KR20030020210A - Error Diffusion Method for display device - Google Patents

Abstract

PURPOSE: An error diffusion method for reducing low gray scale false contour noise of a display device is provided, which reduces false contour noise generated in a low gray scale area of a display image, and improves a brightness repeatability. CONSTITUTION: According to the error diffusion method, a difference between lightness as to a gray scale before gamma correction and lightness as to a gray scale after gamma correction of an input image signal is calculated as a display error, and the display error is weighted and is diffused to a peripheral pixel of a display pixel. An analog image signal inputted from the external is converted into digital data having n bit gray level(ST701,ST702). Display error weighted with an error diffusion coefficient is added to the above digital data and is output as the first input gray(ST703). And the first input gray executes an image display according to the second input gray having gamma correction(ST704).

Description

Error diffusion method for reducing low gradation contour of display device {Error Diffusion Method for display device}

The present invention relates to an error diffusion method for improving the image quality of a display device. In particular, in a display device for displaying a digital image such as a plasma display panel, it is possible to reduce the image quality deterioration generated in the low gradation region of the display image and to improve luminance reproducibility. The present invention relates to an error diffusion method for reducing a low gray pseudo contour of a display device that can be improved.

In general, a plasma display panel (hereinafter referred to as a 'PDP') is configured to express gray scales of a display image by adjusting the number of discharges per unit time of a discharge cell. That is, in the gray scale expression method of the PDP device, when one still picture is one frame, X or more subfields (for example, eight) that display one frame each with 2 ⁰ to 2 ^X-1 discharges each time ( Sub-field), and the luminance of the display image is expressed by, for example, 256 levels of gray scale according to the combination of the number of discharges of each subfield.

Such a PDP device has a linear characteristic as shown in FIG. However, the perceived brightness of the human eye actually has a nonlinear characteristic in which the luminance deviation is large in the low gradation region and the luminance deviation is small in the high gradation region according to the input gradation. There is a problem that the rain is not stable.

Proposed gamma correction is a well known gamma correction. The gamma correction is performed by adjusting the distribution of the input gradation so that the recognition luminance of the input gradation versus the display image has a linear characteristic. For example, the gamma correction may be performed if the input gray level before gamma correction is a uniform distribution having even a system of 1, such as 0, 1, 2, 3, ..., 254, 255. Input of low gradation area such as 0, 0, 0, 0, 1, 1, 1, 1, 1, ......, 20, 20, 21, 21, ...., 250, 252, 255 The gradation is densely distributed, and the input gradation of the high gradation region is adjusted in such a manner that the distribution ratio is distributed so as to distribute even a predetermined gradation.

In this case, since the numerical luminance value for each input gradation value is constant for each PDP device, the luminance deviation in the low gradation region is small, and the luminance deviation in the high gradation region has a relatively large value. Therefore, after gamma correction, the relationship between input gradation and perceived luminance has a linear characteristic as shown in FIG. In this case, however, there is also a problem in that a contour that deteriorates the image quality of the display image is generated in the low gradation region (dark region) a as shown in Fig. 1B.

Hereinafter, a process of generating the pseudo contour will be described with reference to FIG. 2.

The graph of the input gradation versus luminance characteristic shown in FIG. 2 shows a low gradation region (for example, 0 to 60 gradations of 0 to 255 gradations) where pseudo contours are generated for convenience of description.

In FIG. 2, the target gamma b of the display image has an exponentially increasing luminance while the display gamma c actually output has a stepped luminance. Therefore, due to the brightness characteristics of the staircase shape, pseudo contours are hardly generated in the high gradation region (bright region) as the inclination of the gamma curve increases, but pseudo contours are generated in the low gradation region of the display image.

