JP2014127788A - Device of correcting stereoscopic image, and program of the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device of correcting a stereoscopic image capable of correcting a depth of a stereoscopic image within a proper range for an observer.SOLUTION: A stereoscopic image correction device 1 corrects a depth range of an element image group photographed by a stereoscopic image photographing device 100 when a stereoscopic image is displayed by an IP system in a stereoscopic image display device 200. The stereoscopic image correction device 1 comprises: depth range calculation means 11 calculating a display depth range at the time when the element image group is displayed at a stereoscopic image display device 200; depth range determination means 13 determining whether or not the display depth range calculated by the depth range calculation means 11 exceeds a reference depth range; and stereoscopic image processing means 20 correcting the display depth range of the element image group so that the display depth range falls within the reference depth range in the case that the display depth range exceeds the reference depth range.

Description

  The present invention relates to a stereoscopic image correction apparatus for correcting an element image group obtained by photographing a subject by an integral photography method and a program therefor.

  Conventionally, as one of the stereoscopic image display methods capable of freely viewing stereoscopic images from an arbitrary viewpoint, integral photography (Integral Photography) using a group of convex lenses or pinholes arranged in a planar shape is possible. Hereinafter, the IP) method is known.

  Hereinafter, normal stereoscopic image capturing and stereoscopic image display based on the IP method will be described with reference to FIGS. 18 and 19. As illustrated in FIG. 18, the stereoscopic image capturing device 910 includes a lens group 912 in which convex lenses are arranged on the same plane, and an imaging plate 913. FIG. 18 illustrates a subject 911, a photographing direction 914 of the stereoscopic image photographing device 910, and an element image 915 of the subject 911 formed by the lens group 912. When the subject 911 is viewed from the shooting direction 914, the cylinder is in front of the prism.

  The stereoscopic image photographing device 910 photographs a subject 911 through a lens group 912. Then, the same number of element images 915 of the subject 911 as the convex lenses that form the lens group 912 are generated on the imaging plate 913.

  As shown in FIG. 19, the stereoscopic image display device 920 includes a lens group 922 in which convex lenses are arranged on the same plane, and a display element 923. Further, FIG. 19 illustrates a stereoscopic image 921, an observation direction 924 of the observer 926, an element image 925, and an observer 926. The display element 923 displays an element image 925 corresponding to the element image 915 photographed by the imaging plate 913 of the stereoscopic image photographing device 910.

  As a result, as shown in FIG. 19, the stereoscopic image 921 is generated such that the distance from the display element 923 is equal to the distance between the subject 911 and the imaging plate 913 in FIG. At this time, in the stereoscopic image 921 corresponding to the subject 911, when viewed from the observation direction 924, the prism is in front of the cylinder. That is, in normal stereoscopic image shooting based on the IP method, a reverse-view image (stereoscopic image 921) having a depth inverted compared to the subject 911 in FIG. 18 is generated.

In FIGS. 18 and 19, the optical element array is described as the lens group 912 in which minute convex lenses are arranged. However, the optical element array may be an aperture array (spatial filter) in which minute pinholes are arranged. Good.
Further, although the display element 923 has been described as displaying the element image 925 captured by the imaging plate 913, an image generated by a computer (not shown) may be displayed.

  Therefore, an invention for solving the above-described problem of reverse vision has been proposed (for example, Patent Document 1). The invention described in Patent Document 1 performs arithmetic processing on the information acquired by the stereoscopic image capturing device 910 in FIG. 18 and inputs the information after the arithmetic processing to the stereoscopic image display device 920 in FIG. A stereoscopic image 921 having a correct depth is generated.

  Hereinafter, with reference to FIGS. 20 and 21, an image depth conversion device 930 described in Patent Document 1 will be described. As shown in FIG. 20, the image depth conversion device 930 receives the element image 931 captured by the stereoscopic image capturing device 910 of FIG. 18, and this element image 931 is virtually formed through the first virtual lens array 932. The obtained stereoscopic image 933 is obtained by arithmetic processing. Then, the image depth conversion device 930 obtains an element image 935 in which the stereoscopic image 933 is virtually formed through the second virtual lens array 934 by arithmetic processing.

