KR100508306B1 - An Error Diffusion Method based on Temporal and Spatial Dispersion of Minor Pixels on Plasma Display Panel - Google Patents

Abstract

According to the present invention, when the error diffusion method is applied to solve the low gray level reproduction problem caused by the inverse gamma correction in the plasma display (PDP), it is possible to reduce the unpleasant noise by suppressing the redundant light emission for each color channel at the same pixel position. Performing inverse gamma correction on 256 gray levels of each of the R, G, and B channels with respect to the R, G, and B input values of the PDP; Dividing the result of performing the inverse gamma correction into an integer part and a fractional part; Converting the separated fractional part and the weighted error transmitted from the already processed pixel into an updated fractional part; Determining a dither matrix to be used by the vertical and horizontal signals of the frame and the R, G, and B channels and determining a threshold in the dither matrix; Comparing the updated fractional part with the threshold to perform binarization; Adding the binarized result value to the integer part and displaying the result in a PDP; And calculate an error between the updated subdivision value and the binarized result value and process the weighted error weighted by the error diffusion weight determined according to the threshold value used in the already processed pixel and the sign of the calculated error. Delivering to surrounding pixels.

According to the present invention, when the error diffusion method is applied in the PDP, the image is smoothly represented in the dark area by reducing the number of visible pixels, and the noise is reduced as the luminance difference between the neighboring pixels and the minor pixels decreases.

Description

An Error Diffusion Method based on Temporal and Spatial Dispersion of Minor Pixels on Plasma Display Panel}

The present invention relates to a plasma display (Plasma Display or Plasma Display Panel; PDP), and more particularly to the same pixel when the error diffusion method is applied to solve the low gray level reproduction problem caused by inverse gamma correction in a plasma display. The present invention relates to an error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a plasma display which reduces undesired noise by suppressing overlapping light emission for each R, G, and B channel at a position.

Plasma display (PDP) is a display device connected to various video signals (TV, PC, VCR, HDTV, etc.) to receive and reproduce video and audio data. Especially, in the case of color video, most of R, G, B video signals Will be delivered. The color image of R, G, and B represents 256 steps from 0 to 255, respectively. The color image signals of R, G, and B are gray level, or brightness, of various images by pulse number modulation. Is displayed.

As can be seen from the comparison graph of FIG. 1, in a conventional TV or monitor using a CRT, a luminance relationship outputted to an input gray scale value has a gamma curve. On the other hand, unlike the CRT, since the luminance characteristic of the input grayscale value is almost linear within the operating range, the plasma display has an inverse gamma correction to reproduce the image having the same image quality as that of the CRT in the plasma display panel (PDP). Should be done.

When the inverse gamma correction is performed, the gray scale problem may occur especially in a dark area due to the decrease in the number of gray scales that can be expressed. That is, as the number of gray scales decreases, the low gray scale portion is not faithfully reproduced, and thus the number of gray scales that can be expressed in a dark area is not sufficient, thereby generating a pseudo outline in which images are aggregated.

The results are shown in the accompanying FIG. 2, where the target luminance A, which maintains nonlinearity, represents the ideal inverse gamma result to be corrected, and the actual luminance B is measured as the result after inverse gamma correction. It is a luminance value, and plasma display luminance (C) shows the luminance value 3 or less measured in the absence of inverse gamma correction.

In the case of performing substantially inverse gamma correction, the actual luminance B should ideally match the target luminance A. As shown in FIG. 2, the gray scale value is used for the target luminance A aimed at nonlinearity. Each of the 61 steps from 0 to 60 should be expressed by different luminance values, but since the luminance of the actual plasma display panel represents only seven luminance values, the actual luminance B cannot be represented in the form of steps. Therefore, even if the inverse gamma correction is performed in the plasma display panel, due to the luminance characteristic of the actual luminance having the step shape, sufficient gray scale expression cannot be expressed in the dark region, which causes pseudo contours to remain in the displayed image.

Background Art Conventionally, an error diffusion method has been introduced as a method for solving the gray scale expression problem generated during inverse gamma correction. In the conventional error diffusion method, in order to reduce the gray level expression problem and noise due to the error diffusion, multiple error diffusion weights are alternately used for each pixel position, frame, or line (Korean Patent Publication No. 2003-33754). By defining the difference between the target luminance and the actual luminance in the pixel as an error, the average grayscale value of a certain region is matched with the target luminance so that pseudo contours do not occur in the dark region. 2002-18900), the positions of surrounding pixels in which errors are spread are alternately used for each pixel position, frame or line (Japanese Patent Laid-Open No. 2002-135608).

