JP2018112728A - Liquid crystal driving device, image display device, liquid crystal driving method and liquid crystal driving program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress image quality deterioration due to disclination by suppressing deterioration of brightness or gradation.SOLUTION: A liquid crystal driving device comprises: image data generation means 410 that generates a plurality of pieces of sub-frame image data for input frame image data; and driving means 420 that forms a gradation on a pixel by controlling first voltage application and second voltage application to each of a plurality pixels in each of a plurality of subfield periods included in one frame period sequentially on the basis of each gradation value of the plurality of pieces of sub-frame image data. When the gradation value of input frame image data is called as an input gradation value, the image data generation means generates first sub-frame image data having a first gradation value at least higher than the input gradation value as the plurality of pieces of sub-frame image data, and generates second sub-frame image data having a second gradation value lower than the input gradation value.SELECTED DRAWING: Figure 14

Description

  The present invention relates to a liquid crystal driving device that drives a liquid crystal element by a digital driving method.

  Examples of the liquid crystal element include a transmissive liquid crystal element such as a TN (Twisted Nematic) element and a reflective liquid crystal element such as a VAN (Vertical Alignment Nematic) element. The driving method of these liquid crystal elements includes an analog driving method in which brightness is controlled by changing a voltage applied to the liquid crystal layer according to the gradation, and voltage application by binarizing the voltage applied to the liquid crystal layer. There is a digital drive system that controls brightness by changing time. In this digital driving method, one frame period is divided into a plurality of subfield periods on the time axis, and application (on) and non-application (off) of a predetermined voltage to the pixel are controlled for each subfield. There is a sub-field driving method for displaying gradation.

  Here, a general subfield driving method will be described. FIG. 11 shows an example in which one frame period is divided into a plurality of subfield periods (bit length). A numerical value written on each subfield indicates a time weight within one frame period of the subfield. Here, as an example, a case where 64 gradations are expressed is shown. In the description here, the period of time weight 1 + 2 + 4 + 8 + 16 is referred to as an A subfield period, and the subfield period of time weight 32 is referred to as a B subfield period. Further, the subfield period in which the predetermined voltage is turned on is referred to as an on period, and the subfield period in which the predetermined voltage is turned off is referred to as an off period.

  FIG. 12 shows all gradation data corresponding to the subfield division example shown in FIG. The vertical axis represents gradation and the horizontal axis represents one frame period. In the figure, the white subfield period indicates an on period in which the pixels are in a white display state, and the black subfield period indicates an off period in which the pixels are in a black display state. According to this gradation data, two gradations adjacent to each other (hereinafter referred to as adjacent gradations), for example, 32 gradations and 33 gradations, are displayed on two adjacent pixels (hereinafter referred to as adjacent pixels) by the liquid crystal element. In this case, the A subfield period is an on period for 32 gradations and an off period for 33 gradations. The B subfield period is an off period for 32 gradations and an on period for 33 gradations.

  In this way, when an on-period and an off-period overlap in time in an adjacent pixel, that is, when a predetermined voltage is applied to one of the adjacent pixels and not applied to the other in the same period, so-called disclination occurs. Occurring and the brightness of the pixels on the on-period side decreases. The upper part of FIG. 15A shows an image of brightness reduction due to disclination when a horizontal gradation image is projected. When there is no disclination, smooth shading is expressed, but in adjacent gradations where the on period and the off period overlap in adjacent pixels, the brightness decreases due to the influence of disclination and dark lines appear.

  Patent Document 1 observes deterioration in image quality due to disclination by adding a common (uniform) correction amount to gradation data of all pixels for each frame and periodically changing the correction amount. A method for making it difficult for a person to visually recognize is disclosed.

JP2013-050681A

  However, the gradation disclosed in the method disclosed in Patent Document 1 may be impaired. For example, there is a case where a correction amount is set such that the correction amount subtracted in the next frame is the same as the correction amount added to the gradation data of all pixels in the first frame. In this case, since the pixel displaying halftone is subtracted in the next frame by the same amount as the correction amount added in the first frame, the original brightness can be expressed by the average of the two frames. Is kept. However, in a pixel that displays a gradation lower than the correction amount, if the subtraction is performed in the next frame by the same amount as the correction amount added in the first frame, the subtraction is less than the minimum gradation, so that the subtraction cannot be performed. On average, brightness increases from the original gradation. On the other hand, in the pixel displaying high gradation, if the correction amount is added in the first frame, it exceeds the maximum gradation, so the addition cannot be performed, and the brightness is lower than the original gradation due to the average of two frames. To do.

  The present invention provides a liquid crystal drive device and the like that can prevent deterioration in image quality due to disclination from being visually recognized while suppressing a decrease in brightness and gradation.

  A liquid crystal driving device according to one aspect of the present invention drives a liquid crystal element having a plurality of pixels. The liquid crystal driving device sequentially outputs one frame period based on the image data generating means for generating a plurality of subframe image data and the respective gradation values of the plurality of subframe image data for the input frame image data. In each of a plurality of subfield periods included in the pixel, a gradation is formed in the pixel by controlling the application of the first voltage and the application of the second voltage lower than the first voltage to each of the plurality of pixels. Drive means. When the gradation value of the input frame image data is referred to as an input gradation value, the image data generation means includes a first subframe having at least a first gradation value higher than the input gradation value as a plurality of subframe image data. The image data is generated, and second subframe image data having a second gradation value lower than the input gradation value is generated.

  Note that an image display device having the liquid crystal driving device and the liquid crystal element also constitutes another aspect of the present invention.

  According to another aspect of the present invention, there is provided a liquid crystal driving method based on the step of generating a plurality of subframe image data for input frame image data and the respective gradation values of the plurality of subframe image data. Sequentially controlling the application of the first voltage and the application of the second voltage lower than the first voltage to each of the plurality of pixels in each of the plurality of subfield periods included in one frame period. And forming a gradation on the pixel. When the gradation value of the input frame image data is referred to as the input gradation value, first subframe image data having at least a first gradation value higher than the input gradation value is generated as a plurality of subframe image data. The second sub-frame image data having a second gradation value lower than the input gradation value is generated.

  A liquid crystal drive program that is a computer program that causes a computer to execute processing according to the liquid crystal drive method also constitutes another aspect of the present invention.

  According to the present invention, it is possible to make it difficult to visually recognize deterioration in image quality due to disclination while suppressing a decrease in brightness and gradation.

1 is a diagram illustrating an optical configuration of a liquid crystal projector that is Embodiment 1 of the present invention. FIG. FIG. 3 is a cross-sectional view of a liquid crystal element used in the projector according to the first embodiment. FIG. 6 is a diagram illustrating a plurality of subfield periods within one frame period in the first embodiment. FIG. 6 is a diagram illustrating gradation data in an A subfield period according to the first embodiment. FIG. 4 is a diagram illustrating all gradation data in the first embodiment. FIG. 3 is a diagram showing pixel lines in Embodiment 1. FIG. 6 is a diagram illustrating response characteristics of liquid crystal when switching from all white display to black and white display in the first embodiment. FIG. 6 is a diagram illustrating a response characteristic of brightness when switching from all white display to black and white display in the first embodiment. FIG. 6 is a diagram illustrating response characteristics of liquid crystal when switching from full black display to black and white display in the first embodiment. FIG. 6 is a diagram illustrating a response characteristic of brightness when switching from full black to black and white display in the first embodiment. The figure which shows the several subfield period in 1 frame period in the past. The figure which shows the conventional all gradation data. The figure which shows all the gradation data of patent document 1. FIG. FIG. 2 is a block diagram illustrating a configuration of a control circuit according to the first embodiment. The figure which shows the 1st and 2nd gain in Example 1, and the appearance of a disclination dark line. FIG. 3 is a diagram showing a comparative example with respect to Example 1. FIG. 6 is a diagram illustrating a modification example of the first embodiment. FIG. 9 is a block diagram illustrating a configuration of a control circuit according to a third embodiment. The figure which shows the 1st and 2nd gain in Example 2 and a comparative example. FIG. 10 is a diagram illustrating gains applied to 1st, 2nd, 3rd, and 4th subframes according to the third embodiment. The figure explaining the problem which Example 2 solves. The figure which shows how the 1st and 2nd gain and disclination dark line appear in Example 2. FIG. The figure which shows the 1st and 2nd gain in Example 4, and the appearance of a disclination dark line. FIG. 10 is a diagram illustrating effects unique to Example 4;

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 shows an optical configuration of a liquid crystal projector as an image display apparatus that is Embodiment 1 of the present invention. In this embodiment, a projector will be described as an example of an image display device using a liquid crystal element. However, the image display device includes an image display device using a liquid crystal element other than the projector, such as a direct-view monitor.

