JP2009186800A - Display method and flicker determination method of display device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine flicker by quantifying unevenness of luminance and variation of luminance. <P>SOLUTION: Correction of K gradation is performed for N1 frames in N frames, correction of K+1 gradation is performed for N-N1 frames in N frames, one blcok is used as A subpixels, arrangement numbers determining an order of performing the correction of K gradation and the correction of K+1 gradation in N frames of one period are set as 1 to N, luminance of green subpixels is calculated as L times luminance of red subpixels and blue subpixels, a group in which the number of subpixels for determination and their adjacent subpixels becomes B in total is set in one block, when it is assumed that the correction of K gradation is performed for N-1 frames and the correction of K+1 gradation is performed for one frame, correction of all the subpixels of one block is considered as an allocation pattern to one block of the arrangement numbers that there is time when the average luminance of one subpixel of one group becomes approximately the same as the average luminance of one subpixel in one block at least once in N frames in all the subpixels of one block. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

The present invention relates to a display device that displays on the same screen a screen that can discriminate two individual images according to the viewing direction, and more particularly to a display device that performs crosstalk correction of less than one gradation and a flicker determination method for the display device.

Conventionally, liquid crystal display devices and organic EL display devices have been widely used as display devices mounted on television receivers, information devices, and the like. The color pixels of these display devices are composed of three sub-pixels of RGB. Each sub-pixel has digital gradation data, and the driver converts the gradation data into an analog driving voltage, and the sub-pixel emits light with a luminance corresponding to the driving voltage. Thus, the gradation data that specifies the luminance of the sub-pixel is an integer.

Therefore, the gradation data of the image input to the display device is an integer, but may be a decimal when the gradation data is calculated by crosstalk correction or flicker processing. For this,
There is a problem that a desired luminance and error are generated by making the decimal number an integer.

In order to reduce the error, FRC (Frame Rate Control) has been considered (
See Patent Documents and Patent Document 2). In FRC, by using a plurality of frames with different gradations as one cycle, apparently less than an integer gradation display is performed using an afterimage. For example, when 20 gradations are repeated in 3 frames and 21 gradations are repeated in 1 frame, 20.25 gradations appear due to an afterimage.

However, FRC has a problem that flicker occurs because the luminance changes periodically even in a portion where the image is stationary. In view of this, it has been considered to evenly distribute the timing of frames in which the luminance is increased to the sub-pixel arrangement (see Patent Document 1 and Patent Document 2).
JP 2003-5695 A JP 2005-10520 A

However, there are a plurality of layouts that evenly distribute the timing of frames in which luminance increases to the sub-pixel arrangement, and even with the uniform distribution, there is a difference in the flicker reduction effect depending on the layout. The conventional method for determining which layout is superior is to actually display the layout and visually determine the flicker. For this reason, there is a problem that it takes time for the determination. In addition, there is a problem that there is an individual difference in determination, not objective determination.

The present invention has been made in view of the above-mentioned problems, and is intended to objectively perform flicker determination by quantifying luminance non-uniformity and luminance change, which are flicker elements, by a unique method.

In order to solve the above-described problem, the display device of the present invention includes one pixel composed of three sub-pixels of red, green, and blue, and the gradation display of the luminance of the sub-pixel is set, and the correction target sub-pixel level is set. A crosstalk correction unit that corrects the tone based on the gradation of the adjacent sub-pixel,
The crosstalk correcting unit performs the correction of K (K is an integer) gradation in N (N is a positive integer greater than or equal to 2) N frames (N1 is a positive integer less than N) frame, and the correction of K + 1 gradation is performed. N-
N1 frames are performed, and the subpixel is composed of a plurality of blocks in which one block is M1.
Arrangement numbers that define the order in which the correction of the K gradation and the correction of the K + 1 gradation in N frames of one cycle are set from 1 to N, and the arrangement number 1 assigned to one block of sub-pixels is set.
The luminance of the green sub-pixel in the same gradation is approximately L times the luminance of the red sub-pixel and the blue sub-pixel.
The luminance of the green subpixel is calculated as L times the luminance of the red subpixel and the blue subpixel,
A group in which the number of sub-pixels to be determined and the number of adjacent sub-pixels is a total of M2 is set in one block,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
When the correction of +1 gradation is performed for one frame, the average luminance of one sub-pixel in one group is substantially the same as the average luminance of one sub-pixel in one block for all sub-pixels in one block. This is an assignment pattern of the arrangement number to one block at least once in a frame.

In this way, a group including a sub-pixel to be determined and its neighboring sub-pixels is set so that the average luminance of one sub-pixel in one group is substantially the same as the average luminance of one sub-pixel in the block. Since a high green color is taken into consideration, a pattern with reduced flicker can be obtained.

In addition, one pixel is composed of three sub-pixels of red, green, and blue, the gradation display of the luminance of the sub-pixel is set, and the gradation of the sub-pixel to be corrected is corrected based on the gradation of the adjacent sub-pixel. A crosstalk correction unit
The crosstalk correcting unit performs the correction of K (K is an integer) gradation in N (N is a positive integer greater than or equal to 2) N frames (N1 is a positive integer less than N) frame, and the correction of K + 1 gradation is performed. N-
N1 frames are performed, and the subpixel is composed of a plurality of blocks in which one block is M1.
Arrangement numbers that define the order in which the correction of the K gradation and the correction of the K + 1 gradation in N frames of one cycle are set from 1 to N, and the arrangement number 1 assigned to one block of sub-pixels is set.
A display device in which the numbers from N to N are the same,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
When one frame of correction is performed for +1 gradation, the frame order in which all sub-pixels of one block have the same color sub-pixel of the first adjacent pixel of the determination target sub-pixel as K + 1 gradation, and the determination target The frame order in which the sub-pixel has K + 1 gradation and the frame order in which the sub-pixel of the same color of the second adjacent pixel of the determination target sub-pixel has K + 1 gradation are not sequential.
This is an allocation pattern of the arrangement number to one block.