As described above, in the conventional error diffusion method, the difference between the target luminance and the actual luminance is defined as an error to be considered in the surrounding pixels so that the average gradation value of the predetermined region matches the target luminance. However, when the conventional error diffusion method is applied, there is a problem in that unobtrusive noise is generated due to the nonuniformity of the displayed pixel position. For example, assume that inverse gamma correction is performed by Equation 1 below.

[Equation 1]

In Equation 1, x (m, n) is an input gray value and has an integer value between 0 and 255 when the input is 8 bits. Γ represents a preset gamma value. In addition, G (m, n) on the left side represents a target gradation value having an integer part and a fractional part.

When the input image is a single gray scale image having a size of k × k, the result of inverse gamma correction has a constant value regardless of the position of the pixel. In this case, it is assumed that the target gray value is β.xxx. In order to match the average gradation value displayed on the PDP with the target gradation value due to the error diffusion, an integer gradation β is displayed in some of the k × k pixels and an integer gradation (β + 1) in the remaining pixels.

If the fractional value of the target gray value is smaller than 0.5, the number of pixels having an integer gray level (β + 1) is smaller than the number of pixels having a value β. If the value of the fractional part of the target gradation value is greater than 0.5, the number of pixels having an integer gradation β is smaller than the number of pixels having an integer gradation (β + 1). Among the pixels having an integer gradation β and (β + 1) for a given fractional value, the smallest number of pixels is referred to as a fractional pixel.

If the luminance difference between the integer gradation β and (β + 1) is smaller than the distinction limit of the human visual system, the fractional pixel is not recognized as the noise in the error diffusion result image, and on the average, it displays the same value as the target gradation value. However, because one gray level difference in the dark area of the PDP is much larger than the human noticeable difference (just noticeable difference), even if the average luminance of the image resulting from the error diffusion matches the target luminance, the position of the fractional pixel is relatively brighter. Or as a blacker dot. That is, it is recognized as noise. When the same error diffusion method is applied to each of the R, G, and B color channels, the hydrophobic pixels emit light at the same pixel position, and when the plurality of color channels emit the hydrophobic pixels at the same position, it is more easily recognized as noise to human vision. will be.

The present invention has been proposed to solve the above problems, and when the error diffusion method is applied to solve the low gray level reproduction problem caused by the inverse gamma correction, as described above, in the single channel, a time- or frame-based fractional pixel In a plasma display that can reduce undesired noise by avoiding overlapping positions of the R, G, and B color channels, and by preventing a few pixels of a plurality of color channels from overlapping light at the same pixel position in the same frame within the R, G, and B color channels. The purpose is to provide an error diffusion method based on spatiotemporal hydrophobic pixel dispersion.

In order to achieve the above object, an error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a plasma display according to the present invention is performed inversely with respect to 256 gray levels for each of the R, G, and B channels for the R, G, and B input values of the plasma display. Performing gamma correction; Dividing the result of performing the inverse gamma correction into an integer part and a fractional part; Converting the separated fractional part and the weighted error transmitted from the already processed pixel into an updated fractional part; Determining a dither matrix to be used by the vertical and horizontal signals of the frame and the R, G, and B channels and determining a threshold in the dither matrix; Comparing the updated fractional part with the threshold to perform binarization; Adding the binarized result value to the integer part and displaying the result on a plasma display; And calculate an error between the updated subdivision value and the binarized result value and process the weighted error weighted by the error diffusion weight determined according to the threshold value used in the already processed pixel and the sign of the calculated error. Delivering to surrounding pixels.

Hereinafter, an error diffusion method based on spatiotemporal hydrophobic pixel dispersion for gray scale representation of the present invention will be described in detail with reference to the accompanying drawings.

3 is a block diagram illustrating an error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a PDP for smooth grayscale expression in a dark region of a plasma display (PDP) according to the present invention. Referring to FIG. 3, when the R, G, and B values are input, the inverse gamma correction unit 10 performs inverse gamma correction. The inverse gamma correction unit 10 includes an inverse gamma correction LUT (Look Up Table) and the contents of the inverse gamma correction LUT are calculated by Equation 1 above.

