KR100510652B1 - Method for bit-splitting of Digital Light Processing system - Google Patents

Abstract

The invention DLP (Digital Light Processing) system, a bit-plane (bit plane) relates to a bit-plane resolution method of a di LP system so as to improve the image quality by improving the resolution method, intensity, separated pieces of the 2 ^N -1 N bits with N (N = 1,2,3, ..., N) For the bit division of a DLP system, use the central region (m) of the bit plane divided into 2 ^N -1 for the first bit division Use one of the first region, 2 ^N -1st region, the second bit, the third bit division, and use 4,5,6, ....., 2 ^N -1 1st division Select the area in the middle of the areas and divide them sequentially.If the area you want to use is adjacent to the area already used, select the area where the adjacent number of areas used for bit division is minimized and use it for bit division. Characterized in that.

Description

Bit plane splitting method of DLP system {Method for bit-splitting of Digital Light Processing system}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital light processing (DLP) system, and more particularly, to a bit plane splitting method of a DLP system capable of improving image quality by improving a bit plane splitting method.

DLP devices are the next generation of projection TV and projector devices because they have better contrast than LCD and CRT devices that are currently being used, and the optical structure is much simpler to reduce the size and weight of the set. It is attracting attention as.

The DLP system reflects light from a lamp (RAMP) into a number of small mirrors and displays them. Each mirror serves as a pixel, and includes a digital micromirror device (DMD) in which the mirrors are configured.

Hereinafter, a DLP system according to the related art will be described with reference to the accompanying drawings.

1 is a schematic configuration diagram for grayscale and color control in a general DLP system, and FIG. 2 is a format diagram illustrating a bit plane configuration according to pixel values of a DLP system.

The DLP device reflects light from the lamp using an extremely fine mirror made using MEMS (Micro Electro Mechanical System) technology to control the brightness of the pixel corresponding to each mirror.

At this time, since the brightness of the light alone cannot express the color, the light reflected from the mirror passes through the color wheel and has both gradation and color.

The color wheel is composed of R, G, B or R, G, B, W (White), depending on which part of R, G, B or R, G, B, W passes the light reflected from the mirror. It will express the color.

One of these R, G, B, or R, G, B, W is included in one color wheel is a one-plate color wheel, each color wheel is configured to be a three-plate color wheel.

Although not shown in FIG. 1, a color wheel is positioned between the light source and the DMD.

According to the angle of DMD, it can be classified into “Flat State”, “Off State”, and “On State”. In “Flat State”, the angle of DMD is 0 ° and the light from the light source has no effect. to be.

And "On State" is a state in which the angle of the DMD is 10 °, that is, the light from the light source is output to the screen through the projection lens through the DMD.

"Off State" is a state in which the angle of the DMD is -10 ° so that the light enters the heat sink (not shown) to prevent the temperature inside the set from rising.

The DLP device formats the video signal into a specific video format called a bit plane and then delivers it to the Digital Micromirror Device (DMD) panel as a Low Voltage Differential Signaling (LVDS) signal.

The bit plane is expressed in the form of pulse width modulation (PWM).

For example, as shown in FIG. 2, in the case of a 5-bit DLP system, each pixel is represented by a 5-bit number.

The pixel value 1 is 1/31 (3.2%) of the full lumen output, the pixel value 16 is 16/31 (51.6%) of the full lumen output, and the pixel value 26 is 26/31 (84%).

That is, the mirror duty cycle during the refresh time, that is, the time when the mirror reflects light and is reset, determines the light intensity.

Here, the time consumed per given Bit can be calculated as follows.

Here, b denotes a given bit, F denotes a frame rate, n denotes the number of bits b appears during the refresh period, and B denotes the total number of bits.

For example, in the case of the 0th bit, if the frame rate is 60Hz and the number of times the bit b appears during the refresh period is 1, the time consumed in the 0th bit is calculated as follows.

That is, 5.37 * 10 ^-4 seconds.

The human eye must have afterimages to achieve the desired intensity of light. That is, the retina needs about 1/60 seconds or more to obtain all the intensity of light.

In case of driving PWM (Pulse Width Modulation) in DLP system, Contour Noise occurs due to the afterimage effect of retina.

Hereinafter, generation of contour noise due to bit plane splitting and retinal afterimage effects in a DLP system of the related art will be described.

3A and 3B are diagrams for explaining the generation of contour noise due to the retinal afterimage effect in a 4-bit DLP system.