Therefore, a deviation (hereinafter, referred to as "display error") between the input image signal and the displayed image signal in order to reduce the pseudo contour of the low gradation region described above is calculated, and this is referred to as a reference pixel for error diffusion (hereinafter, referred to as a "primary pixel"). A so-called error diffusion method is known in which a low gradation contour of a displayed image is reduced by diffusing to neighboring pixels adjacent to the " "

In other words, the conventional error diffusion method converts an analog image signal input from the outside when displaying 256 gray levels into, for example, 12-bit digital data for each of R, G, and B colors, and then uses the lower 4 bits as display errors as shown in FIG. 3. The peripheral pixels B to E adjacent to (A) are multiplied by a predetermined weight to diffuse. In this case, the input gradation of the display pixel is calculated as the sum of the 8-bit image signal excluding the display error of the pixel and the display error weighted and diffused from the adjacent pixels. Therefore, according to the above method, it is possible to reduce the pseudo contour in the low gradation region of the display image by performing error diffusion on the surrounding pixels of the pixel of interest.

However, in the conventional error diffusion method described above, pseudo contours can be reduced in a low gradation region, but the luminance of an error-diffused display image and a perceived luminance perceived by an actual human vision are obtained by directly using a part of an input image signal when calculating a display error. There is a problem that the luminance reproducibility of the display image is deteriorated due to the deviation between the two.

Accordingly, the present invention has been made in view of the above circumstances, and this is a display device having a linear luminance characteristic with respect to an input gradation such as a PDP, which reduces pseudo contours occurring in a low gradation region of a display image, and its luminance reproducibility. It is an object of the present invention to provide an error diffusion method for reducing the low gray pseudo contour of a display device capable of improving the performance.

Fig. 1A is a diagram showing input grayscale versus luminance characteristics of a PDP apparatus.

Fig. 1B is a diagram showing input grayscale versus perceived luminance characteristics of a PDP apparatus.

2 is a diagram for explaining pseudo contours in a low gradation region of a display image;

3 is a view for explaining an error diffusion method of a conventional display device.

4 is a block diagram schematically illustrating the configuration of a display device to which an error diffusion method for reducing low gray pseudo contours according to the present invention is applied.

FIG. 5 is a diagram illustrating a first lookup table provided in the gamma correction unit 43 shown in FIG. 4.

FIG. 6 is a view showing a second lookup table provided in the recognition luminance reading unit 44 shown in FIG. 4.

7 is a flowchart illustrating an error diffusion method for reducing low gray pseudo contour of the display device according to the present invention.

8 is a view for explaining an error diffusion process when designing the error diffusion method in hardware according to the present invention.

9 is a view for explaining the characteristics of the input luminance vs. recognition luminance in the low gradation region when the error diffusion method according to the present invention is applied.

*** Explanation of symbols for the main parts of the drawing ***

41: A / D conversion unit, 42: addition unit,

43: Gamma Supplementary Government, 44: Awareness and Brightness Reader,

45: error diffusion unit, 46: display control unit,

47: panel.

The error diffusion method for reducing the low gradation pseudo contour of the display device according to the present invention for achieving the above object is an error diffusion method for gradation representation of a display device in which the input gradation versus luminance characteristics are linear, The deviation between the gray level value before gamma correction and the perceived luminance of the gray level value after gamma correction is calculated as a display error, and the display error is weighted with a predetermined weight to diffuse to the surrounding pixels of the pixel of interest.

In addition, the error diffusion method for reducing the low gradation pseudo contour of the display device according to the present invention is an error diffusion method for gradation representation of a display device in which the input gradation versus luminance characteristics are linear, A second step of converting the digital data having the grayscale value of the bit to the digital data having the grayscale value of the bit and adding the display error weighted with a predetermined error diffusion coefficient to output the first input grayscale; And a third step of performing image display according to the second input gradation in which the first input gradation is gamma corrected, wherein the display error is the recognition luminance normalized to the range of the second input gradation and the first input gradation. It is characterized by the deviation between.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

4 is a block diagram schematically illustrating a configuration of a display apparatus to which an error diffusion method for reducing low gray pseudo contours according to the present invention is applied.