In this way, since the frame change movement of the high-luminance sub-pixel of the same color is limited, a pattern with reduced flicker can be obtained.

In order to solve the above problem, the flicker determination method of the display device according to the present invention calculates the luminance of the green subpixel as L times the luminance of the red subpixel and the blue subpixel, A group in which the number of adjacent sub-pixels is a total of M2 is set in one block,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
Point when the average luminance of one sub-pixel in one group is substantially the same as the average luminance of one sub-pixel in one block for all sub-pixels in one block when +1 gradation correction is performed for one frame The higher the point, the more the flicker reduction pattern is determined.

Thus, since it quantifies based on average brightness | luminance, the determination of the flicker by nonuniform brightness | luminance can be performed. In addition, since green with high luminance is taken into account, highly accurate quantification can be performed. A pattern with reduced flicker can be obtained by this determination of quantification.

Further, the correction of all the sub-pixels in the one block is performed by correcting the K gradation in N-1 frames.
When the correction of the K + 1 gradation is performed for one frame, all the sub-pixels of one block are
The frame order in which the subpixels of the same color of the first adjacent pixel of the determination target subpixel have K + 1 gradation, the frame order in which the determination target subpixel has K + 1 gradation, and the second adjacent pixel of the determination target subpixel When the three subframes of the same color are in the K + 1 gradation and the three frame sequences are not consecutive, the point is increased, and the higher the point, the more the flicker reduction pattern is determined.

As described above, since the frame change movement of the high-luminance sub-pixel of the same color is quantified, flicker due to a change in luminance can be determined. A pattern with reduced flicker can be obtained by this determination of quantification.

Hereinafter, the best embodiment of the present invention will be described with reference to the drawings. However, the embodiment described below exemplifies a display device for embodying the technical idea of the present invention, and is not intended to specify the present invention. It is equally applicable to the other embodiments included.

FIG. 1 is a block diagram showing the main part of the display device of this embodiment. The solid line in FIG.
The broken line indicates the navigation device 40 in which the display device 1 is incorporated. 2 is a block diagram showing a main part of the liquid crystal display unit 2 of FIG. 1, FIG. 3 is a block diagram showing a main part of the source driver 21 of FIG. 2, and FIG. 4 is a diagram of the positive-polarity DA converter circuit of FIG. It is a figure which shows the principal part.

1, the navigation device 40 includes a navigation unit 41, a DV
A D reproduction unit 42 and a screen switching circuit 43; The navigation device 40 has an internal HDD (
Map information is read out from a map (not shown), and a map image in which position information received from an internal GPS (Global Positioning System) receiving unit (not shown) or a registered route is superimposed on the map information is output. The DVD playback unit reads a DVD image from the DVD and outputs it. The navigation device 40 has a left view direction (passenger seat direction of a Japanese car) and a right view direction (direction of a driver seat of a Japanese car).
Also have a one-screen mode for visually recognizing the same image and a two-screen mode for visually recognizing different images. 1
In the screen mode, only the navigation image or only the DVD image is visually recognized in both the right viewing direction and the left viewing direction. In the two screen mode, the screen switching circuit 43 visually recognizes the navigation image in the right viewing direction and in the left viewing direction. A DVD image is visually recognized, or a DVD image is visually recognized in the right viewing direction and a navigation image is visually recognized in the left viewing direction. These four screens are switched by the operation of the user, and the screen switching signal is output.

In the configuration of the solid line in FIG. 1, the display device 1 outputs the two source images (navigation image and DVD image) from the liquid crystal display unit 2 and the navigation device 40 to the liquid crystal display unit 2 by performing two-screen composition processing and crosstalk correction. Signal processing circuit 3, EEPROM 4 for storing various data necessary for the operation of the signal processing circuit 3, and power supply circuit 5 for supplying power to the liquid crystal display unit 2
It has. The signal processing circuit 3 includes a two-screen composition unit 6 that synthesizes two source images in the two-screen mode, a crosstalk correction unit 7 that performs crosstalk correction corresponding to each of the one-screen mode and the two-screen mode, An output signal generation unit 8 that controls polarity and timing so that the signal corrected by the talk correction unit 7 can be displayed on the liquid crystal display unit 2, and an EEPROM controller 9 that controls input and output of the EEPROM 4 are provided.

The crosstalk correction unit 7 includes a preprocessing unit 10, a correction data sending unit 11, and a calculation unit 12.
The preprocessing unit 10 sends necessary data from the image signal from the two-screen composition unit 6 to the correction data sending unit 11.
And sent to the arithmetic unit 12. The correction data sending unit 11 includes a LUT (lookup table) 13 that stores a correction table from the EEPROM controller 9 and a data interpolation unit 14 that interpolates data that is not in the LUT 13 to obtain correction data. The calculation unit 12 is a preprocessing unit 10.
The correction gradation corresponding to the correction data of the correction data sending unit 11 is added to the image from the above, and the FRC process corresponding to each of the one-screen mode / two-screen mode is performed. In the FRC process, four frames are defined as one cycle, and a predetermined sub-pixel of a predetermined frame based on the decimal value of the correction gradation corresponding to the correction data from the correction data sending unit 11 is set to −2 gradation, −1 floor One of tone, +1 gradation, and +2 gradation is added. In this way, an apparently less than integer gradation display is performed using an afterimage, which will be described in detail later.

