEDs, liquid crystal and other elements. The method can be used in various image visualization devices, lighting systems, etc., in which the pixels contain many subpixels with an arbitrary spectrum.

, заключающийся в том, что принимают входной векторный сигнал цвета пикселя F, выбирают вектор цвета каждого подпикселя Y_i равным компонентам входного векторного сигнала цвета пикселя.Known [1], pp. 9-10, a method of forming a pixel color containing three subpixels of basic colors (RGB), each of which is characterized by a known maximum color value Y _{max, i} ,

consisting in the fact that they accept the input vector signal of the color of the pixel F, select the color vector of each subpixel Y _i equal to the components of the input vector signal of the color of the pixel.

The basis of this method is the classical idea of pixel color formation due to three basic subpixels of red (R), green (G), blue (B). By changing the brightness of the subpixels, you can achieve the desired color and brightness of the pixel. The linearly independent RGB system is widely used both in the formation of color images in television, computer equipment, photography, and in the display media themselves: cathode ray tubes (CRTs), liquid crystal (LCD) monitors, photo paper, etc.

The disadvantage of this method is the limited functionality associated with the fact that it is applicable only if there are three subpixels of "pure" color R, G and B.

, заключающийся в том, что принимают входной векторный сигнал цвета пикселя F, выбирают векторный сигнал цвета каждого подпикселя Y_i так, чтобы компоненты этого вектора не превосходили входной сигнал цвета F_i и максимальное значение цвета Y_max,i этого подпикселя.Closest to the claimed one is a method [2] of forming a pixel color containing three subpixels of different color and brightness, each of which is characterized by a known vector of maximum color value Y _{max, i} ,

consisting in the fact that the input vector signal of the color of the pixel F is received, the vector color signal of each subpixel Y _i is selected so that the components of this vector do not exceed the input color signal F _i and the maximum color value Y _{max, i of} this subpixel.

In this method, in addition to the basic RGB colors, the pixel of the liquid crystal panel contains one additional - the brightness white subpixel W, which allows to increase the brightness of the image. White color quite often turns out to be an essential component of images, so its use is justified. The implementation of this method is that for normalized RGB signals (figa) determine the minimum level (in this case, G), which corresponds to the level of W - white. The W signal is applied to a white subpixel, and the color of the remaining subpixels does not change. As a result (FIG. 1B), the brightness of the pixel increases while maintaining color.

The disadvantages of this method are the limited scope, which are as follows:

- The method does not take into account the nonlinearity of the dependence of the brightness of the subpixels on the control signal, which leads to color distortion.

- The method is applicable for pixel input signals presented only in the RGB basis, while in many applications other bases are more convenient.

- The method is applicable only in the presence of RGB subpixels, while in many cases, for example, when using LEDs, it is difficult, and sometimes impractical, to obtain “pure” RGB colors. With arbitrary color of the subpixels, the task of dividing the input signal between the subpixels does not have an obvious solution.

- The method is applicable only with 3 (RGB) or 4 (RGBW) subpixels. In the case of LEDs, for example, for large display panels, the number of subpixels can reach tens.

- The method does not take into account the manufacturing cost and operating costs of subpixels. Indeed, in the manufacture of CRT and LCD indicators, the differences in the cost of manufacture and operation of these elements are negligible. When executing subpixels of discrete elements, these differences can be significant.

The problem solved in the claimed method consists in improving the quality of displaying the color of the pixel in various ways of setting input information, an arbitrary number and combination of colors of subpixels, as well as in reducing the cost of manufacturing and operation of display elements.

, заключающийся в том, что принимают входной векторный сигнал цвета пикселя F, выбирают векторный сигнал цвета каждого подпикселя Y_i так, чтобы компоненты этого векторного сигнала не превосходили входной векторный сигнал цвета F_i и максимальное значение векторного сигнала цвета Y_max,i этого подпикселя, в пиксель дополнительно устанавливают произвольное число подпикселей различной цветности и яркости, измеряют зависимость векторного сигнала цвета Y_i каждого подпикселя от управляющего сигнала I_i, устанавливают приоритеты между подпикселями, последовательно от наиболее приоритетного к менее приоритетному вычитают из входного векторного сигнала цвета подпикселя F_i выбранный векторный сигнал цвета Y_i подпикселя и используют результат ΔF_i в качестве входного векторного сигнала цвета для следующего подпикселя F_i+1, в качестве входного векторного сигнала цвета первого подпиксела F_i используют входной векторный сигнал цвета пикселя F, формирование управляющих сигналов I_i,