  As shown in FIG. 21, the stereoscopic image display device 940 has the same configuration as the stereoscopic image display device 920 in FIG. 19, and the element image 941 (the image depth conversion device 930 in FIG. 20 generates via the display element 945 ( That is, the element image 935) of FIG. 20 is displayed. As a result, when the stereoscopic image 943 corresponding to the element image 941 is viewed from the observation direction 944, the cylinder is in front of the prism and the depth is equivalent to the subject 911 in FIG.

JP 2007-114483 A

  A stereoscopic image displayed by the conventional IP method corresponds to an optical aerial image, unlike a stereoscopic image obtained by a display using only binocular parallax. Therefore, in the conventional IP system, the state in which the observer sees the stereoscopic image has the same visibility as the state in which the real object is viewed.

  Here, when looking at the actual object, if the object is placed too close to the eye, or if the place where the object is viewed frequently changes in the depth direction, eye strain may occur. That is, when viewing a stereoscopic image displayed by the conventional IP method, eye strain may occur as in the case of viewing the actual subject.

  Therefore, an object of the present invention is to provide a stereoscopic image correction apparatus and a program thereof that can correct the depth of a stereoscopic image within an appropriate range for an observer.

  In view of the above problems, the stereoscopic image correction apparatus according to the first invention of the present application includes a depth range of a stereoscopic image when an elemental image group captured by the stereoscopic image capturing apparatus by the IP method is displayed on the stereoscopic image display apparatus, and A stereoscopic image correction apparatus that corrects a depth time change that is a change amount of a depth range, and includes a delay unit, a depth range calculation unit, a depth range determination unit, a depth time change calculation unit, and a depth time change determination unit. And a planar light intensity distribution calculating means, a volume light intensity distribution storing means, a depth range correcting means, and a correction element image group generating means.

  According to this configuration, the stereoscopic image correction apparatus receives the element image group by the delay unit, delays the input element image group by a preset delay time, and outputs the delay element image group.

In addition, the stereoscopic image correction apparatus uses a depth range calculation unit to calculate a display depth range and a delayed display depth range when the element image group and the delay element image group are each displayed on the stereoscopic image display device by a predetermined depth range calculation method. Calculated by
This display depth range indicates a range in which a stereoscopic image is displayed in the depth direction when reproducing a stereoscopic video, that is, a depth distance of the stereoscopic image.

In the stereoscopic image correction apparatus, the depth range determination unit determines whether the display depth range exceeds a preset reference depth range (for example, a depth range appropriate for an observer). This reference depth range may be a depth range unique to each observer, or may be a depth range common to all observers.
Here, when the display depth range exceeds the reference depth range and there is a possibility of causing eye strain when reproducing a stereoscopic image, the element image group is corrected by the following procedure.

  First, when the display depth range exceeds the reference depth range by the planar light intensity distribution calculation unit, the stereoscopic image correction device is a virtual element having the same pitch and focal length of the element optical system of the stereoscopic image display device with respect to the element image group. By performing wave optical calculation such as Fresnel approximation through the virtual element optical system group in which the optical systems are arranged two-dimensionally, the light intensity for each preset distance plane having a different distance from the virtual element optical system group A planar light intensity distribution as a distribution is calculated from the element image group. Then, the stereoscopic image correction apparatus uses the planar light intensity distribution calculating means to associate the calculated light intensity distribution with the distance, and to store the volume light intensity distribution storage means as the volume light intensity distribution on the stereoscopic image space for displaying the stereoscopic image. Remember me.

  In this way, the same elemental optical system as that of the stereoscopic image display device is virtually arranged in the space, and wave optical calculation is performed on the elemental image group. A virtual image space in which the apparatus displays a stereoscopic image is virtually shown.

  Next, when the display depth range exceeds the reference depth range by the depth range correction unit, the stereoscopic image correction device converts the volume light intensity distribution coordinates in the element image group into the depth by the ratio of the display depth range to the reference depth range. Correct in the direction. As described above, the corrected volumetric light intensity distribution is obtained by compressing the stereoscopic image space displaying the stereoscopic image in the depth direction.