  The control circuit 303 corresponds to a liquid crystal driving device. The control circuit 303 includes a video input unit (image acquisition unit) 303a that acquires an input video signal (input image) from an external device (not shown), and a floor that will be described later according to the gradation (input gradation) of the input video signal. A driving circuit unit 303b that generates a pixel driving signal corresponding to the tone data. The pixel drive signal is generated for each color of red, green, and blue, and the pixel drive signal for each color is input to the red liquid crystal element 3R, the green liquid crystal element 3G, and the blue liquid crystal element 3B. As a result, the red liquid crystal element 3R, the green liquid crystal element 3G, and the blue liquid crystal element 3B are driven independently of each other. The red liquid crystal element 3R, the green liquid crystal element 3G, and the blue liquid crystal element 3B are vertical alignment mode reflective liquid crystal elements.

  The illumination optical system 301 causes white light from a light source (discharge lamp or the like) to enter the dichroic mirror 305 with the polarization direction aligned. The dichroic mirror 305 reflects magenta light and transmits green light. The magenta light reflected by the dichroic mirror 305 is incident on the blue cross color polarizer 311, where half-wave retardation is given only to the blue light, thereby generating blue light and red light whose polarization directions are orthogonal to each other. . Blue light and red light enter the polarization beam splitter 310, and the blue light passes through the polarization separation film of the polarization beam splitter 310 and is guided to the blue liquid crystal element 3B. The red color component is reflected by the polarization separation film and guided to the red liquid crystal element 3R.

  On the other hand, the green light transmitted through the dichroic mirror 305 passes through the dummy glass 306 for correcting the optical path length, enters the polarization beam splitter 307, is reflected by the polarization separation film, and is guided to the green liquid crystal element 3G. .

  Each liquid crystal element (3R, 3G, 3B) modulates and reflects incident light according to the modulation state of each pixel. The red light modulated by the red liquid crystal element 3R passes through the polarization separation film of the polarization beam splitter 310 and enters the red cross color polarizer 312 where half-wave retardation is given. Then, the red light enters the polarization beam splitter 308, is reflected by the polarization separation film, and travels toward the projection optical system 304.

  Further, the blue light modulated by the blue liquid crystal element 3B is reflected by the polarization separation film of the polarization beam splitter 310, passes through the red cross color polarizer 312 as it is, and enters the polarization beam splitter 308 to enter the polarization separation film. Is reflected toward the projection optical system 304. The green light modulated by the green liquid crystal element 3G passes through the polarization separation film of the polarization beam splitter 307, passes through the dummy glass 309 for correcting the optical path length, enters the polarization beam splitter 308, and the polarization separation thereof. The light passes through the film toward a projection optical system 304 having a projection lens. Thus, red light, green light, and blue light that have been color-combined enter the projection optical system 304. Then, the color light after color synthesis is enlarged and projected onto a projection surface 313 such as a screen by the projection optical system 304.

  In this embodiment, the case of using a reflective liquid crystal element is described, but a transmissive liquid crystal element may be used.

  FIG. 2 shows a cross-sectional structure of the reflective liquid crystal elements (3R, 3G, 3B). Reference numeral 101 denotes an AR coating film, 102 denotes a glass substrate, 103 denotes a common electrode, 104 denotes an alignment film, 105 denotes a liquid crystal layer, 106 denotes an alignment film, 107 denotes a pixel electrode, and 108 denotes a Si substrate.

  The control circuit 303 shown in FIG. 1 drives each pixel by the subfield driving method described above. That is, one frame period is divided into a plurality of subfield periods on the time axis, and on (applied) and off (non-applied) control of a predetermined voltage for each pixel is controlled for each subfield period in accordance with gradation data. A gradation is formed (displayed) on the pixel. One frame period is a period during which one frame image is displayed on the liquid crystal element. In this embodiment, the liquid crystal element is driven at 120 Hz, and one frame period is set to 8.33 ms.

  Note that although no voltage may be applied during the off period, the off period is a period for expressing the pixels in black, and the voltage does not necessarily have to be zero for design reasons. In other words, turning on and off the predetermined voltage can be paraphrased as applying a first voltage (predetermined voltage) and applying a second voltage lower than the first voltage.

  Hereinafter, setting of the subfield period and gradation data in the control circuit 303 will be described. The control circuit 303 may be configured by a computer, and the setting of the following subfield period and on / off of a predetermined voltage for each subfield period may be controlled according to a liquid crystal driving program as a computer program.

  FIG. 3 shows division of one frame period into a plurality of subfield periods (bit length) in the present embodiment. A numerical value written on each subfield indicates a time weight within one frame period of the subfield. In this embodiment, 96 gradations are expressed. In the description here, the period having the time weight of 1 + 2 + 4 + 8 is referred to as an A subfield period (first period), and a bit indicating a gradation expressed in binary in the A subfield period is referred to as a lower bit. In addition, 10 subfield periods having a time weight of 8 are collectively referred to as a B subfield period (second period), and a bit representing a gray scale expressed in the B subfield period is referred to as an upper bit. The time weight 1 corresponds to 0.087 ms, and the time weight 8 corresponds to 0.69 ms.

  Further, the subfield period in which the predetermined voltage is turned on (the first voltage is applied) is referred to as an on period, and the subfield period in which the predetermined voltage is turned off (the second voltage is applied) is referred to as an off period.

  FIG. 4 shows gradation data in the A subfield period shown in FIG. The vertical axis represents gradation and the horizontal axis represents one frame period. In the A subfield period, 16 gradations are expressed. In the figure, the white subfield period indicates an ON period in which the above-described predetermined voltage is applied so that the pixel is in a white display state, and the black subfield period indicates that the predetermined voltage is OFF so that the pixel is in a black display state. Indicates the off period.

  FIG. 5 shows gradation data in the A and B subfield periods (lower and upper bits) in the present embodiment. This gradation data is gradation data for expressing 96 gradations as all gradations. In this gradation data, an A subfield period (lower bit) is arranged at the time center of one frame period, and a B subfield period (upper bit) is divided into 1SF to 5SF and 6SF to 10SF before and after that. Has been placed. That is, the B subfield period is divided into two, and each B subfield period includes two or more subfield periods.

  According to this gradation data, when displaying adjacent gradations that are two gradations adjacent to each other, for example, 48 gradations and 49 gradations, to adjacent pixels that are two adjacent pixels in the liquid crystal element, the A sub The field period is an on period for 48 gradations and an off period for 49 gradations. In the 48 gradations, 1SF, 4SF, 5SF, 6SF, 7SF, and 10SF in the B subfield period are off periods, and 2SF, 3SF, 8SF, and 9SF are on periods. On the other hand, for 49 gradations, 1SF, 5SF, 6SF, and 10SF in the B subfield period are off periods, and 2SF, 3SF, 4SF, 7SF, 8SF, and 9SF are on periods. When such an adjacent gradation is displayed on an adjacent pixel, an ON / OFF adjacent period in which the ON period and the OFF period overlap is generated in the adjacent pixel. Specifically, when 48 gradations and 49 gradations are displayed on adjacent pixels, 4SF and 7SF in the B subfield period are on / off adjacent periods.

  Here, the gradation data of the present embodiment is compared with the conventional gradation data shown in FIG. In the gray scale data in FIG. 13, the B subfield period continues as a unit after the A subframe period, but in the gray scale data of this embodiment shown in FIG. 5, the B subfield period is divided before and after the A subframe period. Are arranged. For example, paying attention to 48 gradations and 49 gradations, in FIG. 13, 5SF and 6SF in the B subfield period are on / off adjacent periods, and 16 on / off adjacent periods continue as time weights. . This is the same for the other adjacent gradations, ie, 16 gradation and 17 gradation, 32 gradation and 33 gradation, 64 gradation and 65 gradation, 80 gradation and 81 gradation, and the like. On the other hand, in this embodiment shown in FIG. 5, in any of the above adjacent gradations, the ON / OFF adjacent period continues in the B subfield period as one time subfield of 8 as a time weight. The period is (= 0.69 ms). A plurality (two) of ON / OFF adjacent periods, which are one subfield period, are separated from each other across the A subframe period.

  Next, the effect obtained by distributing the ON / OFF adjacent periods in a distributed manner as in this embodiment will be described.

  First, as shown in FIG. 6, when the pixels arranged in a matrix form are switched from the all-white display state to the monochrome display state in which white and black are alternately displayed for each pixel line, and from the all-black display state to the monochrome display state. The response characteristics of the liquid crystal when switching to the display state will be described. The 4 × 4 pixels shown in FIG. 6 are arranged in a matrix with a pixel pitch of 8 μm. In the all white display state, both the pixels of the A pixel line and the pixels of the B pixel line in FIG. 6 display white. In the monochrome display state, the pixels of the A pixel line are switched from the white display state to the black display state, and the pixels of the B pixel line are maintained in the white display state.