подпикселей производят с учетом зависимости векторного сигнала цвета каждого подпикселя от управляющего сигнала Y_i(I_i) и так, чтобы сумма векторных сигналов цвета Y_i всех подпикселей стремилась ко входному векторному сигналу пикселя F.To solve this problem according to claim 1, in a method of forming a pixel color containing three subpixels of different color and brightness, each of which is characterized by a known maximum value of a color vector signal Y _{max, i} ,

consisting in the fact that the input vector signal of the color of the pixel F is received, the vector color signal of each subpixel Y _i is selected so that the components of this vector signal do not exceed the input vector signal of the color F _i and the maximum value of the color vector signal Y _{max, i of} this subpixel, a pixel further comprises establishing an arbitrary number of subpixels of different chrominance and luminance measured dependence of the color signal vector Y _i from each sub-pixel control signal i _i, set priorities between subpixels and, sequentially from the highest priority to the less priority is subtracted from the input vector signal subpixel F _i color selected vector Y color signal _i subpixels and used ΔF _i result as an input vector the color signal for the next subpixel F _{i + 1} as the input vector color signals the first subpixel F _i use the input vector signal of the color of the pixel F, the formation of control signals I _i ,

subpixels are produced taking into account the dependence of the vector color signal of each subpixel on the control signal Y _i (I _i ) and so that the sum of the vector signals of color Y _{i of} all subpixels tends to the input vector signal of the pixel F.

According to paragraph 2 of the claims in the method according to claim 1, in the presence of a group of equal priority subpixels, the input color vector signal within the group is distributed according to the established rule, and the difference between the input color signal of the group of subpixels F _{gr is} used as the input color vector signal of the color following the subpixel _{group .} and the sum of the selected vector signals of color Y _i group of subpixels.

According to paragraph 3 of the claims in the method according to claims 1, 2, for each subpixel, the maximum color vector signal Y _{i is} selected.

всех подпикселей, возвращаясь от наименее приоритетного подпикселя к наиболее приоритетному и увеличивая каждый раз модуль векторного сигнала цвета Y_i на выбранный шаг дискретизации.According to paragraph 4 of the claims in the method according to claims 1, 2, the selection of color vectors Y _{i is} repeated many times,

of all subpixels, returning from the lowest priority subpixel to the highest priority and each time increasing the modulus of the vector signal of color Y _i by the selected sampling step.

The basis of the proposed method is based on the following considerations. In recent years, interest in LEDs has increased significantly. These devices have high reliability (operating time up to hundreds of thousands of hours), which allows you to create durable lighting and information display devices on their basis that do not need repair. Such devices can be built into walls, floors and other capital objects. However, the limited brightness, the difficulty of creating LEDs of a given color, as well as the significant nonlinearity of the dependence of brightness on the control signal (current I) create certain problems with their use for creating high-quality color images and lighting. Under these conditions, it is necessary to use pixels consisting not of traditional “pure” subpixels of RGB or W colors, but of subpixels having an arbitrary color and with limited brightness capabilities. However, even in this case, the subpixels must form a linearly independent system, providing any brightness and color of the pixel. With LED pixel execution, a significantly larger number of subpixels than that used in analogs may be required to achieve specified brightness and color values. The cost of LEDs of different brightnesses and colors and their energy efficiency are significantly different, which allows you to find the most profitable option for implementation and application.

Chromaticity is usually understood as a generalized characteristic of the pixel emission spectrum associated with the dominant wavelength and expressed by the coordinates of a point in a given unit plane of the color three-dimensional space, for example, the coordinates (x, y) in the XYZ MCO system [3]. Luminance is usually understood as the amplitude value of light radiation at a given color. By the color vector of a pixel or subpixel we mean a set of parameters characterizing color and brightness. A color vector is usually represented in the form of its coordinates in a 3-dimensional orthogonal basis and consists of the corresponding number of components defining a vector in this basis. The basis should be linearly independent, providing the ability to obtain arbitrary values of brightness and color. In addition, if the basis is orthogonal, then the solution to the problem of the formation of a given color is simplified. In this basis, the desired pixel color is set, as well as the characteristics of the light-emitting elements - subpixels [1], pp. 9-14. Various systems of parameters can serve as such a basis. In particular, in addition to the widespread RGB basis, CMY, CMYK, Lab and others are used. The CMY system is subtractive, in which the components resulting from the subtraction of the RGB components from white W are set as color parameters, i.e. C = (W-R), M = (W-G) and K = (W-B). In the CMYK system, black is added to these components - K. In the Lab system, the brightness is separated from the color, and the color is characterized by the brightness L and two color components: a - a parameter that changes from green to red, and b - from blue to yellow . The choice of a specific basis is carried out depending on the convenience of technical implementation.