The target gray value G (m, n) calculated through Equation 1 is divided into an integer part G _i (m, n) and a fractional part G _f (m, n). The fractional value G _f (m, n) calculates the errors caused by the already processed pixels (40) and adds the updated fractional value UP by adding the errors after they have been weighted by the error diffusion weight (50). converted to _f (m, n). The fractional value UP _f (m, n) is determined by comparing the updated value UP _f (m, n) and the threshold value TH _x (m, n) determined by the threshold comparison unit 30. _If greater than _x (m, n), 1 is output; if the fractional value UP _f (m, n) is less than the threshold value TH _x (m, n), 0 is output. The binary output value O _f (m, n) thus output is added to the integer portion G _i (m, n) and displayed on the PDP. Here, the dither matrix 80 to be used is determined by the vertical horizontal signal 70 of the frame and the R, G and B color channels, and the threshold value in the determined dither matrix 80 is determined. On the other hand, the difference between the binary output value O _f (m, n) and the updated fractional value UP _f (m, n) is defined as an error for error diffusion, and the threshold used for the already processed pixels and the defined error According to the sign of the error diffusion coefficient determiner 60 determines the error diffusion weight 50. Subsequently, the defined error is weighted by the error diffusion weight 50 and transferred to the surrounding pixels to be predetermined in the future to achieve a uniform dot distribution. This error diffusion method is represented by Equation 2.

[Equation 2]

A method of determining the threshold TH _x (m, n) will be described in Equation 2 above. In the present invention, a dither matrix is used as the threshold used in the threshold comparison unit 30. The dither matrix used is a 2x2 matrix containing two elements, 0.9 and 0.1, respectively. A 2x2 dither matrix is arranged so as not to overlap an image of a given frame, and 0.9 or 0.1 is used as a threshold depending on the pixel position. Pixels using the threshold 0.9 have a higher probability that the output value of the error diffusion part O _f (m, n) becomes 0. Pixels using the threshold value of 0.1 will have the output value O _f (m, n) of 1 being the error diffusion part. The probability increases. The dither matrix used as the threshold is shown in Table 1 below. As shown in Table 1 below, the dither matrix is reversed for each frame, and the dither matrix is reversed between the GREEN channel with the highest luminance and the RED and BLUE channels so that the decimal pixels do not overlap between the frames and the channels.

TABLE 1

channel TH _x (m, n) Odd frames Even frame RED GREEN BLUE

Next, a method of determining the error diffusion weight w (k, l) used in Equation 2 will be described. When the threshold value is determined as shown in Table 1, overlapping of the fractional pixel of each pixel or frame channel (G and R or B) of the minority pixel is reduced. However, the error spreading weight 50 is determined according to the threshold and the sign of the error used to further reduce the overlap of the fractional pixels.

That is, when a pixel using a threshold value of 0.1 receives a positive error from the surrounding pixels that have already been processed, the probability that the output value O _f (m, n) of the error diffusion part outputs 1 may be greater, but the value of 0.9 When the threshold value is used, if the pixel receives a positive error from the surrounding pixel, the probability of outputting 0 is reduced. On the contrary, if a pixel using a threshold value of 0.9 receives a negative error from the surrounding pixels, the probability of outputting 0 increases. If a pixel using a threshold value of 0.1 receives a positive error from the surrounding pixels, the value 1 is outputted. The probability is lowered. Therefore, when a positive error occurs in a pixel having a dither matrix value of 0.9 and a negative error occurs in a pixel having a dither matrix value of 0.1, a large amount of error is transmitted in the arrow direction of FIG. 4 (a). On the contrary, when a negative error occurs in a pixel having a dither matrix value of 0.9 and a positive error occurs in a pixel having a dither matrix value of 0.1, a large amount of error is transmitted in the arrow direction of FIG. 4 (b). Table 2 below shows error diffusion coefficients used according to error values and dither matrix values.