  FIG. 7 shows the response characteristics of the liquid crystal. The horizontal axis indicates the position of the pixel, and the vertical axis indicates the brightness at each pixel (however, the ratio when white is 1). 0 to 8 μm on the horizontal axis represents pixels of the A pixel line shown in FIG. 6, and 8 to 16 μm represents pixels of the B pixel line. The plurality of curves indicate the brightness for each elapsed time (0.3 ms, 0.6 ms, 1.0 ms, 1.3 ms) when the switching point from the all white display state to the black and white display state is set to 0 ms.

  As described above, the pixels of the A pixel line are switched from the white display state to the black display state. However, the pixels of the A pixel line are relatively uniformly bright without being affected by disclination because of the pretilt angle direction in the liquid crystal. Changes (darkens). On the other hand, in the pixels of the B pixel line, disclination does not occur in the all white display state. However, the brightness curve gradually becomes distorted as time passes under the influence of disclination after the black-and-white display state is reached, and darkens in the vicinity of 12 μm to 16 μm (dark lines appear).

  In general, the gamma curve (gamma characteristic) that determines the drive gradation of a liquid crystal element relative to the input gradation shows the response characteristics when changing the gradation while displaying the same gradation on the entire surface of the liquid crystal element where no disclination occurs. Created as a premise. For this reason, when a liquid crystal element is driven using such a gamma curve, disclination occurs in a monochrome display state, and it is possible to obtain a brightness lower than the original brightness corresponding to the gamma curve.

  FIG. 8 shows changes in brightness depending on the presence or absence of disclination when the liquid crystal element is switched from the all white display state to the black and white display state. The horizontal axis represents the elapsed time from the switching point, and the vertical axis represents the change in the integrated value of the total brightness (hereinafter simply referred to as brightness) of the pixels of the A and B pixel lines. The brightness is shown as a ratio when the all white display state is 1. When disclination occurs (“with disclination”), the brightness of the pixel of the A pixel line changes with a characteristic close to the response characteristic shown in the vicinity of 1 to 6 μm in FIG. The brightness of the pixels is in a state where white is displayed with 100% brightness. As the time elapses thereafter, the amount of decrease in brightness when disclination occurs is larger than the amount of decrease in brightness when no disclination occurs (“no disclination”). It will become.

  On the other hand, when switching from the all black display state to the black and white display, both the pixel of the A pixel line and the pixel of the B pixel line shown in FIG. 6 are changed from the black display state to the B pixel while the pixels of the A pixel line are kept in the black display state. The line pixels are set to the white display state. FIG. 9 shows the response characteristics of the liquid crystal at this time. The horizontal axis indicates the position of the pixel, and the vertical axis indicates the brightness at each pixel (however, the ratio when white is 1). 0 to 8 μm on the horizontal axis represents pixels of the A pixel line shown in FIG. 6, and 8 to 16 μm represents pixels of the B pixel line. The plurality of curves indicate the brightness for each elapsed time (0.3 ms, 0.6 ms, 1.0 ms, 1.3 ms) when the switching time from the all black display state to the black and white display state is set to 0 ms.

  As described above, the pixels of the B pixel line are switched from the black display state to the white display state, but the pixels of the B pixel line are gradually affected with the disclination after the white display state and gradually. The brightness curve becomes distorted. And it becomes dark especially in the vicinity of 12 μm to 16 μm (dark lines appear). Also, the irregular shape of the brightness curve becomes more prominent with the passage of time.

  As described above, the gamma curve (gamma characteristic) that determines the drive gradation of the liquid crystal element relative to the input gradation generally changes the gradation while displaying the same gradation on the entire surface of the liquid crystal element where no disclination occurs. It is created on the assumption of the response characteristics when For this reason, when a liquid crystal element is driven using such a gamma curve, disclination occurs in a monochrome display state, and it is possible to obtain a brightness lower than the original brightness corresponding to the gamma curve.

  FIG. 10 shows changes in brightness depending on the presence or absence of disclination when the liquid crystal element is switched from the all black display state to the black and white display state. The abscissa represents the elapsed time from the switching point, and the ordinate represents the integrated value of the total brightness of the pixels of the A and B pixel lines (hereinafter simply referred to as brightness and the ratio when the all white display state is 1). ). As for the brightness when disclination does not occur (“no disclination”), the pixels in the A pixel line are always in the black display state, and the pixels in the B pixel line are switched from the black display state to the white display state. It shows the change in brightness as you go. On the other hand, when disclination occurs (“with disclination”), it indicates a change in the integrated value of the sum of the brightness of the pixels of the A pixel line and the B pixel line shown in FIG. .

In FIG. 10, when disclination occurs, the amount of increase in brightness over time is smaller than when disclination does not occur. That is, the longer the disclination occurs after switching from the all black display state to the black and white display state, the darker the disclination does not occur.
Next, a case where 48 gradations are displayed on the pixels of the A pixel line and 49 gradations are displayed on the pixels of the B pixel line using the conventional gradation data shown in FIG. 13 will be described. The period in which the disclination occurs when using this gradation data is in the B subfield period in which the A pixel line pixel is in the black display state and the B pixel line pixel is in the white display state. 5SF and 6SF. 4SF before 5SF is a period in which both the pixels of the A pixel line and the B pixel line pixels are in the white display state, and no disclination occurs.

  The response characteristics of the liquid crystal from 5SF to 6SF are characteristics corresponding to “with disclination” in FIG. Since 4SF is in an all-white display state, 100% brightness is output, and disclination occurs during 1.39 ms from the start of 5SF to the end of 6SF. 8 corresponds to 0 ms, and the end of 6SF corresponds to 1.39 ms. At this time, the brightness decreases to 0.27 from 0.5 when no disclination occurs. As described above, on the basis of the gamma characteristic created on the premise of the same gradation on the entire surface, the ratio becomes as dark as 54% (= 0.27 / 0.5) from 5SF to 6SF at which disclination occurs.

  On the other hand, in this embodiment, 48 gradations are displayed on the pixel (second pixel) of the A pixel line by the gradation data shown in FIG. 5, and 49 gradations are displayed on the pixel (first pixel) of the B pixel line. A case of displaying is described. The period in which the disclination occurs when using this gradation data is 4SF and 7SF in the B subfield period in which the pixels in the A pixel line and the pixels in the B pixel line are in the disclination occurrence display state. 3SF before 4SF is a period in which both the pixels of the A pixel line and the pixels of the B pixel line are in the white display state and no disclination occurs.

  The response characteristic of the liquid crystal at 4SF is a characteristic corresponding to “with disclination” in FIG. Since 3SF is in an all-white display state, 100% brightness is output, and disclination occurs during 0.69 ms of 4SF. Therefore, the start time of 4SF corresponds to 0 ms in FIG. The end time corresponds to 0.69 ms. At this time, the brightness decreases only to 0.65 compared to 0.7 when no disclination occurs.

  Further, the response characteristic of the liquid crystal in 7SF, which is a subfield period in which another disclination occurs, is a characteristic corresponding to “with disclination” in FIG. In 6SF, since the display is all black, the brightness is 0%, and disclination occurs during 0.69 ms of 7SF. Therefore, the start time of 7SF corresponds to 0 ms in FIG. 10 and the end time of 7SF is reached. Corresponds to 0.69 ms. At this time, the brightness decreases only to 0.18, compared to 0.25 when no disclination occurs.

  The sum of brightness when disclination does not occur between 4SF and 7SF is 0.95 (= 0.70 + 0.25), whereas the sum of brightness when disclination occurs is 0.83 (= 0.65 + 0.18). When the gamma characteristic created on the premise of the same gradation on the entire surface is used as a reference, the display becomes dark only up to 87% (= 0.83 / 0.95) in the disclination occurrence display state. That is, according to the present embodiment, it is possible to suppress a decrease in brightness.

  Next, a case where other adjacent gradations are displayed will be described. First, a case where 16 gradations are displayed on the pixels of the A pixel line shown in FIG. 6 and 17 gradations are displayed on the pixels of the B pixel line based on the conventional gradation data shown in FIG. 13 will be described. The period in which the disclination occurs when using this gradation data is in the B subfield period in which the A pixel line pixel is in the black display state and the B pixel line pixel is in the white display state. 1SF and 2SF.

  The response characteristics of the liquid crystal from 1SF to 2SF are characteristics corresponding to “with disclination” in FIG. Disclination occurs during 1.39 ms from the start of 1SF to the end of 2SF. Therefore, the start time of 1SF corresponds to 0 ms in FIG. 10, and the end time of 2SF corresponds to 1.39 ms. At this time, the brightness decreases to 0.27 from 0.5 when no disclination occurs. As described in the first embodiment, on the basis of the gamma characteristic created on the premise of the same gradation on the entire surface, the ratio is 54% (= 0.27 / 0.5) from 1SF to 2SF where disclination occurs. It becomes darker.