Most hardware and software systems generate a pixel color for a display device in one of the traditional linearly independent and orthogonal bases (x, y, z), figure 2, for example, RGB. As a result, the input vector signal of color F enters the pixel in the form of three coordinates (x, y, z), showing the position of the desired color in the selected basis.

, заданных в базисе (x,y,z). Функции Y_i(I)

образуют базис размерности n. Задача отображения цвета состоит в том, чтобы подобрать для каждого i-го подпикселя такое значение управляющего сигнала I_i так, чтобы сумма векторов Y_i(I_i) совпала с (приблизилась к) F, фиг.3. В математическом смысле эта проблема является задачей математического программирования, которая имеет достаточно сложное и трудоемкое точное решение при условии линейной независимости векторов Y_i(i) при любых значениях I. Задача еще более осложняется, если размерность базиса Y_i(I)

велика. Однако можно получить ее приближенное решение, используя соответствующие методы.A pixel is formed by several subpixels of different color and brightness, for example, made on the basis of LEDs. If the colors of the subpixels coincide with the RGB basis, as is the case in the analogue and prototype, then there are no problems with getting to the given point F. If the color of the subpixel does not coincide with any of the coordinates (x, y, z), then its properties can be specified in this basis as a vector function Y (I), which shows how the color of the subpixel depends on the input signal (current) I , as well as the maximum capabilities of this subpixel Y _max . The input signal I acts as a parameter of the function Y. The functions Y (I) of all subpixels form a set of vector functions. Thus, a situation arises when there is a linearly independent orthogonal basis (x, y, z) in which the desired color (vector) F and a set of vector color functions of subpixels Y _i (I) are specified

defined in the basis (x, y, z). Functions Y _i (I)

form a basis of dimension n. The task of color display is to select for each i-th subpixel such a value of the control signal I _i so that the sum of the vectors Y _i (I _i ) coincides with (approaches) F, Fig. 3. In the mathematical sense, this problem is a mathematical programming task that has a rather complicated and time-consuming exact solution provided that the vectors Y _i (i) are linearly independent for any values of I. The task is even more complicated if the dimension of the basis Y _i (I)

great. However, its approximate solution can be obtained using appropriate methods.

One of the simplest approximate (quasi-optimal) is the method according to which at each step the vectors Y _i (I) are selected as maximum, but not exceeding F _i and Y _{max, i in} none of the coordinates of the basis. This procedure is repeated from the highest priority subpixel to the lowest priority. Such an algorithm is the simplest, fastest and most effective if the components Y _i (I _i ) are close to the basis or if the number of subpixels is large. However, as a result of its application, an error may occur, which depends mainly on how much Y _i (I _i ) correspond to the orthogonal basis.

The second approximate method is the sequential, in accordance with the priorities, increase the modules Y _i (I _i ) with the selected step, returning from the lowest priority subpixel to the highest priority. As a result, the approximation error may not exceed the selected step.

Significant differences of the proposed method in comparison with analogues characterize the following features.

An additional number of subpixels of various color and brightness is additionally set in the pixel. The total number of subpixels can be from 3 to an arbitrary number n. As a result, the inventive method is applicable for any number of subpixels. Additional sub-pixels allow, for example, to expand the dynamic range of pixel brightness, i.e. achieve arbitrary brightness due to the parallel operation of several single-color subpixels. In addition, it becomes possible to control the color inside the pixel, providing uniform glow, the concentration of glow at the center, periphery and other options. Thus, the claimed method is applicable for any number of subpixels.

In the prototype, in addition to three regular RGB subpixels, only one brightness subpixel is installed - W. A larger number of subpixels cannot be used fundamentally because of the features of the method [2].

The dependence of the color vector Y _{i of} each subpixel on the control signal I _{i is} measured. As already noted, for LEDs, this dependence is substantially nonlinear, and when choosing the control signal I, taking this dependence into account is necessary to improve color reproduction.

In the prototype, the specified non-linearity is not taken into account.

Prioritize between subpixels. Priorities allow you to choose the best, for example, from an energy point of view, the sequence of choosing the color vector Y _i . This reduces the energy consumption of the pixel, i.e. improve light output. For example, if a pixel consists of four RGBW LEDs, it is energetically more profitable to illuminate one white LED than to achieve the same level of brightness using a combination of three RGB LEDs. In this case, the white LED is assigned the highest priority, and the rest - in an arbitrary sequence. When constructing an LED searchlight containing many (tens) LEDs, emitters located closer to the optical axis of the device should be more priority in relation to peripheral ones, since they provide better conditions for the passage of light through reflectors and lenses.

