JP2013005362A - Image processing device and display device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device and a display device that are capable of performing crosstalk reduction processing causing small side effects.SOLUTION: An image processing device comprises: a correction amount calculation unit that calculates a first correction amount in a case where predetermined filtering processing is applied to each of plural input viewpoint images; and an adjustment unit that adjusts the first correction amount using such a gain as to vary according to a gradation level relation between the plural viewpoint images.

Description

  The present disclosure relates to an image processing device and a display device to which a plurality of viewpoint images are input.

  As a technique for performing 3D (stereoscopic) display, there are a spectacle method using spectacles for stereoscopic vision and a naked-eye method enabling stereoscopic viewing with the naked eye without using special glasses for stereoscopic vision. As the naked eye method, there is a method using an optical element (parallax separation means) such as a parallax barrier or a lenticular lens. In the case of this naked eye method, when an observer views a two-dimensional image displayed on the two-dimensional display device through the parallax separation means, the image seen from a certain direction corresponds to the arrangement pitch of a parallax barrier, a lenticular lens, or the like. Only the pixels at discrete positions of the display device are provided. When viewed from another direction, the pixels at different discrete positions are seen. Thereby, different images can be presented to both eyes, and a stereoscopic display is realized.

  As a typical spectacle method, when displaying a right-eye image and a left-eye image on a two-dimensional display device, image selectivity is provided by time division, space division, wavelength division, or the like. The selectivity corresponding to this is given. As a result, different images can be presented to the right and left eyes of the observer, and stereoscopic display is possible.

JP 2009-251098 A

  As a problem of the stereoscopic display device, there is a so-called crosstalk problem in which the stereoscopic display performance is impaired. Crosstalk is a phenomenon in which not only the viewpoint image that should originally be presented, but also other viewpoint images are observed in one eye of the observer (for example, the right eye is mixed with an image for the left eye). Will be). In the case of the naked eye system, when viewed from a certain direction through the optical element, crosstalk occurs because an image in another direction appears to be mixed without being sufficiently separated by the optical element. In the case of the glasses method, crosstalk occurs when an image to be presented to one eye is mixed with an image to be presented to the other eye without being sufficiently separated by the glasses.

  In order to reduce the crosstalk, there is a method of performing a filtering process using a high-pass filter or the like. However, in this case, the number of gradations that can be processed is limited. Patent Document 1 discloses a technique for suppressing the occurrence of crosstalk by correcting display of pixels displayed in one direction based on image data displayed in another direction. However, with the technique described in Patent Document 1, correction is not sufficient for the adverse effects of crosstalk in a stereoscopic image in which a bright object has parallax in a dark background. This is because the technique described in Patent Document 1 cannot set the correction signal below the black display level of the signal.

  An object of the present disclosure is to provide an image processing device and a display device that can perform crosstalk reduction processing with few side effects.

  An image processing apparatus according to the present disclosure includes a correction amount calculation unit that calculates a first correction amount when a predetermined filtering process is performed on each of a plurality of input viewpoint images, and a first correction amount, And an adjustment unit that adjusts with a gain that changes according to the relationship of the gradation levels between a plurality of viewpoint images.

  A display device according to the present disclosure includes an image processing unit that performs a predetermined filtering process on each of a plurality of viewpoint images, and a display unit that displays an image based on the plurality of viewpoint images after the filtering process is performed by the image processing unit The image processing unit is configured by the image processing apparatus of the present disclosure.

  In the image processing device or the display device according to the present disclosure, the first correction amount when the predetermined filtering process is performed on each of the plurality of input viewpoint images is the relationship between the gradation levels between the plurality of viewpoint images. It is adjusted by a gain that changes in accordance with.

  According to the image processing device or the display device of the present disclosure, the first correction amount when the predetermined filtering process is performed is adjusted by a gain that changes according to the relationship of the gradation levels between the plurality of viewpoint images. Since this is done, it is possible to perform crosstalk reduction processing with few side effects.

It is a block diagram showing an example of 1 composition of a display concerning an embodiment of this indication. It is a top view which shows the 1st example of the combination of the pixel of a display part, and a parallax element. It is a top view which shows the 2nd example of the combination of the pixel and parallax element of a display part. It is explanatory drawing which shows the principle of crosstalk generation | occurrence | production. It is explanatory drawing which shows the 1st example which correct | amended crosstalk by the filtering process. It is explanatory drawing which shows the correction | amendable range of the crosstalk by filtering process. It is explanatory drawing which shows the concept of the correction | amendment of crosstalk in case a correction amount exceeds a dynamic range. It is explanatory drawing which shows the concept which adjusts the amount of correction | amendment when exceeding a dynamic range. It is explanatory drawing which shows the example which adjusted the amount of correction | amendment when exceeding a dynamic range. It is a flowchart which shows an example of the process of crosstalk correction.

  Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[Configuration of display device]
FIG. 1 illustrates a configuration example of a display device according to an embodiment of the present disclosure. The display device includes an image processing unit that performs a predetermined filtering process on the input signal Vin, and a display unit 1 that displays an image based on the output signal Vout after the filtering process is performed by the image processing unit. .

  The input signal Vin is data including a plurality of viewpoint images for stereoscopic display, for example. The output signal Vout is data including a plurality of viewpoint images that have been subjected to a predetermined filtering process by the image processing unit.

  The display unit 1 includes a two-dimensional display such as a liquid crystal display panel, an organic EL (electroluminescence), or a plasma display. A plurality of pixels are two-dimensionally arranged on the display screen of the display unit 1. On the display screen of the display unit 1, an image is displayed according to the stereoscopic display method of the display device based on a plurality of viewpoint images.

  The stereoscopic display method of this display device is not particularly limited. A glasses method may be used, or a naked eye method such as a parallax barrier method or a lenticular lens method may be used. For example, in the case of a glasses system using shutter glasses, for example, left and right two viewpoint parallax images (a left-eye parallax image and a right-eye parallax image) are alternately displayed in time division on the display unit 1 as a plurality of viewpoint images. Is done. Further, for example, in the case of the naked eye method, the display unit 1 displays a parallax composite image obtained by combining a parallax image of two left and right viewpoints or a parallax image of multiple viewpoints in one screen as a plurality of viewpoint images. That is, a plurality of viewpoint images are displayed by being divided into spaces.

  2 and 3 show an example of a stereoscopic display device using a lenticular lens system. In this stereoscopic display device, a lenticular lens 10 as a parallax element is disposed opposite to the display unit 1. 2 and 3, the display unit 1 includes a plurality of pixels including R (red) pixels 11R, G (green) pixels 11G, and B (blue) pixels 11B, and the plurality of pixels. Are arranged in a matrix. 2 and 3 show an example in which 4-viewpoint display is performed. The numbers 0 to 3 given in the pixels 11R, 11G, and 11B indicate the numbers of four types of viewpoint images (viewpoint numbers) assigned to the pixels 11R, 11G, and 11B.

  The lenticular lens 10 has, for example, a plurality of cylindrical divided lenses 10A arranged in parallel. FIG. 2 shows an example in which the split lens 10A is arranged in parallel in the horizontal direction so that the generatrix direction is vertical. FIG. 3 is an example in which the split lens 10A is arranged in parallel in the horizontal direction by tilting so that the generatrix direction is oblique. An example (oblique lenticular method) is shown. The lenticular lens 10 spatially separates a plurality of viewpoint images displayed on the display unit 1 and emits them to the viewer side.

  In addition, it can replace with the lenticular lens 10, and it can be set as the structure of the stereoscopic display apparatus by a parallax barrier system by arrange | positioning a parallax barrier element as a parallax element. The parallax barrier element includes an opening that transmits light and a shielding portion that shields light.

  Returning to FIG. 1, the configuration of the image processing unit will be described. The image processing unit includes a crosstalk correction amount calculation unit 2, a correction amount adjustment unit 9, and an adder 5. The correction amount adjustment unit 9 includes a crosstalk correction amount clip unit 3, a gain calculation unit 4, an adder 6, an adder 7, and a multiplier 8.

The crosstalk correction amount calculation unit 2 calculates a first correction amount (Δ _{i, 1} ) when a predetermined filtering process is performed on each of a plurality of input viewpoint images. The predetermined filtering process here is, for example, a high-pass filter process as in a specific example described later.

The correction amount adjustment unit 9 adjusts the first correction amount (Δ _{i, 1} ) with a gain (G _i ) that changes according to the relationship of the gradation levels between the plurality of viewpoint images. Although a specific example will be described later, the gain (G _i ) is obtained when a predetermined filtering process is performed on a desired first correction amount (Δ _{i, 1} ) before clipping and each of a plurality of viewpoint images. This is based on the ratio to the second correction amount (Δ _{i, 2} ) after clipping, which is the correction limit. The correction amount adjustment unit 9 uses the same gain for each of the plurality of viewpoint images.