In FIG. 2, the liquid crystal display unit 2 includes a liquid crystal panel 20, a plurality of source drivers 21, and a plurality of gate drivers 22. The liquid crystal panel 20 includes subpixels (see FIG. 5) arranged in a matrix corresponding to gate lines and source lines (not shown), and a liquid crystal layer of the subpixels applied to the gate lines is applied to the source lines. Transmittance based on voltage. The source driver 21 applies a voltage based on the gradation data of the RGB signal output from the signal processing circuit 3 to the source line of the liquid crystal panel 20. The gate driver 22 is a signal processing circuit 3
A scanning signal for selecting a gate line is output to the gate line of the liquid crystal panel 20 on the basis of the synchronization signal output from.

In FIG. 3, the source driver 21 includes a shift register 30, a latch 31, a sampling memory 32, a hold memory 33, a level shifter 34, a reference power supply circuit 35, a DA conversion circuit 36, and an output buffer 39. The latch 31 outputs RG output from the signal processing circuit 3.
The gradation signal of the B signal is latched in a time division manner. The sampling memory 32 stores the latched gradation data. The hold memory 33 takes in the gradation data of the sampling memory 32 in synchronization with the synchronization signal, outputs it to the level shifter 34, and holds the gradation data until the next synchronization signal is inputted. The level shifter 34 converts the gradation data from the hold memory 33 into D
Conversion is performed by boosting so as to conform to the A conversion circuit 36.

The DA conversion circuit 36 includes a positive DA conversion circuit 37 that outputs a positive voltage and a negative DA conversion circuit 38 that outputs a negative voltage. Selection of the positive electrode and the negative electrode is performed based on the synchronization signal output from the signal processing circuit 3. A voltage corresponding to the gradation data from the level shifter 34 is output. The output buffer 39 is composed of a differential amplifier circuit and supplies a stable voltage to the liquid crystal panel 20.
Output to.

In FIG. 4, the positive DA conversion circuit 37 is applied with a voltage V0 (voltage 1) from the reference power supply circuit 35.
63 resistors are provided in series between 2V) and V63 (voltage 6.2V). V0 is input 6
Bit 000000 (0 gradation) and V63 correspond to input 6 bits 111111 (63 gradation). The connection terminals V0 to V63 of the respective resistors have resistance values adapted to gamma 2.2. As shown in FIG. 4, the switches for selecting the terminals V0 to V63 are arranged in a six-stage tournament form, and the six bits of the input of each stage are ON / OFF controlled. Thus, V0 to V63 are output corresponding to the input 6 bits 000000 to 111111.

FIG. 5 is a diagram showing the pixels of the liquid crystal panel 20. The liquid crystal panel 20 is a color WVGA, and has 800 pixels in the gate line direction (horizontal direction) and 480 pixels in the source line direction (vertical direction). One pixel is composed of three sub-pixels of RGB. The liquid crystal panel 20 includes a light shielding barrier 20b in which the light shielding pattern of the sub-pixels is a checkered pattern (a black and white pattern of a chess board; see R / L in FIG. 6). As shown in FIG. 7, by this light shielding barrier 20b, one of the checkered patterns of the sub-pixels can be seen only from the right direction (the direction of the driver's seat in Japan), and the other is only viewed from the left direction (the direction of the passenger seat in Japan). It can be visually recognized. For example, as shown in FIG. 8, a navigation image can be viewed from the driver's seat, and a DVD image can be viewed from the passenger seat. The gradation data of the sub-pixels of the input image is 6 bits, and the luminance of RGB is 64 types from 0 gradation to 63 gradations. The drive control of the luminance of the liquid crystal panel 20 is in units of one gradation. That is, non-integer gradation data cannot be specified. The period of one screen (800 pixels × 480 pixels), that is, the frame period is 60 Hz.

FIG. 9 shows a correction data table stored in the EEPROM 4 or the LUT 13. The 64 types of gradation data are from 0 gradation to 63 gradation, and this correction data table has an even gradation (0 gradation, 2 gradations, respectively) for the correction target sub-pixel data and the right-side sub-pixel data. 4 gradations,
... in the (64/2) × (64/2) = 32 × 32 matrix corresponding to 62 gradations (one gradation), in addition to the dummy auxiliary gradation (64 gradations) correction data The matrix is 33 × 33. This is because the final gradation (63 gradations) is obtained by interpolation calculation described later.

In each matrix, correction data experimentally determined from the correction target sub-pixel data and the right-side sub-pixel data is 4 bits of an integer gradation (bit 3 is a sign bit, bit 2 to bit 0) Correction data of -7 to 0 to +7) using the 3 bits of correction data as correction data is stored. Since the FRC has a 4-frame cycle, the correction gradation corresponding to the correction data is 1/4 of the correction data. For example, 0.25 gradation is added when the correction data is +1, 0.5 gradation is added when the correction data is +2, and 1 gradation is added when the correction data is +4. The correction data is stored by converting experimentally obtained data into 0.25 gradation units.