  On the other hand, in the present embodiment, 16 gradations are displayed on the pixel (second pixel) of the A pixel line by the gradation data shown in FIG. 5, and 17 gradations are displayed on the pixel (first pixel) of the B pixel line. A case of displaying is described. The period in which the disclination occurs when using this gradation data is 3SF and 8SF in the B subfield period in which the pixels of the A pixel line and the pixels of the B pixel line are in the disclination occurrence display state. In 2SF before 3SF, both the pixels of the A pixel line and the pixels of the B pixel line are in a black display state, and disclination does not occur. The response characteristic of the liquid crystal at 3SF is a characteristic corresponding to “with disclination” in FIG. Since 2SF is in an all black display state, the brightness is 0%, and disclination occurs during 0.69 ms of 3SF. Therefore, the start time of 3SF corresponds to 0 ms in FIG. 10 and the end of 3SF. Corresponds to 0.69 ms. At this time, the brightness decreases only to 0.18, compared to 0.25 when no disclination occurs.

  Further, the response characteristic of the liquid crystal at 8SF, which is a subfield period in which another disclination occurs, is also a characteristic corresponding to “with disclination” in FIG. In 7SF, since the display is all black, the brightness is 0%, and disclination occurs during 0.69 ms of 8SF. Therefore, the start time of 8SF corresponds to 0 ms in FIG. 10 and the end time of 8SF is reached. Corresponds to 0.69 ms. At this time, the brightness decreases only to 0.18, compared to 0.25 when no disclination occurs.

  The sum of brightness when disclination does not occur between 3SF and 8SF is 0.50 (= 0.25 + 0.25), whereas the sum of brightness when disclination occurs is 0.36 (= 0.18 + 0.18). When the gamma characteristic created on the premise of the same gradation on the entire surface is used as a reference, in the disclination occurrence display state, the ratio becomes dark only up to 72% (= 0.36 / 0.50). That is, according to the present embodiment, it is possible to suppress a decrease in brightness.

  As described above, in this embodiment, a plurality of ON / OFF adjacent periods that are in a disclination display state when displaying adjacent gradations are provided apart (distributed) within one frame period. The continuous on / off adjacent period is shortened. That is, the display state of the disclination occurrence in the adjacent pixels is shifted to another display state before the brightness decrease due to the disclination becomes large. As a result, it is possible to suppress a decrease in brightness due to disclination so that the dark line is not noticeable, and an image with good image quality can be displayed.

  The occurrence of disclination can be suppressed by the liquid crystal element driving method described above (hereinafter referred to as the first driving method). However, in order to make dark lines due to disclination inconspicuous, the following driving method (hereinafter referred to as the second driving method) is also used in this embodiment.

  FIG. 14 shows an internal configuration of the control circuit 303. The input unit 303a takes in an input video signal via a receiver IC (not shown) such as DVI or HDMI (registered trademark). The input unit 303a uses the scaling function to down-convert or up-convert the input video signal and output input image data in a predetermined image format. The input image data is composed of a plurality of continuous input frame image data, and includes a vertical synchronization signal and a horizontal synchronization signal.

  The input unit 303a and the drive circuit unit 303b are connected to the CPU 200 via the register bus 199.

  The drive circuit unit 303b sequentially receives the input frame image data from the input unit 303a, and the plurality of pixels (hereinafter referred to as liquid crystal pixels) of the liquid crystal element 3 (three liquid crystal elements 3R, 3G, and 3B shown in FIG. 1). Drive each one. That is, a pixel drive signal for displaying a gradation on each liquid crystal pixel is generated. The drive circuit unit 303b includes an image generation unit 410 and a panel drive unit 420. The image generation unit 410 corresponds to an image data generation unit, and the panel drive unit 420 corresponds to a drive unit.

  The image generation unit 410 includes a double speed conversion unit 411, an image memory 412, a luminance linear conversion unit 413, a first gain application unit 414, a second gain application unit 415, and an output switching unit 419.

  The double speed conversion unit 411 writes each input frame image data in the image memory 412 and generates a plurality of subframe image data for the input frame image data. Specifically, in the case of conversion to double speed, the double speed conversion unit 411 uses a data width that is twice that at the time of writing to the image memory 412 or twice the speed at the time of writing to the image memory 412. The sub-frame image data is generated so as to be read at a speed of. For example, when the frequency of the vertical synchronization signal of the input frame image data is 60 Hz, the double speed conversion unit 411 generates two subframe image data with a period corresponding to 120 Hz. The plurality of sub-frame image data generated by the double speed conversion unit 411 are all the same image data as the input frame image data. That is, the input frame image data and the plurality of subframe image data generated by the double speed conversion unit 411 have the same gradation value at the same pixel position between them.

  In the present embodiment, the double speed conversion unit 411 generates two subframe image data, of which subframe image data generated earlier is referred to as 1st subframe image data, and the subframe image data generated later is referred to as subframe image data. This is called 2nd subframe image data. The double speed conversion unit 411 generates a field signal for identifying whether the generated subframe image data is 1st subframe image data or 2nd subframe image data, and outputs the field signal to the output switching unit 419.

  The luminance linear conversion unit 413 is configured so that the drive gradation of the liquid crystal pixel increases linearly with an increase in the gradation value (that is, the input gradation value) of each subframe image data input from the double speed conversion unit 411. Tone value conversion is performed on the input tone value. In the following description, the relationship in which the drive gradation of the liquid crystal pixel increases linearly (in proportion) with the increase of the input gradation value is referred to as a luminance linear relationship. Then, the luminance linear conversion unit 413 outputs the input gradation value after the gradation value conversion of the 1st subframe image data to the first gain application unit 414, and the input floor after the gradation value conversion of the 2nd subframe image data. The adjustment value is output to the second gain application unit 415.

  The first gain application unit 414 applies (multiplies) a first gain, which will be described later, to the input gradation value after gradation value conversion of the first subframe image data, thereby obtaining the first gradation value as the first gradation value. The 1st output gradation value is calculated. Then, the first gain applying unit 414 outputs the 1st subframe image data having the 1st output gradation value to the output switching unit 419. Further, the second gain application unit 415 applies 2nd gain described later to the input tone value after the tone value conversion of the 2nd subframe image data, thereby 2nd as the second tone value. The output tone value is calculated. Then, the second gain application unit 415 outputs 2nd subframe image data having a 2nd output gradation value to the output switching unit 419.

  The first and second gain application units 414 and 415 in the present embodiment set the input gradation value area equal to or lower than the predetermined gradation value as the low gradation area, and set the input gradation value larger than the predetermined gradation value. A region (that is, a predetermined region) is set as a high gradation region. Then, the first and second gain application units 414 and 415 set different gains for the low gradation region and the high gradation region in each of the first and second gains. The above calculations performed by the first and second gain applying units 414 and 415 are collectively referred to as gain calculations in the following description and will be described in detail later.

  The output switching unit 419 is synchronized with the vertical synchronization signal of each subframe image data, and 1st and 2nd input from the first and second gain application units 414 and 415 according to the field signal from the double speed conversion unit 411. Subframe image data is alternately switched and output. As a result, the first subframe image data after the gain calculation and the 2nd subframe image data after the gain calculation are alternately input to the panel drive unit 420.

  Panel drive unit 420 includes a VT gamma conversion unit 421 and a PWM conversion unit 422. The VT gamma conversion unit 421 outputs the 1st and 2nd output gradation values of the 1st and 2nd subframe image data so that necessary optical characteristics can be obtained according to the gradation characteristics that change depending on the response characteristics of the liquid crystal in the liquid crystal element 3. Gamma correction is performed for.

  The PWM conversion unit 422 drives the liquid crystal element 3 by the above-described subfield driving method based on the 1st and 2nd subframe image data (1st and 2nd output gradation values) after the gamma correction from the VT gamma conversion unit 421. Output a PWM signal.

  Next, the gain calculation performed by the first and second gain applying units 414 and 415 will be specifically described with reference to FIGS. Here, the case where each of the reflective liquid crystal elements 3G, 3R, and 3B has 96 pixels in the horizontal direction will be described. FIGS. 15A to 15C show the first and second gains applied to the input gradation values by the first and second gain application units 414 and 415, and the gradation image projected on the projection surface ( The relationship between the image quality degradation due to disclination in the projected image) is shown. Degradation of image quality due to disclination is shown as a dark line (black line).