  Finally, the stereoscopic image correction device performs wave optical calculation via the virtual element optical system group for each distance plane of the volume light intensity distribution corrected by the depth range correction unit by the correction element image group generation unit. Thus, a corrected element image group is generated. As a result, the corrected elemental image group does not display a stereoscopic image at a position that is too close to the eyes.

Subsequently, the stereoscopic image correction apparatus calculates a display depth time change which is a change amount between the display depth range and the delayed display depth range by the depth time change calculation unit.
The display depth time change indicates how much the position of the stereoscopic image changes in the depth direction when the stereoscopic video is reproduced.

Further, the stereoscopic image correction apparatus determines whether or not the display depth time change exceeds a preset reference depth time change (for example, a depth time change appropriate for the observer) by the depth time change determination unit.
Here, when the display depth time change exceeds the reference depth time change and there is a possibility of causing eye strain at the time of reproduction of the stereoscopic video, the delay element image group is corrected by the following procedure.

  First, when the display depth time change exceeds the reference depth time change, the stereoscopic image correction device performs wave optical calculation on the delay element image group via the virtual element optical system group by the planar light intensity distribution calculation unit. Thus, the planar light intensity distribution is calculated from the delay element image group, and is associated with the distance and stored in the volume light intensity distribution storage means as the volume light intensity distribution on the stereoscopic image space for displaying the stereoscopic image.

  Next, when the display depth time change exceeds the reference depth time change by the depth range correction unit, the stereoscopic image correction device uses the difference between the display depth time change and the reference depth time change to determine the volume light intensity in the delay element image group. Correct the distribution coordinates in the depth direction. Finally, the stereoscopic image correction device performs wave optical calculation via the virtual element optical system group for each distance plane of the volume light intensity distribution corrected by the depth range correction unit by the correction element image group generation unit. Thus, a corrected delay element image group is generated. Thus, the corrected delay element image group does not frequently change the position of the stereoscopic image in the depth direction.

In the stereoscopic image correction device according to the first invention of the present application, only the depth range may be corrected (second invention of the present application).
In the three-dimensional image correction apparatus according to the first invention of the present application, only the depth time change may be corrected (the third invention of the present application).

  Further, in the stereoscopic image correction apparatus according to the fourth invention of the present application, the depth range determination means sets in advance a unique individual depth range for each observer as the reference depth range, and whether the display depth range exceeds the individual depth range. And the depth range calculation means further calculates the depth range when the corrected elemental image group is displayed on the stereoscopic image display device by the depth range calculation method, and the common depth range common to the observer Is further provided with depth range warning means for determining whether or not the depth range calculated from the corrected elemental image group exceeds the common depth range, and warning if the depth range exceeds the common depth range It is characterized by that.

  According to such a configuration, a stereoscopic image can be reproduced within the optimum depth range for each observer, and an observer can be warned if there is a possibility of causing eye strain when reproducing the stereoscopic image.

  Further, in the stereoscopic image correction apparatus according to the fifth invention of the present application, the depth time change determining means presets an individual depth time change specific to each observer as the reference depth time change, and the display depth time change is the individual depth time change. The depth range calculation means further determines the depth range when the corrected delay element image group is displayed on the stereoscopic image display device by the depth range calculation method, and the depth time change The calculation means further calculates a depth time change that is a change amount between the display depth range and the corrected delay element image group, and a common depth time change common to the observer is preset and corrected. Determine whether the depth time change calculated from the later delay element image group exceeds the common depth time change, and warn if the depth time change exceeds the common depth time change That depth time change warning means, and further comprising a.

  According to such a configuration, a stereoscopic image can be reproduced within an optimum depth time change for an individual observer, and an observer can be warned when there is a possibility of causing eye strain when reproducing the stereoscopic image. .

  Here, the stereoscopic image correction apparatus according to the first invention of the present application is for causing a computer having hardware resources such as a CPU (Central Processing Unit) and storage means (for example, a memory and a hard disk) to operate cooperatively as the above-described means. It can also be realized by the three-dimensional image correction program (the sixth aspect of the present invention). This program may be distributed through a communication line, or may be distributed by writing in a recording medium such as a CD-ROM or a flash memory.