The adder 5 multiplies the difference amount (the output value of the adder 6) between the first correction amount (Δ _{i, 1} ) and the input value of each viewpoint image by the gain (G _i ), and the value of each viewpoint image. It is added to the input value.

[Filtering processing by comparative example]
Prior to describing a specific example of image processing in the present embodiment, the principle of crosstalk generation and filtering processing according to a comparative example for reducing the crosstalk will be described.

  Here, the crosstalk generated in the lenticular lens system shown in FIGS. 2 and 3 will be described as an example. In the case where the split lens 10A is not tilted as shown in FIG. 2, moire tends to occur. As a countermeasure against the moire, there is a method in which the divided lens 10A is inclined as shown in FIG. However, when the split lens 10A is tilted, there is a problem that crosstalk between adjacent viewpoints increases as a side effect.

  Here, in the configuration as shown in FIG. 3, the principle of crosstalk generation when a multi-viewpoint image is output without performing crosstalk improvement image processing (filtering processing) is shown in FIGS. Is shown schematically. FIG. 4B shows the gradation level (signal level) of an arbitrary nth viewpoint image. 4A shows the gradation level of the (n-1) th viewpoint image, and FIG. 4 (C) shows the gradation level of the (n + 1) th viewpoint image. FIG. 4D shows the state of the viewpoint image observed at the nth viewpoint position. In FIG. 4D, the viewpoint images of FIGS. 4A to 4C are combined. At the n-th viewpoint position, it is ideal that the viewpoint image in FIG. 4B is observed, but the adjacent (n−1) -th and n + 1-th viewpoint images are crosstalk components (ghost components). It will be observed.

The crosstalk component at the four viewpoints is expressed by X according to the following (Equation 1). In (Expression 1), for example, V _{1, in} represents the input value of the first viewpoint image, and V _{1, out} represents the output value of the first viewpoint image. As shown in FIG. 4D, it can be seen that the contour is not clear and looks like a ghost due to the crosstalk component of the adjacent viewpoint.

On the other hand, crosstalk can be improved by performing correction by a filtering process (high-pass filter process) by multiplying the inverse matrix X ⁻¹ represented by (Equation 2). 5A to 5D show an example in which such filtering processing is performed. In the state shown in FIG. 5D, crosstalk is suppressed and the image can be clearly seen in the state shown in FIG.

  However, as shown in FIG. 6, there is a limit to the combination of gradations that can be filtered, and in some cases, side effects may be more noticeable. In FIG. 6, for example, the horizontal axis indicates the gradation level (signal level) of another viewpoint image serving as a background (crosstalk component) with respect to the n-th viewpoint image. If the gradation of a certain point is on the high gradation side and the background gradation is on the halftone side, the background gradation has a margin on the low gradation side, so correction can be applied. In the correction of the portion on the key side, the input gradation reaches a peak, reaching the limit to be corrected, and the desired correction cannot be performed, and as a result, for example, FIG. As shown in (D), the correction in the low gradation direction of the background is extremely applied, the crosstalk is deteriorated, and the stereoscopic display may be deteriorated in some cases.

[Specific example of filtering processing of this embodiment]
In the present embodiment, in order to solve the problem of deterioration in crosstalk due to the filtering process according to the comparative example described above, the gain value of the correction amount of the filtering process is changed according to the combination of gradations. When a desired correction amount cannot be applied (or approaches the limit) as shown in FIGS. 7A to 7D, the gain value of the correction amount is decreased. A specific method is shown in FIGS. Although there are many methods for correcting the gain value, as shown in FIGS. 8A and 8B, a desired correction amount Δ1 (first correction amount) and a limit correction amount Δ2 actually applied (second correction) It is desirable to calculate the ratio to the correction amount of the background (crosstalk component). As a result, as shown in FIGS. 9A to 9D, crosstalk components can be reduced while minimizing the side effects of correction.

FIG. 10 shows an example of filtering processing for reducing crosstalk in the present embodiment. A plurality of viewpoint images V _{i, in} (i = 0 to N) are input as the input signal Vin (step S11). The crosstalk correction amount calculation unit 2 uses the first correction amount (when the predetermined filtering process is performed on each of the plurality of viewpoint images V _{i, in} , for example, using the arithmetic expression such as (Expression 2) described above. Δ _{i, 1} ) is calculated (step S12).