  FIG. 15A shows the relationship between the input gradation value and the output gradation value when gain calculation is not performed, that is, when 1.0 times (100%) as the reference gain is applied to the input gradation value. Show. At this time, there are five dark lines (simply referred to as disclination in the figure, hereinafter referred to as disclination dark lines) generated by disclination in the projected image.

  As shown in FIG. 15B, the first gain application unit 414 sets the first gain higher than the target gain as the reference gain shown in FIG. 15A to the input gradation value of the 1st subframe image data. Apply. The first gain is a gain that is higher than the target gain and has a different slope depending on the area of the input gradation value. Specifically, the first gain is 1.1 times (110%) when the input gradation value is a low input gradation value included in the low gradation region from 1 to 87. The first gain is the maximum gradation that can be set in the subframe image data when the input gradation value is a high input gradation value included in a high gradation region from 88 to 96. The gain is a value (saturation gradation value). In other words, the first gain for the high gradation region can set the 1st output gradation value for the high input gradation value in the 1st subframe image data with an increase rate of 0% with respect to the increase of the high input gradation value. This is a gain for obtaining a maximum gradation value.

On the other hand, as shown in FIG. 15C, the second gain application unit 415 applies a second gain lower than the target gain as the reference gain to the input gradation value of the 1st subframe image data. As described above, the second gain is a gain that is lower than the target gain and has a different slope depending on the area of the input gradation value. Specifically, the second gain is a gain whose average with the first gain is 1.0 times (100%), and for the input gradation values in the low gradation region from 1 to 87, Is 0.9 times (90%). For the input gradation values in the high gradation region from 88 to 96, the gain is 1.0 times (100%) on average on average with the first gain for obtaining the saturation gradation value.
The output switching unit 419 alternately outputs the first subframe image data after the gain calculation output from the first gain application unit 414 and the 2nd subframe image data after the gain calculation output from the second gain application unit 415. The data is output to the gamma conversion unit 421. Thereby, on each liquid crystal element and the projection surface, a 1st subframe image that is an image corresponding to the 1st subframe image data after gain calculation and a 2nd subframe that is an image corresponding to 2nd subframe image data after gain calculation are displayed. Images are displayed sequentially (alternately). The brightness of these two sub-frame images is averaged, so that brightness and gradation are not deteriorated compared to a projection image (frame image) projected without gain calculation, and the disclination dark line However, it is possible to display a projected image that is difficult to be visually recognized by an observer.

  The input gradation value is X, the output gradation value from the first gain application unit 414 is Y1, and the output gradation value from the second gain application unit 415 is Y2. Further, the maximum value of the output gradation value is Ymax, and the gain applied to the low gradation region in the first gain application unit 414 is A (1 <A ≦ 2). At this time, the gain applied by the first gain applying unit 414 and the second gain applying unit 415 to the input gradation value of each subframe image data is expressed as follows.

When X <Ymax / A, Y1 = AX (1)
When X ≧ Ymax / A, Y1 = Ymax (2)
When X <Ymax / A, Y2 = (2-A) X (3)
When X ≧ Ymax / A, Y2 = 2X−Ymax (4)
The reason why the disclination dark line becomes difficult to be visually recognized by using the gain as described above will be described.

  When the gain calculation is not performed, as shown in FIG. 15A, output gradation values 64 and 65 are displayed on adjacent liquid crystal pixels whose horizontal coordinates (that is, input gradation values) H are 64 and 65, respectively. At this time, for the reasons described above, a disclination dark line is generated at the position of the adjacent liquid crystal pixel at the horizontal coordinates H = 64,65.

  On the other hand, when the first gain shown in FIG. 15B is applied to the input gradation values 64 and 65, the output gradation value is at the position of the adjacent liquid crystal pixel whose horizontal coordinate H is 58 and 59. 64, 65 are displayed. A disclination dark line is generated at the position of the adjacent liquid crystal pixels at the horizontal coordinates H = 58 and 59.

  On the other hand, when the second gain shown in FIG. 15C is applied to the input gradation values 64 and 65, the output gradation values 64 and 65 are located at the positions of the adjacent liquid crystal pixels whose horizontal coordinates H are 71 and 72, respectively. Is displayed. A disclination dark line is generated at the position of the adjacent liquid crystal pixels at the horizontal coordinates H = 71 and 72.

  By alternately projecting (displaying) the 1st subframe image and the 2nd subframe image after gain calculation obtained in this way, the generation position of the disclination dark line changes on the projection surface. For this reason, the disclination dark line visually recognized by the observer is about ½, and the disclination dark line can be made difficult to see as shown in FIG.

  Further, as shown in FIG. 15B, all the 87 to 96 gradations (high input gradation values) in the high gradation region in the 1st subframe image data are converted to the same saturation gradation value. Accordingly, no disclination dark line is generated in the high gradation region of the 1st subframe image. On the other hand, there is a possibility that a disclination dark line may occur in the high gradation region of the 2nd subframe image, but no disclination dark line occurs in the high gradation region of the 1st subframe image. For this reason, the disclination dark line does not occur continuously at the same position (pixel) in the continuous 1st and 2nd subframe images. Therefore, the disclination dark line is about ½ in any region in the projected image, and the disclination dark line can be made difficult to be visually recognized.

  Further, the first gain and the second gain are gains that average 100% with respect to each input gradation value, that is, gains corresponding to the reference gain shown in FIG. Is set. This means that the luminance is reduced by the gain calculation using the second gain by the increase in the luminance by the gain calculation using the first gain. Therefore, it is possible to express brightness and gradation equivalent to the brightness and gradation of the projected image displayed without performing gain calculation for all input gradation values of the input frame image data.

  Further, when displaying the first sub-frame image including the high gradation region converted into the saturation gradation value by the gain calculation, the liquid crystal pixel displaying the saturation gradation value does not generate disclination. It is desirable that the width of the high gradation region converted into the gradation value is large. However, if the gradation value is converted to a saturation gradation value up to a low gradation region of 50% or less of the maximum gradation value that can be input, the gradation property at the average between the 1st and 2nd subframe images cannot be maintained.

  For example, FIG. 16A shows a first gain for converting an input gradation value that is 50% or more of the maximum gradation value into a saturation gradation value. The second gain with respect to the first gain is calculated as shown in FIG. In order to maintain the gradation in the average between the 1st and 2nd subframe images, it is necessary to set the second gain so that the input gradation value in the low gradation region is corrected to a negative output gradation value. . However, since a negative output gradation value cannot actually be expressed, the first and second gains as shown in FIG. 16C are set, and thereby, between the 1st and 2nd subframe images. The gradation of the average of the image quality decreases. Therefore, it is desirable to set the first gain so that a gradation value lower than 50% of the maximum gradation value that can be set in the subframe image data is not converted to a saturation gradation value.

  In order to make the average luminance of the two sub-frame images displayed through the gain calculation equal to the luminance of the projection image displayed without the gain calculation, the input gradation value is set to the drive gradation of the liquid crystal pixel. On the other hand, it is desirable to perform gain calculation after conversion so as to have a luminance linear relationship. However, if the difference between the average luminance of the two sub-frame images displayed through the gain calculation and the luminance of the projection image displayed without the gain calculation is within an allowable range, the luminance linear conversion unit 413 is omitted. The gain calculation may be performed without performing the conversion to the luminance linear relationship. By omitting the luminance linear conversion unit 413, the processing can be simplified, the circuit can be reduced in size, and the heat generated from the circuit can be reduced.

  Further, in order to obtain an effect of making the disclination dark line less visible, it is not always necessary to use the first and second gains having the shapes shown in FIGS. For example, as shown in FIG. 17A, the slopes of the first and second gains may be changed smoothly. In particular, when it is difficult to obtain a smooth gradation due to a calculation error caused by omitting the luminance linear conversion unit 413, the first gain with a smoothly changed slope is set as shown in FIG. By doing so, a smoother gradation can be obtained. In the same case, as shown in FIG. 17B, the gain applied to the high gradation region of the first gain may be a gain that does not convert the high gradation region into the saturated gradation value. In other words, the first gain for the high gradation region may be a gain for generating the 1st output gradation value in which the increase rate with respect to the increase of the high input gradation value is less than 100%.

  However, the disclination dark line may be generated in the high gradation region by not converting the high gradation region into the saturated gradation value, and therefore, the gain gradient with respect to the high gradation region is less likely to be generated. It is desirable to keep the tilt small.

  In this embodiment, a gradation value equal to or lower than a predetermined gradation value is set as a low gradation area, a gradation value larger than the predetermined gradation value is set as a high gradation area, and each of the first and second gains is a low gradation area. The case where the gain is different between the high gradation region and the high gradation region has been described. However, other gains may be set. For example, the image generation unit 410 holds a plurality of lookup tables (LUTs) as a first data table and a second data table that respectively indicate a 1st output tone value and a 2nd output tone value with respect to an input tone value. May be. In this case, the double speed conversion unit 411 can generate a plurality of subframe image data having different gains by changing the LUT that is referred to for each subframe image data to be generated.