In FIG. 4, reference numeral 41 denotes an A / D converter for converting an input analog image signal into 8-bit digital data for each channel of R, G, and B colors, respectively, and is digitally output through the A / D converter 41. The data is used as, for example, a gray scale value for displaying 256 gray scales. Reference numeral 42 denotes an adder which adds the output data of the A / D converter 41 and the output data of the error diffusion unit 44 to be described later and outputs the same through the adder 42. The bit digital data will be referred to as a first input grayscale. The first input gray scale represents an input gray scale before gamma correction.

In FIG. 4, reference numeral 43 denotes a gamma correction unit for outputting the gamma correction unit as a second input gradation when the first input gradation is input, and the gamma correction unit 43 is represented by the graph shown in FIG. 5. 1 Gamma correction is performed using the Look Up table. In the first lookup table, a second input gradation value is previously designated to correspond to the first input gradation value divided into 256 steps from 0 to 255, and the gamma correction unit 43 is configured as the first lookup table. And a ROM for storing the data value of. Since the data value of the first lookup table for performing the gamma correction uses a general gamma correction algorithm, a detailed description thereof will be omitted.

In FIG. 4, reference numeral 44 denotes an input luminance after the gamma correction outputted from the gamma correction portion 43, that is, an input luminance connected to an output terminal of the gamma correction portion 43, that is, a recognition luminance corresponding to a second input gray scale. As a recognition luminance reading unit for reading and outputting, the recognition luminance reading unit 44 uses the second look-up table shown in the graph shown in FIG. 6 to determine the recognition luminance for the second input grayscale. Read and output.

The recognition luminance reading unit 44 includes a ROM for storing data values of the second lookup table, and the data values of the second lookup table are 256 as shown in FIG. 6. The recognition luminance value normalized in 256 steps is specified in advance corresponding to the second input gray level of the step. Since the algorithm for calculating the recognition luminance value corresponding to the input gray scale value uses a general recognition luminance calculation method, detailed description thereof will be omitted.

On the other hand, as shown in (B) of FIG. 1, the relationship between the input gradation and the recognition luminance after gamma correction generally has a linear characteristic, and the input gradation after the gamma correction having 256 stages as shown in FIG. Normalized by, the perceived luminance can be compared with the input gradation before gamma correction. In this case, as shown in FIG. 6, the graph of the second lookup table shows a large luminance deviation between adjacent input gradations in the low gradation region as it has a sharp slope in the low gradation region.

In FIG. 4, reference numeral 45 denotes a display error of a deviation between the first input gradation output from the adder 42 and the recognition luminance output from the recognition luminance reading unit 44, and spreads a predetermined error in the display error. The error diffusion unit 45 multiplies a coefficient and outputs the output error to an input terminal of the adder 42. The error diffusion unit 45 displays a display error in neighboring pixels B to E adjacent to the pixel A of interest as shown in FIG. Diffused.

In FIG. 4, reference numeral 46 denotes a second input gradation output for each channel of the R, G, and B colors from the error diffusion unit 46, for example, based on the subfield information. A display control unit which controls a digital image to be displayed by applying drive pulses such as an address pulse and a scan pulse to a PDP panel described later.

The subfield information is information indicating whether each pixel is turned on / off at one subfield time of the PDP display image. The display controller 46 has a frame memory therein to display the subfield information for each pixel, Each vertical line, subfield, and frame are stored. In addition, the display control unit 46 is configured to include an electrode driver (not shown) for applying the driving pulse to the PDP panel described later.

In FIG. 4, reference numeral 47 denotes a plurality of scan / sustain electrodes and sustain electrodes (not shown) arranged in parallel with each other, and a plurality of address electrodes (not shown) define a predetermined space with the scan / sustain electrodes and sustain electrodes. It is a panel of a PDP which is arranged orthogonal to each other and performs image display in accordance with a predetermined driving pulse applied from the display control unit 46.