In the prototype, all subpixels are equal in priority.

Subsequently, the selected color vector Y _{i of the} subpixel is subtracted from the highest priority to lower priority from the input vector signal of the subpixel color F _i and the result ΔF is used as the input color vector signal for the next subpixel F _{i + 1} , as the input vector color signal of the first subpixel F ₁ using the input signal vector color pixel F. This approach (Figure 3) allows gradually bring the pixel color vector Y = {Y _1, Y _2, ..., Y _n} to the desired input vector F. Moreover, color parameters first vector u ETA F try to meet the expense of higher-priority sub-pixels. The share of less priority subpixels is only the remainder, not implemented by the previous ones. As a result, priority subpixels are most involved, which reduces the cost of operation, such as energy consumption.

In the prototype, the sub-pixels are interdependent and there is no need for prioritization.

подпикселей производят с учетом зависимости вектора цвета каждого подпикселя от управляющего сигнала Y_i(I_i). В результате этой процедуры нужно получить сигнал I_i (ток, который надо подать в i-й подпиксель), являющийся нелинейной обратной функцией цвета подпикселя I(Y_i) и известной в результате измерений, описанных выше. Эта процедура позволяет скомпенсировать нелинейную зависимость цвета подпикселя от управляющего сигнала I_i,

и за счет этого адекватно отобразить цвет подпикселя. Таким образом, пиксель оказывается независимым от индивидуальных особенностей подпикселей. Нелинейность подпикселей учитывается в его устройстве управления, и не требуется предварительного преобразования входного векторного сигнала F в какую-либо форму. Такой прием позволяет получить высокое качество отображения и использовать пиксель (матрицу пикселей) при обычных входных сигналах.The formation of control signals I _i

subpixels are produced taking into account the dependence of the color vector of each subpixel on the control signal Y _i (I _i ). As a result of this procedure, it is necessary to obtain a signal I _i (the current that must be supplied to the i-th subpixel), which is a nonlinear inverse function of the color of the subpixel I (Y _i ) and is known as a result of the measurements described above. This procedure allows you to compensate for the nonlinear dependence of the color of the subpixel on the control signal I _i ,

and due to this, adequately display the color of the subpixel. Thus, the pixel is independent of the individual characteristics of the subpixels. The non-linearity of the subpixels is taken into account in its control device, and no preliminary conversion of the input vector signal F into any form is required. This technique allows you to get high quality display and use a pixel (matrix of pixels) with conventional input signals.

In the prototype, the indicated nonlinearity of the radiating properties of the subpixel is not taken into account.

The selection of vector signals of the color of the subpixels Y _{i is} performed in such a way (Fig. 3) that their total value tends to the input vector signal of the pixel F. This sign means that the total color signal Y _{i of} all subpixels should be as close as possible to the input signal of the pixel F.

In analogue [1], the three components (RGB) of the input vector signal of the pixel F are supplied to the corresponding sub-pixels. As a result, the total output signal of the RGB subpixels coincides with the input F, so this feature in the present method and analogue matches. In the prototype [2], an additional luminance signal W is formed, as a result, the output total RGBW signal is larger than F, but while maintaining color.

According to claim 2. formulas sign: if there is a group of equal priority subpixels, the input color vector signal inside the group is distributed according to the established rule, and the difference ΔF _gr between the input color signal of the group of subpixels F _gr and the sum of the selected vector color signals is used as the input color vector signal after the subpixel group Y _{i of the} group of subpixels or an equal sum of differences between the vector input signal of each subpixel in the group F _i and the vector signal of color Y _i selected for this subpixel. Let us illustrate this with an example. Let a pixel include several K subpixels of the same color, which are set to increase brightness. These subpixels are assigned the same priority, and they form a group having a common input signal F _gr . The specified signal can be distributed between them using various, a priori established rules. For example, evenly. In this case, the input vector signal of each subpixel becomes equal to F _gr / K. This option provides uniformity in the distribution of pixel color. In a LED floodlight, for a greater range of illumination, it is advisable to give preference to subpixels located closer to the optical axis, and to obtain diffused light to peripheral subpixels. Other distribution rules are possible. In this case, the input of the subpixel following the group must be supplied, as in previous cases, the difference between the input signal F _gr and the sum of the selected vector signals of color Y _i groups.