In FIG. 9, a portion where the code “0” is entered is a portion where correction is not necessary since the sub-pixel data to be corrected and the pixel data on the right side have the same value and horizontal crosstalk does not occur. The grayscale data of the correction target sub-pixel is “0” as the minimum value or “63” as the maximum value, and horizontal crosstalk does not occur regardless of the pixel data on the right side. is there. In FIG. 9, all the correction data other than those that do not need to be corrected because the above-described horizontal crosstalk does not occur is omitted, but an integer value of −7 to 0 to +7 is entered.

The EEPROM 4 stores data excluding the portion of the code “0” entered in FIG. 9, that is, (33 × 33−33 × 3 + 2) × 4 bits = 992 × 4 bits. The data determined experimentally in advance and stored in the EEPROM 4 is read from the EEPROM 4 to the LUT 13 when the power switch is turned on, and the data shown in FIG.
Expanded as shown in

  The image processing of the display device having the above configuration will be described.

As shown in FIG. 5, the navigation image output from the navigation unit 41 of the navigation device 40 and the DVD image output from the DVD playback unit 32 are data of 800 pixels × 480 pixels, which is the same as the pixel array of the liquid crystal panel. Yes. The signal output from the screen switching circuit 43 of the navigation device 40 is a signal NN for visually recognizing a navigation image in both the left viewing direction (passenger seat direction of a Japanese car) and the right viewing direction (direction of a driver seat of a Japanese car). A signal DD for visually recognizing a DVD image in both the right and right viewing directions, a signal DN for visually recognizing a DVD image in the left viewing direction, a signal DN for visually recognizing a navigation image in the right viewing direction, and a navigation image viewed in the left viewing direction. There are four types of signals ND for visually recognizing a DVD image. The signal NN and the signal DD are one screen mode, and the signal DN and the signal ND are two screen modes.

The two-screen composition unit 6 outputs only the navigation image and the navigation image input from the navigation device 40 to the crosstalk correction unit 7 when the screen switching signal is NN which is the one-screen mode. The screen switching signal is another one-screen mode D
In the case of D, only the DVD image among the navigation image and the DVD image input from the navigation device 40 is output to the crosstalk correction unit 7.

When the screen switching signal is DN or ND in the two-screen mode, as shown in FIG. 6, the navigation image of 800 pixels × 480 pixels and the DVD playback unit 32 of 800 pixels × 480 pixels are selected in a checkered pattern, and 1 Two 800 pixel × 480 pixel images are synthesized.

The pre-processing unit 10 of the crosstalk correction unit 7 outputs the correction target sub-pixel data from the composite image input from the two-screen composition unit 6 to the correction data transmission unit 11 and the calculation unit 12, and the sub-pixel data on the right side is output. It outputs to the correction data sending part 11 and the calculating part 12.

The correction data sending unit 11 starts the supply of power to the display device 1 and at the same time the EEPROM.
4 is stored in the LUT 1 via the EEPROM controller 9.
3 is read.

The correction data sending unit 11 reads out corresponding correction data from the LUT 13 from the gradation data of the correction target sub-pixel and the gradation data of the sub-pixel adjacent to the right. At this time, since the rows and columns of the correction data table are every other gradation, the correction data is interpolated by the data interpolation unit 14 in the case of an interval (odd gradation).

The operation of the data interpolation unit 14 of the correction data sending unit 11 is as follows. That is,
For example, as shown in FIG. 10, the data of the four portions in the Z portion in FIG. 9 are LU, RU, LD, and RD.
If the correction data falls between LU and RU, the correction data is (LU + RU) / 2. If the correction data falls between LU and LD, the correction data is RU. (RU + RD) / 2 if the correction data falls between RD and RD, (LD + RD) / 2 if the correction data falls between LD and RD, and further, the correction data becomes LU and R
If it falls between D, it can be calculated as (LU + RU + LD + RD) / 4.
Although data for all gradations may be developed in the LUT 13, if such a configuration is adopted, the amount of data stored in the EEPROM 4 can be reduced.

In the above interpolation calculation, the addition of LU, RU, LD, and RD is divided by 2 or 4, so that the correction data obtained by interpolation is corrected to be an integer of -7 to +7 (gradation in units of 0.25). Is done.

The calculation unit 12 adds the gradation data of the right subpixel to the gradation data of the correction target subpixel. However, since the correction data corresponds to a gradation of 0.25 unit, the driving of the liquid crystal panel 20 must be an integer gradation, so that it becomes an apparent 0.25 unit in a cycle of 4 frames. FRC processing is performed. This FRC process will be described.

Since 1 (cycle) ÷ 0.25 (unit) = 4 (frame), the calculation unit 12 mixes correction of K (K is an integer) gradation and correction of K + 1 gradation with a period of 4 frames. In addition, if all the sub-pixels in one frame are set to K + 1 gradation, flicker is likely to occur.
As shown in FIG. 11, the arrangement of the sub-pixels for correcting the K + 1 gradation is divided into four types of 1, 2, 3, and 4, and is distributed evenly over the four frames.