4A is a diagram of a bit plane without bit division, and FIG. 4B is a diagram of a bit plane with bit splitting in the refresh period of the prior art.

If one pixel has 7/15 (47%) intensity in a 4-bit DLP system, the PWM at this time is 7/15, and thus has the same structure as in FIG. 3A.

Here, part (a) is a PWM part which actually turns on the mirror.

However, the intensity of the pixel is 8/15 (53%) in the t1 part of FIG. 3B, and the bright flash is generated due to the afterimage effect of the retina.

This is due to the fact that the retina is perceived to be much larger than the intensity of 8/15, which is 7/15, since 7/15 intensity still remains as an afterimage, and 7/15 intensity is added to 8/15 intensity.

Bright flashes like this are often found on screens where the gradation changes smoothly, such as the outline of a face or the setting sun.

That is, the portion having such a smooth gradation is distorted, so that the portion appears to be greatly different in gradation.

In order to eliminate such distortion of the screen, a technique of bit-spliting, that is, dividing a bit plane into a plurality of divided bits and disposing them widely during a refresh time is required.

The bit division arrangement of the prior art is made as follows.

The bit plane of FIG. 4A shows an intensity structure without bit division.

As described above, when the bit division is not performed, contour noise is caused by the bright flash as described above.

As a method for removing such a cause of contour noise, bit division is performed as shown in FIG. 4B.

However, in the bit plane structure in which bit division is performed as in FIG. 4B, the contour noise is not completely removed.

That is, contour noise can be completely removed only if the regions used for bit division are not as close as possible, but there is a limit in eliminating contour noise since bit division is not considered at all.

However, the bit plane splitting method in the prior art DLP system has the following problems.

In the bit plane structure of the prior art, a bright flash is generated due to the afterimage effect of the retina in a region where the intensity is concentrated.

Such bright flash distorts portions having a smooth gradation, so that the portions appear to be significantly different in gradation.

This is a major cause of image quality deterioration.

The present invention is to solve the problems of the prior art DLP system, to provide a bit plane segmentation method of the DLP system to improve the image quality by improving the bit plane segmentation method. have.

In order to achieve the above object, a bit plane splitting method of a DLP system according to the present invention includes an N-bit (N = 1,2,3, ..., N) DLP system having an intensity divided by 2 ^N −1. For bit division, use the center region (m) of the bit plane divided into 2 ^N -1 for the first bit division, and the second bit, the third bit division for the first region, and the 2 ^N -1 region Use one by one, and divide 4,5,6, ....., 2 ^N -1st bit division sequentially by selecting the area in the middle of unused areas, and use the area you want to use. In the case of adjoining the divided region, an area in which an adjacent number of regions used for bit division is minimized is selected and used for bit division.

Hereinafter, a bit plane splitting method of a DLP system according to the present invention will be described in detail with reference to the accompanying drawings.

5A to 5O are block diagrams illustrating bit splitting methods of a DLP system according to the present invention.

The present invention is to eliminate the contour noise of the DLP system, which is spotlighted as a next-generation panel in a projection TV or a projector. This is to improve image quality by dividing properly.

Bit division of the present invention is performed by the following principle to completely remove contour noise.

In performing bit-spliting in N-bit DLP system, the total intensity can be divided into 2 ^N -1. When each area is divided based on the center area (m) located in the center of the whole bit plane, As follows.

That is, 1,2,3, ..., (m-2), (m-1), (m), (m + 1), (m + 2), ..., (2 ^N -2) , (2 ^N -1).

Here, the same number of areas are provided to the left and right with respect to the center area m.

First, the first bit division must use the center region (m) of the bit plane.

Second, in the second and third bit divisions, bit division is selected by selecting one of the regions located farthest from the central region m, that is, the first region and the ( ^2N- 1) th region.

Third, when the fourth bit split from the last bit split (2 ^N -1), the bits are divided by selecting regions located in the middle of regions not used for bit splitting.

This is for selecting and using an area where the area to be used for bit division is as far as possible from the areas already used for bit division.

Fourth, if the area to be used for bit division is adjacent to the area already used when bit division using the above-mentioned third principle (only the unused area is adjacent to the area already used) The area of the part with the smallest number of adjacent areas is used for bit division.

Using this principle, bit division of a 4-bit DLP system is described as follows.

In this case, the total intensity of the bit plane can be distinguished from 2 ^N -1, that is, 15.

If the pixel has an intensity of 1/15 (6.7%), bit division is performed using the eighth region, which is the center region m, as shown in FIG. 5A.