  Further, although the LUT may be composed of output gradation value data for all input gradations, a large storage capacity is required, so that the LUT is only output gradation value data for typical input gradation values. The tone value conversion may be performed for other input tone values by interpolation processing.

Further, the first and second gains may be changed for each input frame image data.
According to the present embodiment, it is possible to make it difficult to visually recognize deterioration in image quality due to disclination by moving the disclination dark line for each sub-frame image while suppressing a decrease in brightness and gradation of the projected image.

  Next, a second embodiment of the present invention will be described. In the present embodiment, the same reference numerals as those in the first embodiment are assigned to components common to the first embodiment.

  The configuration of the drive circuit unit 303b in the present embodiment is the same as that in the first embodiment, but the operations of the first gain application unit 414 and the second gain application unit 415 are different from those in the first embodiment. Specifically, the first and second gain application units 414 and 415 in the present embodiment perform a first gain calculation for an input gradation value equal to or lower than a predetermined first gradation threshold value, The second gain calculation is performed for an input gradation value that is larger than the gradation threshold value and less than or equal to a predetermined second gradation threshold value. Further, the first and second gain application units 414 and 415 perform a third gain calculation for an input gradation value larger than a predetermined third gradation threshold value. That is, in the present embodiment, each of the first and second gain applying units 414 and 415 can set two change points of the first gain and the second gain applied to the input gradation value. The advantages of adopting such a configuration will be described below.

  In the first embodiment, the appearance of the disclination dark line when displaying a gradation image in which the gradation value is increased by 1 for each pixel in the horizontal direction has been described as an example. That is, the gradation difference between adjacent liquid crystal pixels was 1. When the gradation difference between adjacent liquid crystal pixels is small, the difference between the first and second gains (hereinafter referred to as gain swing width) applied to the input gradation values of the 1st and 2nd subframe image data is small. However, the generation position of the disclination dark line changes. However, the position of the disclination dark line is less likely to change as the difference in gradation value between adjacent liquid crystal pixels increases.

  For example, FIGS. 21A to 21C show an example of displaying a gradation image in which the horizontal gradation value increases by 6 for each pixel. In FIG. 21A, similarly to FIG. 15A of the first embodiment, when gain calculation is not performed, that is, 1.0 times (100%) as the reference gain is applied to the input gradation value. The relationship between the input gradation value and the output gradation value is shown. At this time, there are seven disclination dark lines in the projected image.

  FIGS. 21B and 21C show the horizontal position of the disclination dark line when the first and second gains are set as in FIGS. 15B and 15C of the first embodiment. (Horizontal coordinates). In this case, since the gain amplitude is small with respect to the gradation difference between adjacent liquid crystal pixels in which the disclination dark line is generated, the change in the output gradation value by the gain calculation is small, and the position of the disclination dark line is Is hard to change. As a result, the disclination dark line may be visually recognized by the observer as shown in FIG.

  On the other hand, in this embodiment, the gain swing width is made larger than that in the first embodiment. The first and second gains when A in Formula (1) described in Embodiment 1 is set to 1.4 and the appearance of the disclination dark line in the projected image when these are applied are shown in FIG. d) and (e). In this case, the gain swing width is sufficiently large with respect to the gradation difference between adjacent liquid crystal pixels in which the disclination dark line is generated. For this reason, the change in the gradation value due to the gain calculation also increases, and the position of the disclination dark line changes for each subframe image. As a result, as shown in FIG. 22G, the disclination dark line can be made difficult to be visually recognized.

  Since various gradation values can be adjacent to each other in actual input frame image data, the number of combinations of gradation values that can make it difficult to see the disclination dark line can be increased by increasing the gain swing width difference. Can do.

  For the reason described above, a case where the gain swing width in the first embodiment is increased will be described. Specifically, in the first embodiment, a case where the gain swing width is set large as shown in FIG. 19A and the disclination dark line generated in the vicinity of the input gradation value X1 is made less visible will be described. . In this case, for the reason described above, it is effective to apply a larger gain to the input gradation value X1 in order to make it difficult to visually recognize the disclination dark line. An example in which a larger gain is applied to the input gradation value X1 is shown in FIG. The gain swing width at the input gradation value X1 is sufficiently larger than that in FIG.

  On the other hand, paying attention to the vicinity of the input gradation value X2 on the higher gradation side than the input gradation value X1, the gain swing width is considerably larger than that in FIG. In this way, if the gain swing width becomes too large, the difference in brightness between the 1st and 2nd subframe image data output from the output switching unit 419 increases, and thus the brightness of the large difference is flickered by the observer. There is a risk of being visually recognized.

  Therefore, the first and second gain application units 414 and 415 in the present embodiment set the first and second gains having two change points as shown in FIG. 19C. When attention is focused on the vicinity of the input tone value X1, a sufficiently large gain amplitude is secured as in FIG. 19B, so that the disclination dark line can be made difficult to visually recognize. Further, in the vicinity of the input gradation value X2 on the high gradation side, the gain swing width is set to be equal to that in FIG. 19A, so that the difference in brightness between the 1st and 2nd subframe image data does not increase. Visual recognition of the flicker can be suppressed.

  As described above, in the present embodiment, the change point of the first and second gains is increased to two (or more), thereby suppressing the flicker and suppressing the low gradation side. Disclination dark lines can be made more difficult to see.

  Next, Embodiment 3 of the present invention will be described. In the present embodiment, the same reference numerals as those in the first embodiment are assigned to components common to the first embodiment.

  FIG. 18 shows the internal configuration of the drive circuit section 303b in the present embodiment. The third embodiment is different from the first embodiment in that a third gain application unit 416 and a fourth gain application unit 417 are added between the luminance linear conversion unit 413 and the output switching unit 419. The basic operations of the third gain application unit 416 and the fourth gain application unit 417 are the same as those of the first gain application unit 414. Hereinafter, a different part from Example 1 in a present Example is demonstrated.

  In the first embodiment, the double speed conversion unit 411 generates the 1st and 2nd subframe image data by doubling the frequency of the vertical synchronization signal of the input frame image data. In this case, for example, if the frequency of the vertical synchronization signal of the input frame image data is 24 Hz, the frequency of the vertical synchronization signal of each subframe image data is as low as 48 Hz, and flicker may be visually recognized. For this reason, the frequency of the vertical synchronization signal of each subframe image data may be set high (for example, about 90 Hz or more) so that the flicker is not visually recognized.

  Therefore, in this embodiment, the double speed conversion unit 411 generates a plurality of subframe image data so that the frequency of the vertical synchronization signal of each subframe image data is equal to or higher than a predetermined frequency. The number of subframe image data generated by the double speed conversion unit 411 may be determined by the CPU 200 based on the frequency of the vertical synchronization signal measured by the input unit 303a, or the double speed change unit 411 may perform vertical synchronization of the input frame image data. The frequency of the signal may be measured and determined.

As a specific example, when the frequency of the vertical synchronization signal of each input frame image data is 24 Hz, the double speed conversion unit 411 in the present embodiment uses four subframe images at a cycle corresponding to 96 Hz from the input frame image data. Generate data. The four subframe image data are referred to as the first subframe image data, the second subframe image data, the 3rd subframe image data, and the 4th subframe image data in order from the one generated earlier.
The double speed conversion unit 411 generates a field signal for identifying which subframe image data is the 1st, 2nd, 3rd, and 4th subframe image data, and outputs the generated field signal to the output switching unit 419. To do.

  As in the first embodiment, the first and second gain application units 414 and 415 respectively apply the first gain and the second gain to the input gradation values of the 1st subframe image data and the 2nd subframe image data. Perform the gain processing to be applied. Then, the 1st and 2nd subframe image data after the gain processing is output to the output switching unit 419. The third gain application unit 416 performs gain processing that applies the third gain to the input tone value of the 3rd subframe image data (third subframe image data), and outputs the 3rd output tone value after the gain processing. The 3rd subframe image data having the same is output to the output switching unit 419. The fourth gain application unit 417 performs gain processing for applying the fourth gain to the input tone value of the 4th subframe image data (fourth subframe image data), and outputs the 4th output tone value after the gain processing. The 4th subframe image data having the same is output to the output switching unit 419.

  The output switching unit 419 is synchronized with the vertical synchronization signal of each subframe image data, and is output from the first to fourth gain application units 414 to 417 in accordance with the field signal from the double speed conversion unit 411. Subframe image data is sequentially switched and output. As a result, the 1st to 4th subframe image data after the gain calculation are sequentially input to the panel drive unit 420.