The present invention has the following excellent effects.
According to the first and sixth inventions of the present application, the stereoscopic image correction apparatus corrects the depth (depth range and depth time change) of the stereoscopic image within an appropriate range for the observer. Is not displayed, and the position of the stereoscopic image does not frequently change in the depth direction, and eye strain can be suppressed.

According to the second invention of the present application, since the stereoscopic image correction apparatus corrects the depth (depth range) of the stereoscopic image within an appropriate range for the observer, the stereoscopic image may be displayed at an excessively close position of the eyes. In addition, eye strain can be suppressed.
According to the third invention of the present application, since the stereoscopic image correction apparatus corrects the depth (change in depth time) of the stereoscopic image within an appropriate range for the observer, the position of the stereoscopic image may change frequently in the depth direction. In addition, eye strain can be suppressed.

According to the fourth aspect of the present invention, a stereoscopic image can be reproduced within an optimum depth range for each observer, and if there is a possibility of causing eye strain when reproducing this stereoscopic image, the observer is warned. it can.
According to the fifth invention of the present application, a stereoscopic image can be reproduced within an optimum depth time change for each observer, and warning is given to the observer when there is a possibility of causing eye strain when reproducing the stereoscopic image. Can do.

It is a block diagram which shows the structure of the stereo image correction apparatus which concerns on 1st Embodiment of this invention. It is a block diagram which shows the structure of the three-dimensional image processing means of FIG. It is explanatory drawing for demonstrating the production | generation of the light intensity distribution in the planar light intensity distribution calculation means of FIG. It is a block diagram which shows the structure of the planar light intensity distribution calculation means of FIG. It is explanatory drawing for demonstrating the light wave until it produces | generates the light intensity distribution in the planar light intensity distribution calculation means of FIG. It is explanatory drawing for demonstrating the planar light intensity distribution which the planar light intensity distribution calculation means of FIG. 2 produced | generated. It is explanatory drawing for demonstrating the volume light intensity distribution memorize | stored in the volume light intensity distribution memory | storage means of FIG. (A) And (b) is explanatory drawing for demonstrating correction | amendment of the depth range in the depth range correction | amendment means of FIG. It is explanatory drawing for demonstrating the concept which produces | generates the correction element image group in the correction element image group generation means of FIG. It is a block diagram which shows the structure of the correction | amendment element image group production | generation means of FIG. It is a block diagram which shows the structure of the stereo image correction apparatus which concerns on 2nd Embodiment of this invention. It is explanatory drawing for demonstrating correction | amendment of the depth time change in the three-dimensional image correction apparatus of FIG. It is explanatory drawing for demonstrating the concept which produces | generates the correction | amendment element image group in the stereo image correction apparatus of FIG. It is explanatory drawing for demonstrating the depth time change correct | amended with the stereo image correction apparatus of FIG. 11, (a) shows the depth time change before correction | amendment, (b) shows the depth time change after correction | amendment. 12 is a flowchart illustrating an operation of the stereoscopic image correction apparatus in FIG. 11. 12 is a flowchart illustrating an operation of the stereoscopic image correction apparatus in FIG. 11. It is a block diagram which shows the structure of the stereo image correction apparatus which concerns on 3rd Embodiment of this invention. It is a figure explaining imaging | photography of a to-be-photographed object in the conventional IP system. It is a figure explaining the display of a three-dimensional image in the conventional IP system. It is a figure explaining cancellation of depth inversion in the conventional IP system. It is a figure explaining the display of the element image group which the depth inversion canceled in the conventional IP system.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate. In each embodiment, means having the same function are denoted by the same reference numerals and description thereof is omitted.

(First embodiment)
[Configuration of stereoscopic image correction apparatus]
With reference to FIG. 1, the structure of the three-dimensional image correction apparatus 1 which concerns on 1st Embodiment of this invention is demonstrated.
The stereoscopic image correction apparatus 1 corrects the depth range of the element image group captured by the stereoscopic image capturing apparatus 100 when displaying a stereoscopic image by the IP method on the stereoscopic image display apparatus 200.