TABLE 2

Error Dither Matrix Values 0.9 0.1 Positive negative

FIG. 5 is an image diagram of a digital camera photographing a result of directly displaying a plasma display (PDP) by applying an error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a plasma display according to the present invention. 5 (a) is a display image in the PDP when the error diffusion method is not applied, Figure 5 (b) is a display image in the PDP when the conventional error diffusion method of the Flyod-Steinberg method is applied 5 (c) is a display image of a PDP for a case where an error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a PDP according to the present invention is applied. Referring to Figure 5 (a) ~ (c), it can be seen that the step phenomenon when the error diffusion method is not applied is reduced when the conventional error diffusion method of the Flyod-Steinberg method is applied. However, when the error diffusion method based on the spatiotemporal hydrophobic pixel dispersion according to the present invention is applied, the error difference method of the conventional Flyod-Steinberg method is applied by reducing the luminance difference from the surrounding pixels by not overlapping the hydrophobic pixels between frames and channels. It can be seen that the noise is reduced more than the case.

As described above, in the present invention, when the error diffusion method is applied to solve the low gradation representation problem caused by performing the inverse gamma correction in the plasma display, the threshold diffusion used in the binarization process among the error diffusion methods is used. When using an element of a predetermined dither matrix and transferring the error generated by the binarization to surrounding pixels to be processed in the future, an error diffusion weight that is different depending on the threshold used for the processed pixel and the sign of the calculated error is used. This suppresses redundant light emission for each of the R, G, and B color channels at the same pixel position, thereby reducing undesired noise.

The above detailed description and contents of the drawings relate to embodiments for describing the present invention, but the present invention is not limited thereto. In particular, the components of the block diagram for applying the error diffusion method according to the present invention described in the detailed description and drawings may be substituted, modified or changed within the scope of the technical idea of the present invention.

Accordingly, the scope of the present invention should be determined by the appended claims rather than by the foregoing description and drawings.

According to the present invention, by not overlapping the hydrophobic pixels between frames and channels in the plasma display, the number of hydrophobic pixels generated when the error diffusion method is applied is reduced, thereby rendering the image smoothly in the dark region.

In addition, according to the present invention, as the luminance difference between the neighboring pixels and the hydrophobic pixels is reduced in the plasma display, there is an effect of reducing noise due to error diffusion.

1 is a graph comparing luminance characteristics of a plasma display and a CRT with an input gray scale value.

2 is a graph showing the result of the inverse gamma correction and the target luminance for the low gradation region.

3 is a block diagram illustrating an error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a PDP according to the present invention.

4 is an exemplary diagram of a dither matrix showing a direction in which a large number of error values are transmitted in order to increase a probability of generating a desired minority pixel according to the present invention.

5 is a display image diagram of a PDP according to an application of the present invention.

 Explanation of symbols on the main parts of the drawings

10: reverse gamma correction unit 20, 40, 90: addition unit

30: threshold comparison unit 50: error diffusion weighting

60: error diffusion coefficient determiner 70: vertical and horizontal signals for each frame and channel

80: dither matrix determination unit

Claims (5)

Performing inverse gamma correction on 256 gray levels of each of the R, G, and B channels with respect to the R, G, and B input values of the plasma display; Dividing the result of performing the inverse gamma correction into an integer part and a fractional part; Converting the separated fractional part and the weighted error transmitted from the already processed pixel into an updated fractional part; Determining a dither matrix to be used by the vertical and horizontal signals of the frame and the R, G, and B channels and determining a threshold in the dither matrix; Comparing the updated fractional part with the threshold to perform binarization; Adding the binarized result value to the integer part and displaying the result on a plasma display; And Calculate the error between the updated subdivision value and the binarized result and surround the weighted error weighted by the error diffusion weight determined according to the threshold used in the already processed pixel and the sign of the calculated error. Delivering to a pixel; Spatiotemporal hydrophobic pixel dispersion based error diffusion method in a plasma display comprising a. The method of claim 1, An error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a plasma display, using elements of a dither matrix having a predetermined number of values as thresholds. The method according to claim 1 or 2, The dither matrix is a 2 x 2 matrix including two elements each of two elements, wherein the error diffusion method based on the spatial and temporal hydrophobic dispersion in the plasma display. The method of claim 1, An error diffusion method based on spatiotemporal hydrophobic pixel dispersion in a plasma display, characterized by having different error diffusion weights for each frame and color channel according to the threshold value of the dither matrix. The method of claim 1, A method for spreading errors based on spatiotemporal hydrophobic pixels in a plasma display, characterized in that the dither matrix is inversely used between a G channel and an R or B channel to prevent the pixel from intermingling between frames and channels.