  FIGS. 20A to 20G show the first to fourth gains applied by the first to fourth gain application units 414 to 417 in the present embodiment and the projected image visually recognized by the observer. The disclination dark line is shown. In FIG. 20A, similarly to FIG. 15A of the first embodiment, when gain calculation is not performed, that is, 1.0 times (100%) as the reference gain is applied to the input gradation value. The relationship between the input gradation value and the output gradation value is shown. At this time, there are seven disclination dark lines in the projected image.

  FIGS. 20B and 20C respectively show the first and second gains applied by the first and second gain application units 414 and 415, respectively. These first and second gains are the same as the first and second gains shown in FIGS. 15B and 15C of the first embodiment.

  When these first and second gains are also applied to the 3rd and 4th subframe image data, respectively, the first gain is applied to the 1st and 3rd subframe image data, and the second gain is applied to the 2nd and 4th subframe image data. Will be applied. In this case, a disclination dark line is generated at the same position in at least two subframe images among the four subframe images projected (displayed). For this reason, the darkness of the disclination dark line visually recognized by the observer is about ½ compared to the case where the gain calculation is not performed, and it is possible to make the disclination dark line difficult to see.

  However, in this embodiment, by using gains different from the first and second gains as the third and fourth gains, respectively, the disclination dark line is further made difficult to be visually recognized, as will be described below.

  20D and 20E show the third gain and the fourth gain, respectively. The high gradation region and the low gradation region set by the third and fourth gain application units 416 and 417 are different from the high gradation region and the low gradation region in the first and second gain application units 414 and 415. Specifically, the third and fourth gain application units 416 and 417 use the low gradation region as the region having the input gradation value of 1 to 67 and the high gradation region as the region having the input gradation value of 68 to 96. To do. Then, gains having different slopes are set in the high gradation region and the low gradation region.

  The third and fourth gains are set to gains such that the average of the third and fourth gains is 1 time (100%). Further, unlike the first gain, the third gain is set to a gain that does not convert the input gradation value in the high gradation region of the 3rd subframe image data into the saturated gradation value. The reason will be described below.

  If the high gradation region of the 3rd subframe image data is converted into a saturated gradation value, the gradient of the fourth gain applied to the high gradation region of the 4th subframe image data is 100% on average with the third gain. Therefore, it must be 2.0 times (200%). On the other hand, the slope of the second gain applied in the high gradation region of the 2nd subframe image data is also 2.0 times (200%). As a result, a disclination dark line is generated at the same position in the displayed 2nd and 4th subframe images. As described above, when a disclination dark line is generated at the same position in two subframe images among the four subframe images, the density of the disclination dark line visually recognized by the observer does not become lower than about 1/2. In order to further enhance the effect of making the disclination dark line less visible, it is desirable that the gain gradients applied to the 1st to 4th subframe image data are all different.

  As shown in FIG. 20D, by setting the third gain to a gain that does not convert the high gradation region of the 3rd subframe image data into the saturated gradation value, the slopes of the first to fourth gains are all different. Can be made. As a result, the occurrence positions of the disclination dark lines in the 1st to 4th subframe images to be displayed are different from each other, and the darkness of the disclination dark lines visually recognized by the observer is about 1/4. Therefore, it is possible to further enhance the effect of making the disclination dark line less visible.

  Thus, in this embodiment, four subframe image data are generated from input frame image data. Even in this case, the disclination dark line can be made difficult to be visually recognized by moving the disclination dark line for each sub-frame image while suppressing a decrease in the brightness and gradation of the projected image.

  Here, depending on the order in which the 1st to 4th subframe image data after the gain calculation from the output switching unit 419 is output, flicker may be visually recognized.

  The 1st and 3rd subframe images displayed through the application of the first and third gains exceeding 100% are defined as bright subframe images. In addition, the 2nd and 4th subframe images displayed through the application of the second and fourth gains below 100% are set as dark subframe images. At this time, if the images are displayed in the order of bright subframe image → bright subframe image → dark subframe image → dark subframe image, flicker may be visually recognized. Therefore, it is desirable that the output switching unit 419 alternately output subframe image data corresponding to each of the bright subframe image and the dark subframe image. In other words, it is desirable not to continuously output the subframe image data corresponding to the bright subframe image or to continuously output the subframe image data corresponding to the dark subframe image.

That is, in the present embodiment, it is desirable to cause the output switching unit 419 to output the 1st to 4th subframe image data in any of the following orders.
“1st (bright) → 2nd (dark) → 3rd (bright) → 4th (dark)”
"1st (bright)-> 4th (dark)-> 3rd (bright)-> 2nd (dark)"
“2nd (dark) → 3rd (bright) → 4th (dark) → 1st (bright)”
“2nd (dark) → 1st (bright) → 4th (dark) → 3rd (bright)”
“3rd (bright) → 4th (dark) → 1st (bright) → 2nd (dark)”
"3rd (bright)-> 2nd (dark)-> 1st (bright)-> 4th (dark)"
“4th (dark) → 1st (bright) → 2nd (dark) → 3rd (bright)”
“4th (dark) → 3rd (bright) → 1st (bright) → 2nd (dark)”
In the first to third embodiments, the case where the double speed conversion unit 411 generates an even number of subframe image data for the input frame image data has been described. However, an odd number of subframe image data may be generated. In this case, a gain (reference gain) of 1 (100%) is applied to one of the generated odd number of subframe image data, and the other subframe image data will be described in the first to third embodiments. The same effect can be obtained by applying a gain different from 1 ×.

  Specifically, when the double speed conversion unit 411 generates five subframe image data for the input frame image data, a gain of 1 (100%) is applied to one of the five subframe image data. To do. Then, a gain different from 1 is applied to the other four subframe image data as in the third embodiment.

  In addition, since the output switching unit 419 switches the output of these subframe image data at a speed that is not visually recognized by the observer, the observer is averaged between the five displayed subframe images even if the order of output is changed. A projected image having luminance will be observed. For this reason, the gradation of the projected image can be maintained.

  The sub-frame image data to which the above-described gain of 1 (100%) is applied may be output in what order. However, it is desirable to output the bright subframe images in an order that does not continue.

  As described above, even when an odd number of subframe image data is generated, as in the case of generating an even number of subframe image data, disclination is achieved while suppressing reduction in brightness and gradation of the projected image. The dark line is moved for each sub-frame image. As a result, it is possible to make it difficult to visually recognize deterioration in image quality due to disclination.

  Next, a fourth embodiment of the present invention will be described. In the present embodiment, the same reference numerals as those in the first embodiment are assigned to components common to the first embodiment. The parameter determination method applied to the first and second gain application units 414 and 415 is the same as that in the first embodiment. In this embodiment, first, a dark subframe is output, and then a bright subframe is output. Is different from the first embodiment. Here, the dark subframe is a subframe to which a gain lower than 100% is applied, and the bright subframe is a subframe to which a gain higher than 100% is applied.

  As in the present embodiment, the dark subframe is output first, and then the bright subframe is output, so that the dark line due to the disclination can be made less visible. The reason will be explained below.

  As described in the first embodiment, the disclination takes a time on the order of several milliseconds due to a transient response. On the other hand, there is a transient response in the time from when disclination occurs until the dark line due to disclination is no longer visible (referred to as relaxation time). In order to alleviate disclination, it is necessary to continue a state in which discnation does not occur, that is, a time during which white display is performed in both adjacent liquid crystal pixels or black display is performed in both adjacent liquid crystal pixels for a certain period of time. In the present embodiment, description will be made assuming that the reflective liquid crystal elements 3G, 3R, and 3B are used which have a relaxation time longer than the generation time until the disclination occurs and the darkness becomes visible. As described above, when a liquid crystal element having a long relaxation time is used, the dark line due to disclination can be made less visible by displaying the first subframe as a dark subframe first.

  The reason why it is desirable to display the dark subframe first will be described below with reference to FIG. In this embodiment, the PWM converter 422 is assumed to drive the liquid crystal elements 3G, 3R, and 3B with the PWM shown in FIG. FIG. 23A shows a projection image when the gain calculation is not performed. In addition, the first and second gain applying units 414 and 415 in the present embodiment apply the first and second gains applied to the input gradation value, and the discriminating in the gradation image (projected image) projected on the projection surface. FIGS. 23B and 23C show the relationship between the appearance of image quality degradation due to nations. Further, FIG. 23 (d) shows the state of the disclination dark line visually recognized by the observer.

  Description will be made by paying attention to the projected image area in which the horizontal coordinate H is 87 to 96 in FIGS. For the sake of explanation, this region is referred to as a region of interest. The region of interest is a region where a saturated gradation value is output by calculation using the 2nd subframe applied gain of this embodiment when the gradation image of FIG. 23A is input. In the region of interest of the 1st subframe, disclination occurs between adjacent liquid crystal pixels at horizontal coordinates H of 88 and 89 as shown in FIG. On the other hand, in the region of interest of the 2nd subframe, the saturation gradation value is output, so that no disclination occurs.