In the prototype, a pixel consists of only 4 subpixels of different (RGBW) color, and this problem is not solved.

According to claim 3 of the claims, a feature: a vector signal of color Y _{i of} each sub-pixel is selected as maximum, which means the following. When choosing a control signal I _{i of} each subpixel, it is necessary to ensure that at least one component of the vector signal of color Y _i coincides with the corresponding component of either the input vector signal of color F _i or the maximum color value Y _{max, i of} this subpixel. This means that the capabilities of each sub-pixel are used to the full to achieve the desired brightness. Thus, moving from the highest priority subpixel to the lowest priority at each step, the maximum modulus signals Y _i are selected. As a result, in one pass, it is possible to obtain all the values of the control signals I _i .

The prototype can also solve the problem of maximizing Y _i , but only for one - brightness - subpixel. For the rest of the subpixels, there is a strict correspondence between the input and output RGB signals.

всех подпикселей, возвращаясь от наименее приоритетного подпикселя к наиболее приоритетному и последовательно увеличивая каждый модуль векторного сигнала цвета на выбранный шаг дискретизации, означает следующее. При выборе векторного сигнала цвета Y_i максимальным на каждом шаге, как это предусматривалось в предыдущем пункте, может возникнуть значительная ошибка - рассогласование ΔF (фиг.3), между входным сигналом F и суммой векторных сигналов цвета Y_i. Эта ошибка связана с тем, что в рассматриваемом случае базис подпикселей является линейно-независимым, но не ортогональным. В ортогональном базисе (x,y,z) движение к желаемой точке F (фиг.1) может происходить по любому из маршрутов: x-y-z, y-z-x и т.п. с одинаковым успехом. При не ортогональном базисе продвижение максимальными шагами по векторам базиса в произвольном порядке может не дать хорошего результата. Для получения более качественного результата целесообразно продвигаться к F небольшими дискретными шагами, постепенно увеличивая модули векторных сигналов Y_i,

. В результате после нескольких итераций ошибка может быть уменьшена до величины шага дискретизации.According to claim 4 of the claims, a sign: repeatedly select the color vectors Y _i ,

of all subpixels, returning from the lowest priority subpixel to the highest priority and sequentially increasing each module of the vector color signal by the selected sampling step, means the following. When choosing a vector signal of color Y _i maximum at each step, as provided for in the previous paragraph, a significant error may occur - a mismatch ΔF (Fig. 3) between the input signal F and the sum of vector signals of color Y _i . This error is due to the fact that in the case under consideration, the basis of subpixels is linearly independent, but not orthogonal. In the orthogonal basis (x, y, z), movement to the desired point F (Fig. 1) can occur along any of the routes: xyz, yzx, etc. with equal success. With a non-orthogonal basis, advancing with maximum steps along the basis vectors in an arbitrary order may not give a good result. To obtain a better result, it is advisable to advance to F in small discrete steps, gradually increasing the modules of vector signals Y _i ,

. As a result, after several iterations, the error can be reduced to the value of the sampling step.

The prototype uses the orthogonal basis RGB, and this problem is not solved.

The inventive method is illustrated by the following graphic materials:

Figure 1 - the principle of operation of the prototype.

Figure 2 - the position of the input vector signal F and the function Y (I) in the basis (x, y, z).

Figure 3 - the process of generating a given input vector signal F from the functions Y (I).

4 is a subpixel control circuit.

Consider the implementation options of the proposed method.

Depending on the purpose, an arbitrary number of subpixels of different color and brightness is set in the pixel. The number of subpixels depends, for example, on the requirements for a dynamic range of brightness. The brightness capabilities of one subpixel are limited by a certain value of Y _max . By setting several identical sub-pixels, you can expand the dynamic range an appropriate number of times. In classical applications (cathode ray tubes, liquid crystal monitors, etc.), the developers spent a lot of money on creating a pixel structure containing 3-4 subpixels of basic colors. These costs paid off by mass production. When creating billboards, placards and other visualization devices, due to the variety of requirements, one has to be content with the existing elemental base and carry out such objects from discrete elements, for example, from LEDs. To achieve a given color, these elements must have different color characteristics, such as RGB. However, as will be shown below, subpixels of other colors, including mixed ones, which are cheaper, can be used. A pixel is usually regarded as a point source with indistinguishable details for the observer - subpixels. However, in some applications, such as LED spotlights, it is necessary to control the internal structure of the pixel. In particular, in the presence of many multi-colored subpixels, their relative position, methods of controlling the same subpixels, and other circumstances become important.