For example, when the correction data calculated by interpolation is 1.25 gradation of 5, 2 gradation correction is performed for the arrangement 1 of the sub-pixels in the first frame (4n-3 frame) of the four frames, and the arrangement 2
, 3 and 4 are corrected by one gradation. Two gradation corrections are performed for the arrangement 2 of the sub-pixels in the second frame (4n-2 frame) of the four frames, and one gradation correction is performed for the arrangements 1, 3, and 4. Two gradation corrections are performed for the arrangement 3 of the sub-pixels in the third frame (4n-1 frame) of the four frames, and one gradation correction is performed for the arrangements 1, 2, and 4. Two gradation corrections are performed for the arrangement 4 of the sub-pixels in the fourth frame (4n frame) of the four frames, and one gradation correction is performed for the arrangements 1, 2, and 3.

Accordingly, the correction order of the four frames is 1, 0, 0, 0 for the arrangement 1, 0, 1, 0, 0, 0 for the arrangement 2, and 0, 0, 1, 0 for the arrangement 3. , Arrangement 4 is 0, 0, 0, 1 and the timing of K + 1 gradation correction is different.
Since the frame period is 60 Hz, an apparent retinal image effect can be obtained by mixing K gradation correction and K + 1 gradation correction in a frame of N (N is a positive integer of 2 or more) period in this way. Correction in units of less than one gradation can be performed. In the case of correction with 0.5 gradations below the decimal point, the timings of K + 1 gradation correction for arrangement 1 and arrangement 3 and arrangement 2 and arrangement 4 are the same.

In this way, the flicker is reduced by evenly distributing the sub-pixels in the arrangements 1, 2, 3, and 4 in one frame. However, the degree of flicker reduction is slightly different depending on the distribution patterns of arrangement 1 to arrangement 4. The present invention is objectively performed by quantifying the subtle flicker determination.

As flicker elements, there are non-uniform luminance within a frame and changes in luminance within the frame.
Flicker is less likely to occur when the brightness is evenly distributed without shifting. In addition, flicker is less likely to occur if the sub-pixels with high luminance do not appear to move as the frame changes, as in a moving image.

[Single screen mode (uniform brightness)]
The FRC processing method of the calculation unit 12 is different between the 1-screen mode and the 2-screen mode. First, the uniform brightness of the FRC process in the single screen mode will be described. If the screen switching signal from the navigation device 40 is NN or DD, the calculation unit 12 determines the one screen mode, and if the screen switching signal is DN or ND, determines the two screen mode.

As a block having the same minimum configuration of one frame (one screen), a block of 3 × 4 = 12 subpixels is set. “3” is the gate line direction (horizontal direction) and is the number of sub-pixels of one pixel. “4” is the source line direction (vertical direction), which is the number of frames in one cycle. Combining multiple blocks makes one frame. Thus, arrangements 1, 2, 3,
The distribution pattern of 4 is uniformly distributed by a plurality of block configurations, and the occurrence of flicker can be further reduced. As one group of luminance determination, a correction target sub-pixel,
A total of nine of the adjacent eight sub-pixels is set.

As shown in FIG. 11, there are four types of correction gradations corresponding to the correction data from the correction data sending unit 11 (0, 0.25, 0.5, and 0.75). for,
It is assumed that the value after the decimal point of the correction gradation corresponding to all the correction data is 0.25. Also,
Here, the integer value of the correction gradation corresponding to the correction data is exemplified as 20 (K = 20). That is, the average gradation (apparent gradation) in one cycle is 20.25, and as shown in FIG. 12, all sub-pixels in four frames in one cycle are 21 gradations in the first frame in arrangement 1. The other is 20 gradations, in arrangement 2 the 21st gradation is in the second frame and the others are the 20 gradations, in arrangement 3, the 21st gradation is in the third frame and the others are the 20th gradation, and in arrangement 4 is the 4th frame. There are 21 gradations and 20 others.

FIG. 13 shows a distribution pattern of the arrangements 1, 2, 3, and 4 determined to reduce flicker. Since one pixel is a square, the sub-pixel is a rectangle. The upward slanted line is R (red),
About 10% (thinner) shading indicates G (green), and about 20% (darker) shading indicates B (blue). Numbers 1 to 4 in FIG. 13 are arrangement numbers in FIG. For example, in FIG. 13, 2 sub-pixels have 21 gradations with high luminance in the second frame.

As shown in the gradation-luminance curve of FIG. 14, in the same gradation, the luminances of red and blue are substantially the same, and green has a luminance about four times that of red and blue. Since green is a color that people feel high in luminance, the green luminance is set higher than red and blue in the same gradation so that the liquid crystal panel feels bright. In the quantification of flicker determination, the luminance of red and blue is the same in the same gradation, and green is calculated as four times the luminance of red and blue.

Here, a luminance value is calculated at which the luminance of 20.25 gradation of one group is higher than that of 20 gradations. FIG. 13 is a diagram showing the luminance difference of the first frame. As for 20.25 gradation, there are 21 gradations in each of the first to fourth frames, one each for red, green, and blue. Red 2
If the difference in luminance between 1 gradation and 20 gradations is A, the luminance of 1 block of 20.25 gradations in any frame is 1A (red) + 4A (green) + 1A (blue) = 6A.

Therefore, when a value is obtained in which the average luminance of one subpixel in one group is the same as the average luminance of one subpixel in one block, the number of subpixels in one block is twelve and the number of subpixels in one group is nine. Therefore, 6A ÷ 12 × 9 = 4.5A. Since the calculation for flicker determination is an integer, a value that makes the average luminance of one sub-pixel in one group substantially the same as the average luminance of one sub-pixel in one block is 5A, and flicker is reduced when it becomes 5A. Point up as a pattern.