  If the 1st subframe is a bright subframe and the 2nd subframe is a dark subframe, a sufficient disclination relaxation period cannot be secured depending on the gradation value of the frame image input next. There is. On the other hand, by setting the 1st subframe as a dark subframe and the 2nd subframe as a bright subframe as in the present embodiment, at least in a region where a gradation of a high gradation region is input, that is, a region of interest in the present embodiment. In other words, even when disclination occurs in the 1st subframe, the white display period continues between the adjacent elephant pixels in the 2nd subframe for one subframe period. Therefore, it is possible to ensure the time for the disclination to be relaxed, and to make it difficult to visually recognize the dark line due to the disclination.

  With reference to FIG. 24, the reason why the time for relaxation of disclination can be surely secured will be described in detail. A square in FIG. 24 represents a pixel, and a number in the square represents a gradation value in each pixel. As shown in FIG. 24A, when input images A and B having a size of 1 × 10 columns are continuously input from the input unit 303a, the 1st subframe is a bright subframe and the 2nd subframe is a dark subframe. An image output from the output switching unit 419 in this case is shown in FIG. FIG. 23C shows an image output from the output switching unit 419 when the 1st subframe is a dark subframe and the 2nd subframe is a bright subframe.

  It should be noted that the hatched pixels in FIGS. 23B and 23C indicate that there are periods in which white display and black display differ between adjacent liquid crystal pixels in PWM in one subframe period. In addition, pixels not hatched in FIGS. 23B and 23C indicate that there is no period in which white display and black display differ between adjacent liquid crystal pixels in PWM in one subframe period. In other words, it is shown that the disclination can be relaxed in the non-hatched pixels in FIGS.

  FIG. 23B shows an output image when the 1st subframe is a bright subframe and the 2nd subframe is a dark subframe. In this case, the 2nd subframe image corresponding to the input image A is a dark subframe, and disclination occurs between adjacent liquid crystal pixels that output the gradation values 80 and 82 of the horizontal coordinates 2 and 3. In the subsequent subframes, the pixels of the horizontal coordinates 2 and 3 are indicated by hatching, which means that there are periods in which white display and black display are different in adjacent pixels in PWM in one subframe period. . Therefore, the generated disclination is difficult to relax.

  As shown in FIG. 23C, when the 1st subframe is a dark subframe and the 2nd subframe is a bright subframe, the 1st subframe image corresponding to the input image A is a dark subframe, Disclination occurs between adjacent liquid crystal pixels that output gradation values 80 and 82. In the next subframe, the 2nd subframe image corresponding to the input image A, all pixels including the pixels of the horizontal coordinates 2 and 3 are not hatched, and white display is performed between adjacent liquid crystal pixels in the PWM of one subframe period. This means that the black display is a pixel that does not have different periods. Therefore, as compared with FIG. 23B, a sufficient time for relaxing the generated disclination is ensured, and even when a liquid crystal element having a long disclination relaxation time is used, the disclination causes It is possible to make the dark line less visible.

  Therefore, according to the present embodiment, it is possible to make it difficult to visually recognize deterioration in image quality due to disclination by moving the disclination dark line for each sub-frame image while suppressing reduction in brightness and gradation of the projected image. it can. Further, as an effect peculiar to the present embodiment, since the dark subframe and the light subframe are output in this order, the disclination relaxation time can be reliably ensured in the high gradation region. Can be made difficult to see.

It is needless to say that the shape of the gain curve in FIG. 23 is merely an example, and the shape described in other embodiments such as FIGS.
(Other examples)
In the present invention, a program that realizes one or more functions of the above-described embodiments is supplied to a system or apparatus via a network or a storage medium, and one or more processors in the computer of the system or apparatus read the program. It can also be realized by executing the processing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

  Each embodiment described above is only a representative example, and various modifications and changes can be made to each embodiment in carrying out the present invention.

303 control circuit 3G, 3R, 3B reflection type liquid crystal element 410 image generation unit 411 double speed circuit 412 image memory 414 first gain application unit 415 second gain application unit 416 third gain application unit 417 fourth gain application unit 419 output switching unit 420 Panel drive unit

Claims (16)

A liquid crystal driving device for driving a liquid crystal element having a plurality of pixels,
Image data generation means for generating a plurality of subframe image data for input frame image data;
The first voltage is applied to each of the plurality of pixels in each of the plurality of subfield periods included in one frame period sequentially based on the gradation values of the plurality of subframe image data. Driving means for controlling the application of a second voltage lower than the voltage of the pixel to form a gradation in the pixel,
When the gradation value of the input frame image data is referred to as an input gradation value,
The image data generation means generates first subframe image data having at least a first gradation value higher than the input gradation value as the plurality of subframe image data, and from the input gradation value A liquid crystal driving device that generates second sub-frame image data having a lower second gradation value.   2. The liquid crystal driving device according to claim 1, wherein an average of the first and second gradation values with respect to the same input gradation value is a gradation value corresponding to the input gradation value.   The image data generation means generates the first sub-frame image data having the first gradation value by applying a first gain exceeding 100% to the input gradation value, The second sub-frame image data having the second gradation value is generated by applying a second gain lower than 100% with respect to the input gradation value. The liquid crystal driving device described.   The liquid crystal driving device according to claim 3, wherein the first and second gains are gains having an average of 100%.   The first gradation value with respect to the high input gradation value included in a predetermined high gradation region among the input gradation values is a gradation value with an increase rate of less than 100% with respect to the increase of the high input gradation value. The liquid crystal driving device according to claim 1, wherein the liquid crystal driving device is a liquid crystal driving device.   The first gradation value with respect to the high input gradation value is a maximum gradation value that can be set in the first sub-frame image data while the increase rate is 0%. Item 6. The liquid crystal drive device according to Item 5.   7. The liquid crystal according to claim 5, wherein the high gradation region is a region having a gradation value of at least 50% or more of a maximum gradation value that can be set in the first sub-frame image data. Drive device.   8. The liquid crystal driving device according to claim 1, wherein the first and second gradation values are different gradation values for each region of the input gradation value. 9. The image data generating means includes a first data table including data of the first gradation value for the input gradation value and a second data including data of the second gradation value for the input gradation value. A data table,
9. The liquid crystal driving device according to claim 1, wherein the first and second table data are referred to in the generation of the first and second sub-frame image data, respectively. 10.   The said image data generation means produces | generates 1st sub-frame image data, after producing | generating the said 2nd sub-frame image data with respect to the said input frame image data, The any one of Claim 1 to 9 characterized by the above-mentioned. The liquid crystal drive device according to one item. The image data generation means includes, in addition to the first and second subframe image data, third subframe image data having a third gradation value higher than the input gradation value, and the input gradation Generating fourth sub-frame image data having a fourth gradation value lower than the value,
11. The third gradation value is different from the first gradation value, and the fourth gradation value is different from the second gradation value. The liquid crystal drive device according to item.   The image data generation means generates an odd number of frame image data including subframe image data having the same gradation value as the input gradation value in addition to the first and second subframe image data. The liquid crystal driving device according to claim 1, wherein A light source for inputting light to the liquid crystal element;
The liquid crystal driving device according to claim 1, further comprising a projection lens that projects light modulated by the liquid crystal element onto a projection surface. A liquid crystal element;
An image display device comprising the liquid crystal driving device according to claim 1. A liquid crystal driving method for driving a liquid crystal element having a plurality of pixels,
Generating a plurality of sub-frame image data for the input frame image data;
The first voltage is applied to each of the plurality of pixels in each of the plurality of subfield periods included in one frame period sequentially based on the gradation values of the plurality of subframe image data. Controlling the application of a second voltage lower than the voltage of the pixel to form a gradation in the pixel,
When the gradation value of the input frame image data is referred to as an input gradation value, the first subframe image data having at least a first gradation value higher than the input gradation value as the plurality of subframe image data. And generating second sub-frame image data having a second gradation value lower than the input gradation value. In a computer that controls driving of a liquid crystal element having a plurality of pixels,
A process of generating a plurality of subframe image data for the input frame image data;
The first voltage is applied to each of the plurality of pixels in each of the plurality of subfield periods included in one frame period sequentially based on the gradation values of the plurality of subframe image data. A computer program for executing a process of forming a gradation in the pixel by controlling application of a second voltage lower than the voltage of
When the gradation value of the input frame image data is referred to as an input gradation value,
Causing the computer to generate first subframe image data having at least a first gradation value higher than the input gradation value as the plurality of subframe image data, and lower than the input gradation value. A liquid crystal driving program for generating second sub-frame image data having a gradation value of 2.