The dependence of the vector signal of color Y of each subpixel on the control signal I is measured. The signal of color Y _i emitted by the subpixel can be considered monochromatic and linearly dependent on the control signal only in the simplest cases, for example, in classical RGB systems, each color is considered to be “pure”, and the brightness of each subpixel - a linear function of the input signal I. However, in many cases, for example, for LEDs, this approach is not justified. To evaluate Y (I), either the passport data of the LEDs can be used, or appropriate measurements can be made. In this case, the dependence Y (I) turns out to be a vector function given in some basis (x, y, z), for example (RGB).

Prioritize between subpixels. The reasons for setting priorities can be energy considerations, for example, one white subpixel consumes less energy than the sum of RGB, geometric considerations - an LED located in the center of the pixel is more profitable than located on the periphery and other considerations. If there are several, for example, the same sub-pixels, they can be given the same priority.

An input vector signal of the color of the pixel F is received. The task of all sub-pixel control circuits is to find the control signals I _i for all sub-pixels, such that the sum of the vector signals Y _i (I _i ) tends to the input vector signal of the pixel F.

To solve this problem, the input vector signal of the color of the pixel F is first fed to the highest priority subpixel. The task of the control circuit of this and all subsequent subpixels is to select the value of the vector function Y _i (I _i ) and the corresponding control signal I _i . This is possible, since the dependence Y _i (I) is known. The value of the vector function (its coordinates in the selected basis) Y _i (I _i ) must satisfy a number of conditions. Firstly, Y _i (I _i ) must not exceed the maximum color value of the corresponding subpixel Y _{max, i} . This condition means that a subpixel can only produce a limited color. Secondly, Y _i (I _i ) must not exceed the input signal of the subpixel F _i in any of the coordinates, i.e. subpixel color should not exceed the job. In the first subpixel, this is the value of F, and in the rest, F _i .

Subtract the selected color vector signal Y _{i of} this subpixel from the input color vector signal of the color of each subpixel F _i . The resulting value ΔF _i , figure 2, shows how close the sum of signals Y _i is close to F.

The result obtained is used as an input color vector signal for the next lower priority subpixel F _{i + 1} . Thus, the control circuit of the next subpixel continues to approximate the sum of the signals Y _i to F.

The procedure indicated for one subpixel is repeated, moving from the highest priority subpixels to lower priority ones. At the end of the procedure, control signals I of all subpixels will be generated, which provide some approximation of the sum of the colors of the subpixels to a given color F.

A pixel may include several subpixels that are assigned the same priority. These subpixels are grouped together. For example, if there is a group of several identical red subpixels, the task of achieving the required brightness for a given color can be solved in various ways: by jointly and uniformly illuminating these pixels, making maximum use of the brightness of some of them, and in other ways. Depending on the selected method, a rule is established and the input color signal of this group F _{gr is} distributed between subpixels, for example, dividing it evenly between them. In this case, the principle of operation of the sub-pixel control circuit remains the same, i.e. in each subpixel, a control signal I _i and the difference ΔP _i between the input signal F _{i of} this subpixel distributed according to the established rule and the selected value of the color vector Y _i will be generated. Then, as the input vector signal of color F of the next after the subpixel group, you need to use the difference between the input signal of the color of the group of subpixels F _gr and the sum of the selected vector signals of color Y _{i of the} group of subpixels or, equivalently, the difference between F _gr and the sum ΔF _i in the group.

When implementing the proposed method, a different strategy for choosing a vector signal of color Y _i (I _i ) can be used.

The simplest option is to select the maximum value of Y _i (I _i ) for each subpixel. At the same time, the capabilities of priority subpixels will be used to the greatest extent compared with less priority ones. This method provides a quick approximation to the desired pixel color. However, in this case, lower priority pixels may remain unused, and the approximation error F _OSH , FIG. 2, may turn out to be large, especially with mixed colors of the subpixels.

подпикселей предполагает возвращение от наименее приоритетного подпикселя к наиболее приоритетному, увеличивая каждый раз модуль векторного сигнала цвета Y_i на выбранный шаг дискретизации. Такая многократно повторяющаяся процедура обеспечивает постепенное продвижение суммы векторов Y_i к желаемому значению F. При этом окажутся задействованными все необходимые подпиксели, а точность приближения окажется выше.Another option is the selection of vector signals of color Y _i ,

subpixels implies a return from the lowest priority subpixel to the highest priority, increasing each time the modulus of the vector signal of color Y _i by the selected sampling step. Such a repeated procedure ensures the gradual advancement of the sum of vectors Y _i to the desired value F. In this case, all necessary subpixels will be involved, and the accuracy of the approximation will be higher.