FIG. 15 is a diagram illustrating an example of a calculation method for flicker determination based on uniform luminance. 9 shown in FIG.
The nine sub-pixels are the nine sub-pixels above the block in FIG. 13 and represent one group. The sub pixel in the arrangement 1 in the center is a determination target. The first frame is 4A (G) + A (B) = 5A,
The second frame is A (R) + A (B) = 2A, and the third frame is A (R) + 4A (G) + A (
B) = 6A, and the fourth frame is A (R) + 4A (G) = 5A. Since there are two 5A,
There are 2 points for this sub-pixel.

In this way, 48 points of 4 frames × 12 subpixels are obtained for one block. It is determined that the larger the total point is, the less likely flicker occurs due to uneven brightness.

In this way, a group including a sub-pixel to be determined and its adjacent sub-pixels is set, and when the average luminance of one sub-pixel in one group is substantially the same as the average luminance of one sub-pixel in the block, the point of addition is set. Points are obtained for all sub-pixels in one block, and further points are obtained for 4 frames and summed. Thus, since it quantifies based on average brightness | luminance, the determination of the flicker by nonuniform brightness | luminance can be performed. In addition, since green with high luminance is taken into account, highly accurate quantification can be performed. A pattern with reduced flicker can be obtained by this determination of quantification.

[Single screen mode (change in brightness)]
Next, a change in luminance will be described. 16 and 17 are diagrams showing examples of bad patterns. In FIG. 16 and FIG. 17, the frame number of the same color is incremented by 1 downward, and appears to wave downward as the frame changes.

Therefore, the arrangement number of the same color sub-pixel of the upper adjacent pixel and the same color sub-pixel of the lower adjacent pixel (see the position in FIG. 17) can be a number that forms three sequential numbers with the arrangement number to be determined as the center position. If the number is not three consecutive numbers, the point is set to 1 because it is difficult to generate flicker. Here, the adjacent pixels refer to eight pixels that are slanted in the arrangement of FIG. The three serial numbers may be either ascending order or descending order, and 1 and 4 are serial numbers.

Twelve points (luminance change) of 12 subpixels are obtained for one block. It is determined that flicker due to a change in luminance is less likely to occur as the total point is larger.

Thus, the frame order in which the sub-pixels of the same color of the pixels above the determination target sub-pixel have 21 gradations, the frame order in which the determination target sub-pixel has 21 gradations, and the sub-pixels to be determined A point of addition is determined when the three sub-pixels of the same color of the pixel No. are not sequentially numbered in the frame order of 21 gradations. That is, since the frame change movement of the high-luminance sub-pixel of the same color is quantified, flicker due to a change in luminance can be determined. A pattern with reduced flicker can be obtained by this determination of quantification. The above moving directions are ascending order and descending order, that is, when the ascending order is 1, 2, 3 for the upper neighbor, the judgment target, and the lower neighbor. , Judgment target, lower adjacent arrangement number is 3, 2, 1
When descending, move from top to top with reference to the bottom neighbor. However, the moving direction may be another direction. In other words, the point of addition may be used when the arrangement is such that the frame does not move to two adjacent pixels (up, down, left, right, and diagonal eight pixels) due to a frame change.

Forty-eight luminance uniform points and twelve luminance change points are summed. It is determined that the total flicker is less likely to occur as the total point is larger.

FIG. 18 is a diagram illustrating an example of a pattern having a large total point. In this pattern, one green is arranged in the same group, and the adjacent arrangement number of the same color is the serial number 3
It arrange | positions so that it may not become connected. As shown in FIG. 18, the luminance uniformity point of each sub-pixel is 2, the luminance change is point 1, and the total of the points is 36.

[2 screen mode (uniform brightness)]
Next, the uniform luminance in the two-screen mode will be described. In the two-screen mode, since individual images are adjacent to each other, a large crosstalk occurs. Since the sub-pixels visible in the two-screen mode are arranged in a checkered pattern, the positions of adjacent or adjacent sub-pixels are different from those in the one-screen mode.

FIG. 19 is a diagram showing one block in the two-screen mode. One block has 3 × 4 = 12 sub-pixels viewed from one viewing direction. “3” is the gate line direction (
“4” is the source line direction (vertical direction).

FIG. 20 is a diagram showing one group in the two-screen mode. The number of sub-pixels that form one group is seven including the determination target and six neighboring sub-pixels on the left, right, upper left, upper right, lower left, and lower right. The subpixel is a rectangle (three subpixels of RGB are square). Therefore, the upper and lower adjacent subpixels are not included in the group because they are separated from the determination target subpixel.

As in the case of one screen, if the difference in luminance between 21 gradations and 20 gradations of red is A, the luminance of one block of 20.25 gradations is 1A (red) + 4A (green) + 1A (blue) ) = 6
A.

Since the number of sub-pixels in one block is 12 and the number of sub-pixels in one group is 7, 6A
÷ 12 × 7 = 3.5A. Since the calculation of flicker determination is an integer, a value that makes the average luminance of one subpixel in one group substantially the same as the average luminance of one subpixel in one block is 4A, and flicker is reduced when it becomes 4A. Point up as a pattern.

FIG. 20 is a diagram illustrating an example of a calculation method for flicker determination based on uniform luminance. 7 shown in FIG.
The 19 sub-pixels are the 7 sub-pixels in the upper left of the block of FIG. 19 and indicate one group.
The sub pixel in the arrangement 1 in the center is a determination target. The first frame is A (R) + 4A (G) = 5A
The second frame is A (R) + A (B) = 2A, the third frame is 4A (G) = 4A, and the fourth frame is 4A (G) + A (B) = 5A. Since there is one 4A, there is one point for this sub-pixel.