In other words, the present embodiment according to the above-described configuration defines a deviation between the normalized recognition luminance output from the recognition luminance reading unit 44 and the first input gradation as a display error, and displays the display error as a display pixel (the main pixel). To spread to surrounding pixels. According to this method, since the average gradation value of the peripheral pixels approaches the recognition luminance after gamma correction, it is possible to reduce the occurrence of the contour error in the low gradation region of the display image.

Hereinafter, an error diffusion method for reducing the low gray pseudo contour of the display device according to the present invention will be described in detail with reference to FIG. 7.

First, when an analog image signal is input to the PDP device (step ST701), the A / D converter 41 of FIG. 4 converts the input analog image signal by n bits (for example, 8 bits) for each color of the R, G, and B channels. The digital data is converted into digital data representing the gray scale value of (ST702).

Then, the adder 42 of FIG. 4 adds the output data of the error diffusion unit 45 (data added with a predetermined error diffusion coefficient to the display error) to n-bit digital data output through the A / D converter 41. The sum is output as the first input gray level which is the gray level data before gamma correction. A detailed description of the error diffusion coefficient and the display error will be described later in step ST706 (ST703). The gamma correction unit 43 of FIG. 4 queries the first lookup table of FIG. In response to the first input gradation output from the adder 42, the second input gradation is read and gamma correction is performed on the first input gradation.

On the other hand, the second input gradation output in accordance with step ST704 is input to the display controller 46 of FIG. 4, and the display controller 46 converts the second input gradation into n-bit (eg, 8-bit) subfield information. After that, predetermined drive pulses such as an address pulse and a scan pulse are applied to the panel 47 of FIG. 4 in accordance with this subfield information to perform display control on the digital image.

Meanwhile, the recognition luminance reading unit 44 of FIG. 4 reads and outputs the recognition luminance corresponding to the second input gradation output from the gamma correction unit 43 according to the second lookup table provided therein. At this time, the recognition luminance output from the recognition luminance reading unit 44 of FIG. 4 uses a value normalized to the range of the second input gradation as described above (step ST705).

Thereafter, the error diffusion unit 45 of FIG. 4 calculates an error value between the first input gradation output from the adder 42 and the recognition luminance output from the recognition luminance reading unit 44 as a display error and displays the error. By applying a predetermined error diffusion coefficient to the error and applying it to the input terminal of the adder 42, the next first input gradation output of step ST703 is prepared.

In the above process, the adder 42, the gamma correction unit 43, and the error diffusion unit 45 of FIG. 4 perform predetermined arithmetic processing required for error diffusion. In other words, arithmetic processing performed through the adder 42, the gamma correction unit 43, and the error diffusion unit 45 of FIG. 4 is represented by the following equations (a), (b) and (c), respectively. same.

: Expression of the adder 42

(b) graylevel ' _{i, j} = LUT1 (U _{i, j} ): The equation of the gamma correction unit 43

(c) e _{i, j} = graylevel _{i, j} -LUT2 (graylevel ' _{i, j} ): calculation formula of the error diffusion unit 45

In Equation 1, U is a first input gradation, graylevel is output data (gradation value) of the A / D converter 41 of FIG. 4, graylevel 'is a second input gradation, and e is a recognition luminance and a first input. The display error between gray levels, W denotes an error diffusion coefficient, LUT1 denotes a first lookup table described with reference to FIG. 5, and LUT2 denotes a second lookup table described with reference to FIG. 6. I and k represent rows of pixels displayed on the PDP panel, and j and l represent columns of pixels.