  The stereoscopic image correction apparatus 1 receives an element image group that is a video signal captured by the stereoscopic image capturing apparatus 100 using the IP method, and outputs a corrected element image group (correction element image group). The correction element image group is used as an element image group when displaying a stereoscopic image in the stereoscopic image display apparatus 200 using the IP method.

The stereoscopic image capturing apparatus 100 is an image capturing apparatus using a general IP system that captures a subject as an element image group, and is similar to the stereoscopic image capturing apparatus 910 shown in FIG. 18, for example. The stereoscopic image display device 200 is a general display device that displays a group of element images as a stereoscopic image, and is similar to the stereoscopic image display device 920 shown in FIG. 19, for example.
The element image group input to the stereoscopic image correction apparatus 1 does not have to be a video signal obtained by capturing an actual subject, and the stereoscopic image capturing apparatus 100 is schematically illustrated by a computer (not shown) such as computer graphics. It may be generated by reproducing.

  In addition, the stereoscopic image correction apparatus 1 generates standard depth range information, reference depth range information, and element image group generation as information necessary for correcting the element image group together with the element image group video signal from the stereoscopic image capturing apparatus 100. Information and standard display information are input. However, the standard display device information may be set in the stereoscopic image correction device 1 in advance.

The standard depth range information is information indicating a preset depth range (standard depth range) when a stereoscopic image is displayed on a standard display device (not shown).
The reference depth range information is information indicating a preset depth range (reference depth range) appropriate for the observer. In the present embodiment, the reference depth range is a depth range (common depth range) common to all observers.

The element image group generation information is information indicating the characteristics and arrangement of an element optical system (hereinafter referred to as an element lens) that is an optical element included in the stereoscopic image capturing apparatus 100.
Specifically, the element image group generation information includes the distance from the element lens group of the stereoscopic image capturing device 100 to the imaging surface of the imaging plate (focal length of the element lens), and the center distance between the element lenses constituting the element lens group ( Pitch), the size of the elemental image of the subject imaged by the imaging plate, and the arrangement position of the element lens.

In the stereoscopic image correction apparatus 1, display device information is set in advance as information related to the stereoscopic image display device 200.
This display device information is information indicating the characteristics and arrangement of the element lenses included in the stereoscopic image display device 200.
Specifically, the display device information includes the distance from the element lens group of the stereoscopic image display device 200 to the display surface of the display element (focal length of the element lens), and the center interval (pitch) of the element lenses constituting the element lens group. And the size of the elemental image of the subject displayed on the display element and the arrangement position of the element lens.

The standard display device information is information indicating the characteristics and arrangement of the element lenses included in the standard display device.
Specifically, the standard display device information includes the distance from the element lens group of the standard display device to the display surface of the display element (focal length of the element lens), and the center interval (pitch) of the element lenses constituting the element lens group. ), The size of the element image of the subject displayed on the display element, and the arrangement position of the element lens.

  As illustrated in FIG. 1, the stereoscopic image correction device 1 is configured to generate an element image group based on standard depth range information, reference depth range information, element image group generation information, display device information, and standard display device information. Display state calculation means 10 and stereoscopic image processing means 20 are provided.

  The display state calculation unit 10 calculates a state (depth range) in which the element image group is displayed in the stereoscopic image display device 200, and includes a depth range calculation unit 11 and a depth range determination unit 13.

  The depth range calculation unit 11 calculates a display depth range when the element image group is displayed on the stereoscopic image display device 200 by a predetermined depth range calculation method. Specifically, the depth range calculation unit 11 calculates a display depth range using the following formulas (1) and (2), and generates display depth range information indicating the display depth range.

Specifically, the depth range calculation unit 11, of the standard depth range, a point closest to the observer depth position of (nearest point) by substituting Z _c of the formula (1), in the three-dimensional image display device 200 The closest point of the generated stereoscopic image is calculated. Further, the depth range calculation unit 11, of the standard depth range, the point far from the nearest to the observer the depth position of the (farthest point) by substituting Z _c of the formula (1), is generated by the stereoscopic image display apparatus 200 The farthest point of the stereoscopic image is calculated. Then, the depth range calculation unit 11 calculates a range from the nearest point to the farthest point of the stereoscopic image generated by the stereoscopic image display apparatus 200 as the display depth range.