If the pixel has an Intensity of 2/15 (13.3%), the first region is used for bit division as shown in FIG. 5B.

Likewise, pixels having an intensity from 3/15 (20%) to 15/15 (100%) also have the 15th region, the 4th region, the 12th region, the 6th region, and the 1st region as shown in FIGS. The bit division is performed by using the order of the tenth, third, thirteenth, seventh, eleventh, fifth, fourteenth, second, and ninth regions.

As such, the first bit is located in the eighth region, which is the central region m of the bit plane, and the second bit is located in the first region or the fifteenth region, which is the rightmost or leftmost region of the bit plane.

Similarly, the third bit is located in the fifteenth region or the first region, which is the leftmost or rightmost side of the bit plane not used for bit division of the second bit.

The fourth bit is located in the middle region m and the rightmost middle region or the middle region m and the leftmost middle region of the bit plane.

And the fifth bit is located in the middle region m and the leftmost middle region or the middle region m and the rightmost middle region of the bit plane which are not used for bit division of the fourth bit.

In this way, all bits are divided.

This bit division method can be applied to a DLP system of 4 bits or more as it is.

In addition, in the bit division embodiment of the present invention shown in Figs. 5A to 5O, the following subdivision criteria are performed.

That is, after finally using the region adjacent to the first region and the 2 ^N -1 th region sequentially for bit division, the region adjacent to the right side m + 1 based on the center region m is used for bit division.

In the bit division, the 2 ^N- 2 th bit division from the second bit division is an odd number (3, 5) in an area symmetrical with the area where the even (2, 4, 6, ...) bit division is made. , 7, ..) Bit division is performed.

If the symmetric area is not the area that minimizes the number of consecutive adjacent used areas, another area is used for bit division.

In the bit division of the present invention as described above, the most important principle is to select the area of the part spaced apart as far as possible from the area already used for the bit division.

This is to completely eliminate the contour noise caused by the bright flash by minimizing the density of the areas used for the bit division.

The bit plane splitting method of the DLP system according to the present invention has the following effects.

In the PWM driving method used in the DLP system, the density of the regions used for the bit division is minimized as much as possible during the bit division of the bit plane.

 This has the effect of improving the image quality by completely removing the contour noise caused by the bright flash.

1 is a schematic configuration diagram for grayscale and color control in a general DLP system

2 is a format diagram illustrating a bit plane configuration according to pixel values of a DLP system;

3A and 3B are diagrams for explaining the generation of contour noise due to the retinal afterimage effect in a 4-bit DLP system.

4A is a bit plane configuration diagram without bit division

FIG. 4B is a diagram of a bit plane configuration using bit splitting in a refresh period of a prior art. FIG.

5A to 5O are block diagrams illustrating a bit splitting method of a DLP system according to the present invention.

Claims (5)

In the bit splitting of an N bit (N = 1,2,3, ..., N) DLP system having intensity divided into 2 ^N -1 through a bit plane driven in a PWM form, Use the middle area (m) of the bit plane separated by 2 ^N -1 for the first bit split, The second bit, the third bit division, the first region, one of the 2 ^N -first region, and one of the 4,5,6, ....., 2 ^N -the first bit division unused region Select the areas in the middle of the field If the area to be used is adjacent to the area already used, the bit plane partitioning method of the DLP system, characterized in that for selecting the area to minimize the adjacent number of areas used for bit division is used for bit division. 2. A region according to claim 1, wherein the region adjacent to the first region and the ^2N- 1 th region is finally used for bit division, and then the region adjacent to the right side (m + 1) with respect to the center region (m) is used for the bit division. Bit plane segmentation method of a DLP system, characterized in that used. 2. The method of claim 1, wherein the 2 ^N- 2 th bit divisions from the second bit division are odd (3, 5) in an area symmetric to an area where even (2, 4, 6, ...) bit divisions are made. , 7, ..) Bit plane splitting method of a DLP system, characterized in that bit splitting. 4. The method as claimed in claim 3, wherein another area is used for bit division if the area to be mirrored is not an area that minimizes the number of consecutive adjacent used areas. The method according to claim 1, wherein when the N value is 4, the area of the bit plane used for bit division is divided into 1,2,3, ....., 14,15, Bit plane segmentation method of the DLP system characterized in that the bit partitioning using the sequence of 8,1,15,4,12,6,10,3,13,7,11,5,14,2,9 .