The crosstalk correction amount clip unit 3 calculates a _second correction amount (Δ _{i, 2} ) after clipping that becomes a correction limit when a predetermined filtering process is performed (step S13). In the gain calculation unit 4, for example, a gain (G _i ) expressed by a ratio between a desired first correction amount (Δ _{i, 1} ) before clipping and a second correction amount (Δ _{i, 2} ) after clipping. ) Is calculated (step S14). The multiplier 8 finally obtains a value obtained by multiplying the amount of difference (output value of the adder 6) between the first correction amount (Δ _{i, 1} ) and the input value of each viewpoint image by the gain (G _i ). The correction amount xi is obtained (step S15). The adder 5 outputs an output value V _{i, out} obtained by adding the final correction amount xi to each of the plurality of viewpoint images V _{i, in} as an output signal Vout (step S16).

[effect]
As described above, according to the display device according to the present embodiment, the first correction amount when the predetermined filtering process is performed is changed according to the relationship between the gradation levels between the plurality of viewpoint images. Since the gain is adjusted by such a gain, crosstalk reduction processing with few side effects can be performed. Thereby, it is possible to perform a stereoscopic display with less blur and ghost.

<Other embodiments>
The technology according to the present disclosure is not limited to the description of the above embodiment, and various modifications can be made.
For example, in the above-described embodiment, a case where stereoscopic display is performed based on a plurality of viewpoint images having parallax with each other is described as an example, but a multi-view display that presents different images in a plurality of directions instead of stereoscopic display In addition, the present technology is applicable.

For example, this technique can take the following composition.
(1)
A correction amount calculation unit that calculates a first correction amount when a predetermined filtering process is performed on each of a plurality of input viewpoint images;
An image processing apparatus comprising: an adjustment unit that adjusts the first correction amount by a gain that changes in accordance with a relationship between gradation levels between the plurality of viewpoint images.
(2)
The image processing apparatus according to (1), wherein the predetermined filtering process is a high-pass filter process.
(3)
The above (1) or (2) further comprising an adding unit that adds a value obtained by multiplying the difference between the first correction amount and the input value of each viewpoint image by the gain to the input value of each viewpoint image. An image processing apparatus according to 1.
(4)
The gain is
Based on the ratio between the first correction amount and the second correction amount that is a correction limit when the predetermined filtering process is performed on each of the plurality of viewpoint images. The image processing apparatus according to any one of (3).
(5)
The image processing apparatus according to any one of (1) to (4), wherein the adjustment unit uses the same gain for each of the plurality of viewpoint images.

  DESCRIPTION OF SYMBOLS 1 ... Display part, 2 ... Crosstalk correction amount calculation part, 3 ... Crosstalk correction amount clip part, 4 ... Gain calculation part, 5 ... Adder, 6 ... Adder, 7 ... Adder, 8 ... Multiplier, 9 ... Correction amount adjustment unit, 10... Lenticular lens, 10A... Splitting lens, 11R... Red pixel, 11G ... green pixel, 11B .. blue pixel, Vin ... input signal, Vout ... output signal.

Claims (6)

A correction amount calculation unit that calculates a first correction amount when a predetermined filtering process is performed on each of a plurality of input viewpoint images;
An image processing apparatus comprising: an adjustment unit that adjusts the first correction amount by a gain that changes in accordance with a relationship between gradation levels between the plurality of viewpoint images. The image processing apparatus according to claim 1, wherein the predetermined filtering process is a high-pass filter process. The image processing according to claim 1, further comprising: an addition unit that adds a value obtained by multiplying a difference amount between the first correction amount and an input value of each viewpoint image by the gain to an input value of each viewpoint image. apparatus. The gain is
The method according to claim 1, wherein the first correction amount is based on a ratio between the first correction amount and a second correction amount that is a correction limit when the predetermined filtering process is performed on each of the plurality of viewpoint images. Image processing apparatus. The image processing apparatus according to claim 1, wherein the adjustment unit uses the same gain for each of the plurality of viewpoint images. An image processing unit that performs a predetermined filtering process on each of the plurality of viewpoint images;
A display unit configured to display an image based on the plurality of viewpoint images after being filtered by the image processing unit,
The image processing unit
A correction amount calculation unit that calculates a first correction amount when the predetermined filtering process is performed on each of the input viewpoint images;
A display device comprising: an adjustment unit that adjusts the first correction amount with a gain that changes in accordance with a relationship between gradation levels between the plurality of viewpoint images.

Images