The inventive method can be implemented programmatically using a microcontroller and a hardware-analog or digital method. In the first embodiment, one microcontroller can control all subpixels, which makes this option cheaper and preferable. However, for greater clarity, consider the diagram of the sub-pixel control device 7 shown in FIG. 4, which implements the inventive method for one sub-pixel in the analogue version, where:

1. emitter (LED).

2. driver control signal I _i .

3. shaper of the vector signal of the color of the subpixel Y _i .

4. vector subtractor.

5. vector subtractor.

6. zero detection circuit.

7. diagram of the sub-pixel control device.

The emitter 1 (LED) is designed to generate a light signal.

Shaper 2 of the control signal I _{i is} designed to generate current LED. Shaper 1 can be made in the form of an increasing current generator based on an operational amplifier. The generator has start and stop inputs, which can be implemented by the corresponding switching elements (keys).

Shaper 3 of the vector signal of the color of the subpixel Y _{i is} intended to convert the current I _i into a vector signal Y _i , which displays the nonlinear dependence of the color of a given subpixel on the input signal. Shaper 3 in a three-dimensional basis contains three independent blocks. Each of the blocks forms one coordinate Y _i in the selected basis. Blocks can be implemented as an operational amplifier with non-linear feedback.

The vector subtractor 4 is designed to find the difference ΔF _i , figure 2, between the color vector signal Y _i and the input vector signal F _i . Subtraction is carried out separately according to the coordinates of the selected basis using operational amplifiers.

The vector subtractor 5 is designed to find the difference between the vector color signal Y _i and the maximum vector color signal of a given subpixel Y _{max, i} . Subtraction is carried out separately according to the coordinates of the selected basis using operational amplifiers.

The zero detection circuit 6 is used to determine the moment when a zero value is detected at one of the outputs of the subtractors 4 and 5, and to generate a signal for completing the operation of the sub-pixel control circuit. The zero detection circuit 6 can be implemented on the basis of comparators detecting the appearance of zero on one of the outputs connected by the outputs of the OR logic circuit.

The sub-pixel control device 7 is designed to control the color of one sub-pixel. All subpixels have similar schemes, and embodiments of the proposed method differ in external schemes.

подаются выходные сигналы ΔF_i с вычитателей 4 предшествующих подпикселей F_i-1. Сигнал запуска первого подпикселя подается извне и служит для запуска схемы. Выходной сигнал "окончание i" схемы обнаружения нуля 6 i-го,

подпикселя подан на вход запуска следующего подпикселя и на собственный вход "остановка i". Перед началом работы измеряют свойства каждого подпикселя. Для этого оценивают максимальные значения векторов Y_max,i и зависимость вектора Y_i(I_i). Эту информацию можно получить путем измерений или из справочников. Векторную величину Y_max,i подают на вход вычитателя 5 каждого подпикселя, а зависимость Y_i(I_i) вводят в виде параметров формирователей 3 векторного сигнала цвета подпикселя.The device operates as follows. We assume that a pixel consists of an arbitrary number of subpixels with known characteristics. In the initial state, the drivers of the control signal I _{i of} all subpixels are zeroed. An input vector signal of a pixel F is fed to the input of the subtractor 4 of the highest priority - the first - subpixel. At the inputs F _{i of the} subtractors are 4 subsequent subpixels

ΔF _i outputs from the subtractors 4 of the preceding subpixels F _i-1 . The start signal of the first subpixel is supplied externally and serves to start the circuit. The output signal "end i" circuit detection zero 6 of the i-th,

subpixel is fed to the start input of the next subpixel and to its own input "stop i". Before starting work, the properties of each subpixel are measured. For this, the maximum values of the vectors Y _{max, i} and the dependence of the vector Y _i (I _i ) are estimated. This information can be obtained by measurement or from reference books. The vector value Y _{max, i is} fed to the input of the subtractor 5 of each subpixel, and the dependence Y _i (I _i ) is introduced in the form of parameters of the formers 3 of the subpixel color vector signal.