On the other hand, the error diffusion coefficient refers to a weight multiplied by the display error of the pixel A of interest described in FIG. 3, and the weight is used for a predetermined value for each peripheral pixel B to E. FIG. At this time, the weight is 7/16 for the peripheral pixel (B), 3/16 for the peripheral pixel (C), 3/16 for the peripheral pixel (D), and 5/16 for the peripheral pixel using the Floyd-Steinberg coefficient. It may be desirable to set 1/16 as a weight for the pixel E. Since the Floyd-Steinberg coefficient is a general coefficient value used for image display, a detailed description thereof will be omitted.

In the present embodiment, for convenience of explanation, the error diffusion process is described as the display error is spread from the pixel A of FIG. 3 to the peripheral pixels B to E. However, the configuration of FIG. 4 is designed as hardware. As shown in Fig. 8, the error is added to the pixel A of interest from each of the peripheral pixels B 'to E'. This is equivalent to the difference depending on whether the reference is the pixel of interest A or the peripheral pixels B to E. On the other hand, the error diffusion coefficients of the peripheral pixels B 'to E' are applied to the error diffusion coefficient values of the peripheral pixels B to E as they are.

9 illustrates the characteristics of the first input gradation versus the recognition luminance in the low gradation region of the display image when the error diffusion method according to the present embodiment is applied. For example, the low gradation region having the first input gradation ranging from 0 to 35 is shown. In this paper, the luminance reproducibility of the displayed image is expressed as experimental data. In FIG. 9, reference numeral a denotes a target recognition luminance graph without contour errors in a low gradation region, b denotes a recognition luminance graph when only gamma correction is performed, and c denotes a recognition luminance when the error diffusion method according to the present embodiment is performed. It is a graph. Therefore, when the error diffusion method according to the present invention is applied, a smooth luminance reproducibility in which the pseudo contour is close to the target recognition luminance in the low gradation region as shown in b of FIG. 9 can be obtained.

In other words, the above-described embodiment defines a deviation of the perceived luminance between the grayscale value before the gamma correction (first input grayscale) of the input image signal and the grayscale value after the gamma correction (second input grayscale) as the display error. The error diffusion of neighboring pixels is performed on the basis of this method. The average gradation of the neighboring pixels can be matched to the perceived brightness after gamma correction, thereby reducing pseudo contours generated at low gradations and improving luminance reproducibility. You can do it.

As described above, according to the present invention, in a display device in which input gradation to luminance characteristics are linear as in a PDP apparatus, pseudo contours occurring in a low gradation region of a display image can be reduced and the luminance reproducibility can be improved. It is possible to provide an error diffusion method for reducing the low gray pseudo contour of the display device.

Claims (4)

In the error diffusion method for gradation representation of a display device in which the input gradation to luminance characteristics are linear, The difference between the gray level value before the gamma correction of the input image signal and the perceived luminance of the gray level value after the gamma correction is calculated as a display error, and the display error is weighted by a predetermined weight to diffuse to the surrounding pixels of the pixel of interest. Error diffusion method for reducing the low gray pseudo contour of the display device. In the error diffusion method for gradation representation of a display device in which the input gradation to luminance characteristics are linear, A first step of converting an analog image signal input from the outside into digital data having an n-bit gray value; A second step of adding a display error weighted with a predetermined error diffusion coefficient to the digital data having a gray value of n bits and outputting the first input grayscale; A third step of performing image display according to the second input gradation gamma-corrected to the first input gradation, And the display error is a deviation between the recognition luminance normalized to the range of the second input gradation and the first input gradation. The method of claim 2, When the pixels in row _{i and} column j of the display device are A _{i, j} The display error weighted by the error diffusion coefficient is a first pixel of A _{i, j + 1} , a second pixel of A _{i + 1, j-1} , a third pixel of A _{i + 1, j} , A _{i + 1, An} error diffusion method for reducing low gray pseudo contours of a display device, wherein the fourth pixel of _{j + 1} is diffused to each other. The method of claim 3, wherein The error diffusion coefficients diffused into the first to fourth pixels are 7/16, 3/16, 5/16, and 1/16, respectively. .