In this way, 48 points of 4 frames × 12 subpixels are obtained for one block. It is determined that the larger the total point is, the less likely flicker occurs due to uneven brightness.

[Dual screen mode (brightness change)]
Next, a change in luminance will be described. The arrangement number of the same color sub-pixel of the upper right adjacent pixel and the same color sub pixel of the lower right adjacent pixel (see the position in FIG. 19) is 3 with the arrangement number to be determined being the center position.
If it is a number that forms one serial number, the point is 0, and if it is not 3 serial numbers, the point is set to 1 because it is difficult for flicker to occur. Here, the adjacent pixels refer to eight pixels that are slanted in the arrangement of FIG. The three serial numbers may be either ascending order or descending order, and 1 and 4 are serial numbers.

Twelve points (luminance change) of 12 subpixels are obtained for one block. It is determined that flicker due to a change in luminance is less likely to occur as the total point is larger.

Forty-eight luminance uniform points and twelve luminance change points are summed. It is determined that the total flicker is less likely to occur as the total point is larger.

FIG. 21 is a diagram illustrating an example of a pattern having a large total point. In this pattern, one green is arranged in the same group, and the adjacent arrangement number of the same color is the serial number 3
It arrange | positions so that it may not become connected. As shown in FIG. 21, the luminance uniformity point of each sub-pixel is 1, the luminance change is point 1, and the total of the points is 24.

In the above-described embodiment, one frame out of four frames is one gradation higher than FRC with a decimal number of 0.25 gradation. However, this may be applied to a decimal number 0.5 gradation or a decimal number 0.75 gradation. it can. Decimal 0.
It is sufficient to change the arrangement number 4 to 2 and 3 to 1 considering that the 5 gradation is twice the decimal number 0.25 gradation,
The decimal number of 0.75 gradation may be applied as it is, considering that it is positive / negative of the decimal number of 0.25 gradation.

In this way, the calculation unit 13 performs the FRC process, and adds the correction gradation corresponding to the correction data sent from the correction data sending unit 12 to the correction target sub-pixel data sent from the preprocessing unit 11, The corrected sub-pixel data is sent to the output signal generator 7. The output signal generation unit 7 controls the polarity and timing so that the signal corrected by the crosstalk correction unit 6 can be displayed on the liquid crystal panel 2 and outputs the signal to the liquid crystal panel 2. Such correction of subpixel data is sequentially performed in the right direction by one subpixel for all subpixel data.

Although the above group is a product (3 × 4) of the number of RGB sub-pixels and the number of frame periods, it is not limited to this and can be arbitrarily set. The group is set by adding a proximity sub-pixel to the determination sub-pixel, and the number of the sub-pixels can be arbitrarily set.

Although the above-described light-shielding pattern is a checkered pattern, the present invention can be applied to various patterns such as a stripe pattern.

In addition, although the above-described embodiments are liquid crystal panels, they can be applied to electro-optical devices such as organic EL (electroluminescence) devices and plasma displays as long as they are compatible with the driving method of the present invention.

It is a block diagram which shows the principal part of the display apparatus of an Example. It is a block diagram which shows the principal part of the liquid crystal display part of FIG. FIG. 3 is a block diagram illustrating a main part of the source driver in FIG. 2. It is a block diagram which shows the principal part of the positive electrode DA conversion circuit of FIG. It is a figure which shows the pixel arrangement | sequence of a liquid crystal panel. It is a figure which shows the synthesis | combination of the separate image with respect to two viewing directions, and the sub pixel arrangement | sequence of a checkered pattern. It is a figure which shows the principle of 2 screens by the slit of a light shielding barrier. It is a figure which shows the display of the separate image with respect to two viewing directions. It is a figure which shows a correction data table. It is a figure which shows the method of carrying out the interpolation calculation of the gradation correction data which are not stored in the correction data table. It is a figure which shows each flame | frame gradation difference of FRC which makes 4 frames 1 period. It is a block diagram which shows the gradation assumed for quantization of flicker determination. It is a figure which shows the block in 1 screen mode. It is a gradation value-luminance diagram showing that the luminance of green is high. It is a figure which shows the group in 1 screen mode. It is a figure which shows the example which is easy to produce the flicker by the change of the brightness | luminance in 1 screen mode. It is a figure which shows FIG. 16 by a frame number. It is a figure which shows the calculation result of the point in 1 screen mode. It is a figure which shows the block in 2 screen mode. It is a figure which shows the group in 2 screen mode. It is a figure which shows the calculation result of the point in 2 screen mode.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Display apparatus 2 Liquid crystal display part 3 Signal processing circuit 4 EEPROM
Reference Signs List 5 Power supply circuit 6 Two-screen composition unit 7 Crosstalk correction unit 8 Output signal generation unit 9 EEPROM controller 10 Preprocessing unit 11 Correction data transmission unit 12 Calculation unit 13 LUT
14 Data Interpolation Unit 20 Liquid Crystal Panel 21 Source Driver 22 Gate Driver 36 DA Conversion Circuit 37 Positive DA Conversion Circuit 38 Negative DA Conversion Circuit 40 Navigation Device 41 Navigation Unit 42 DVD Playback Unit 43 Screen Switching Circuit