When a trigger signal is input to the shaper 2 of the first subpixel, the signal I ₁ at its output starts to increase and the brightness of the LED 1 increases. The shaper 3 generates a rising vector signal Y ₁ (I ₁ ), which corresponds to the current brightness and color of the first LED. This signal by subtraction in blocks 4 and 5 is compared with the maximum vector Y _{max, 1} and with the input signal of the subpixel F ₁ = F. If the signal Y ₁ (I ₁ ) at some instant in at least one of the coordinates Y _{max, 1} or F ₁ matches the indicated values, this means that either the color of the first pixel has reached the maximum value for this subpixel, or the output signal of the LED on this coordinate reached the desired value. The coordinate coincidence is fixed by a zero detection circuit 6, which, through the “Stop i” input, stops the shaper 2 of this subpixel and starts the control device for the next subpixel, the input of which is a vector signal ΔF = (F ₁ -Y ₁ (I ₁ )). This signal indicates how much the color of the first subpixel does not match the input signal. Subsequent subpixels work similarly. After the last least-priority sub-pixel finishes, a color will be formed in the pixel that approximately matches the specified color. The described option corresponds to a sequential selection scheme for the maximum values of Y _i (I _i ), according to claims 1 and 3 of the claims.

To implement paragraph 2 of the formula scheme, you must use the following scheme. The input signal F of the group of equal priority subpixels is divided between the group using scaling operational amplifiers with given coefficients. The condition for the end of color formation in a group is the completion of all subpixels, i.e. the arrival of “end” signals from all devices 7. The output signal for starting the control circuit by the next subpixel after the group is formed by the logic circuit I. To form the input vector signal of color F of the next subpixel, it is necessary to add all the signals ΔF _i inside the group. This operation can perform a summing operational amplifier.

To implement claim 4 of the formula, the start and stop signals of the sub-pixel control circuits 7 are supplied externally for a short time sequentially from the highest priority sub-pixel to the lower priority, followed by the return and repetition of the procedure. Such signals can be generated by a pulse distributor (shift register). A sign of the end of the entire circuit is the appearance of signals "end" at the outputs of all subpixels.

Thus, the described device confirms the possibility of technical implementation of the proposed method.

The advantages of the proposed method in comparison with the known authors analogues are:

1. Advanced pixel brightness.

2. High quality color rendering.

3. The reduced cost of operation.

4. Possibility of application for any type of subpixels (from incandescent lamps to LEDs).

5. Ability to use with any method of setting the source information (arbitrary basis).

6. The ability to build various display and lighting devices with desired properties from low-cost components.

Sources of information

1. Glushakov S.V., Knabe G.A. Computer graphics. Training course, Kharkov - "Folio", Moscow "ACT", 2001, 500 p.

2. Patent WO 0 137251 "LIQUID CRYSTAL DISPLAY DEVICE WITR HIGH BRIGHTNESS" G 09 G 3/36, published 2001-05-25, priority JP 19990321902 19991112; JP 20000330859 20001030.

3. Reference book on lighting engineering / ed. Yu.B. Aizenberg. M .: Energoatomizdat, 1995 .-- 528 p.

Claims (4)

1. The method of forming the color of a pixel containing three subpixels of different color and brightness, each of the subpixels is characterized by a known maximum value of the color vector signal Y _{max, i} , namely, that an input vector signal of the color of the pixel F is received _{, a} vector color signal of each subpixel Y is selected _i so that the signal components of the vector do not exceed the input color vector signal F _i and the maximum value Y _max color vector _{signal, i} of the sub-pixel, wherein the pixel in a surcharge but set an arbitrary number of subpixels of different chrominance and luminance measured dependence Y _i vector color signals of each sub-pixel on the control signal I _i, set priorities between subpixels, sequentially from the highest priority to the lower priority sub-pixel is subtracted from the input vector signal subpixel color F _i selected vector signal colors Y _{i of the} subpixel and use the result ΔF as the input color vector signal for the next subpixel F _{i + 1} , as the input vector with the color needle of the first subpixel F ₁ uses an input vector signal of the color of the pixel F, the selection of control signals I _{i of the} subpixels is made taking into account the dependence of the vector color signal of each subpixel on the control signal Y _i (I _i ) and so that the sum of the vector signals of color Y _{i of} all subpixels tended to the input vector signal of the pixel F. 2. The method according to p. 1, characterized in that, if there is a group of equal priority subpixels, the input color vector signal within the group is distributed according to the established rule, and the difference between the input color signal of the group of subpixels F _{gr is} used as the input color vector signal of the color following the subpixel _group. and the sum of the selected vector signals of color Y _i group of subpixels. 3. The method according to p. 1 or 2, characterized in that the maximum color vector signal for each subpixel Y _{i is} selected. 4. The method according to p. 1 or 2, characterized in that repeatedly repeat the selection of vector signals of color Y _i ,

of all subpixels, returning from the lowest priority subpixel to the highest priority and each time increasing the modulus of the vector signal of color Y _i by the selected sampling step.

Images