Claims (4)

One pixel is composed of three sub-pixels of red, green, and blue, the gradation display of the luminance of the sub-pixel is set, and the gradation of the correction target sub-pixel is corrected based on the gradation of the adjacent sub-pixel. A talk correction unit,
The crosstalk correcting unit performs the correction of K (K is an integer) gradation in N (N is a positive integer greater than or equal to 2) N frames (N1 is a positive integer less than N) frame, and the correction of K + 1 gradation is performed. N-
N1 frames are performed, and the subpixel is composed of a plurality of blocks in which one block is M1.
Arrangement numbers that define the order in which the correction of the K gradation and the correction of the K + 1 gradation in N frames of one cycle are set from 1 to N, and the arrangement number 1 assigned to one block of sub-pixels is set.
The luminance of the green sub-pixel in the same gradation is approximately L times the luminance of the red sub-pixel and the blue sub-pixel.
The luminance of the green subpixel is calculated as L times the luminance of the red subpixel and the blue subpixel,
A group in which the number of sub-pixels to be determined and the number of adjacent sub-pixels is a total of M2 is set in one block,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
When the correction of +1 gradation is performed for one frame, the average luminance of one sub-pixel in one group is substantially the same as the average luminance of one sub-pixel in one block for all sub-pixels in one block. The display device according to claim 1, wherein the display device is a pattern in which the arrangement number is assigned to one block at least once in a frame. One pixel is composed of three sub-pixels of red, green, and blue, the gradation display of the luminance of the sub-pixel is set, and the gradation of the correction target sub-pixel is corrected based on the gradation of the adjacent sub-pixel. A talk correction unit,
The crosstalk correcting unit performs the correction of K (K is an integer) gradation in N (N is a positive integer greater than or equal to 2) N frames (N1 is a positive integer less than N) frame, and the correction of K + 1 gradation is performed. N-
N1 frames are performed, and the subpixel is composed of a plurality of blocks in which one block is M1.
Arrangement numbers that define the order in which the correction of the K gradation and the correction of the K + 1 gradation in N frames of one cycle are set from 1 to N, and the arrangement number 1 assigned to one block of sub-pixels is set.
A display device in which the numbers from N to N are the same,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
When one frame of correction is performed for +1 gradation, the frame order in which all sub-pixels of one block have the same color sub-pixel of the first adjacent pixel of the determination target sub-pixel as K + 1 gradation, and the determination target The frame order in which the sub-pixel has K + 1 gradation and the frame order in which the sub-pixel of the same color of the second adjacent pixel of the determination target sub-pixel has K + 1 gradation are not sequential.
A display device characterized in that the arrangement number is a pattern assigned to one block. One pixel is composed of three sub-pixels of red, green, and blue, the gradation display of the luminance of the sub-pixel is set, and the gradation of the correction target sub-pixel is corrected based on the gradation of the adjacent sub-pixel. A talk correction unit,
The crosstalk correcting unit performs the correction of K (K is an integer) gradation in N (N is a positive integer greater than or equal to 2) N frames (N1 is a positive integer less than N) frame, and the correction of K + 1 gradation is performed. N-
N1 frames are performed, and the subpixel is composed of a plurality of blocks in which one block is M1.
Arrangement numbers that define the order in which the correction of the K gradation and the correction of the K + 1 gradation in N frames of one cycle are set from 1 to N, and the arrangement number 1 assigned to one block of sub-pixels is set.
To N, and the luminance of the green sub-pixel in the same gradation is approximately L times the luminance of the red sub-pixel and the blue sub-pixel. Because
The luminance of the green subpixel is calculated as L times the luminance of the red subpixel and the blue subpixel,
A group in which the number of sub-pixels to be determined and the number of adjacent sub-pixels is a total of M2 is set in one block,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
Point when the average luminance of one sub-pixel in one group is substantially the same as the average luminance of one sub-pixel in one block for all sub-pixels in one block when +1 gradation correction is performed for one frame The flicker determination method of the display device is characterized in that the pattern is determined to be flicker reduction as the point increases. One pixel is composed of three sub-pixels of red, green, and blue, the gradation display of the luminance of the sub-pixel is set, and the gradation of the correction target sub-pixel is corrected based on the gradation of the adjacent sub-pixel. A talk correction unit,
The crosstalk correcting unit performs the correction of K (K is an integer) gradation in N (N is a positive integer greater than or equal to 2) N frames (N1 is a positive integer less than N) frame, and the correction of K + 1 gradation is performed. N-
N1 frames are performed, and the subpixel is composed of a plurality of blocks in which one block is M1.
Arrangement numbers that define the order in which the correction of the K gradation and the correction of the K + 1 gradation in N frames of one cycle are set from 1 to N, and the arrangement number 1 assigned to one block of sub-pixels is set.
A display device in which the numbers from N to N are the same,
The correction of all the sub-pixels of the one block is performed by correcting the K gradation by N-1 frames, and the K
When one frame of correction is performed for +1 gradation, the frame order in which all sub-pixels of one block have the same color sub-pixel of the first adjacent pixel of the determination target sub-pixel as K + 1 gradation, and the determination target This is a point when the frame order in which the sub-pixel has K + 1 gradation and the frame order in which the sub-pixel of the same color of the second adjacent pixel of the determination target sub-pixel has K + 1 gradation are not sequential. The flicker determination method of the display device is characterized in that the pattern is determined to be flicker reduction as the point increases.

Images