Here, the distance from the element lens of the stereoscopic image display apparatus 200 to the stereoscopic image is Z _r. The distance from the lens when the stereoscopic image is displayed by the standard display until the stereoscopic image is Z _c.

The distance from the display surface of the display device in a standard display until the element lens is d _c, the pitch P _c of the element lens group, the size of the element image displayed by the display device K _c It is. These d _c , P _c , and k _c can be acquired from the standard display device information.

In the stereoscopic image display apparatus 200, the distance from the display surface of the display element to the element lens group is _dr , the pitch of the element lens group is _Pr , and the size of the element image displayed by the display element is K. _r . These d _r , P _r , and k _r can be acquired from the display device information.

Since the depth range calculation method described above is described in the following references, detailed description thereof will be omitted.
References: J.Arai, M.Okui, M.Kobayashi, and F.Okano: "Geometrical effects of positional errors in integral photography", J. Opt. Soc. Am. A, Vol. 21, pp.951-958 , 2004

The depth range determination unit 13 determines whether or not the display depth range calculated by the depth range calculation unit 11 exceeds the reference depth range.
When the display depth range exceeds the reference depth range, the depth range determination unit 13 generates a command signal for correcting the element image group.
On the other hand, when the display depth range does not exceed the reference depth range, the depth range determination unit 13 generates a command signal indicating that the element image group is not corrected.

  The command signal generated by the depth range determination unit 13 is output to the stereoscopic image processing unit 20 together with the display depth range information, the standard depth range information, the reference depth range information, the element image group generation information, and the display device information. Is done.

The stereoscopic image processing means 20 performs processing (depth range correction) on the video signal of the element image group based on the command signal.
When a command signal for correcting the element image group is input, the stereoscopic image processing unit 20 corrects the element image group.
On the other hand, when a command signal indicating that the element image group is not corrected is input, the stereoscopic image processing unit 20 outputs the same to the stereoscopic image display apparatus 200 without correcting the depth range.

  As shown in FIG. 2, the stereoscopic image processing unit 20 includes a planar light intensity distribution calculating unit 30, a volume light intensity distribution storing unit 40, a depth range correcting unit 50, and a correction element image group generating unit 60.

  The planar light intensity distribution calculating means 30 is a virtual element lens (virtual element optical system) having the same pitch and focal length of the element lens of the stereoscopic image display device 200 indicated by the display device information from the element image group which is an input video signal. ) Are virtually arranged in a two-dimensional manner, by performing wave optical calculation via a virtual element lens group (virtual element optical system group), the distance from the virtual element lens group is different for each preset distance plane. A plane light intensity distribution, which is a light intensity distribution, is calculated.

The planar light intensity distribution calculating unit 30 calculates a planar light intensity distribution for each different distance plane and writes it in the volume light intensity distribution storage unit 40 in association with the distance.
That is, the planar light intensity distribution calculating unit 30 generates light intensity distributions that differ depending on a plurality of distance planes from the element image group constituted by one plane, and thereby the light intensity when the element image group is displayed as a stereoscopic image. A space (stereoscopic image space) corresponding to the distribution is reproduced.
The plane light intensity distribution calculating unit 30 calculates the light intensity distribution in a plurality of predetermined distance planes, writes the volume light intensity distribution in the volume light intensity distribution storage unit 40, and then generates the volume light intensity distribution. This is notified to the depth range correction means 50.

Here, with reference to FIG. 3, the light intensity distribution calculated by the planar light intensity distribution calculating unit 30 for each distance plane will be described.
As shown in FIG. 3, the planar light intensity distribution calculating unit 30 is virtual with respect to the image surface of the element image group G by the same distance as the distance _dr between the element lens group and the display surface in the stereoscopic image display device 200. through the virtual element lens unit V which is disposed spaced, by performing wave optics calculation, arbitrary distance _{L 1,} L 2, ..., the light intensity of each _{L n} distribution _{t 1, t 2, ...,} t _{Find n} .

Here, the distances L ₁ , L ₂ ,..., L _n are a plurality of predetermined distances, and the number and interval can be arbitrarily determined. If a large number of distance planes are set, the effect of correction can be increased accordingly, but the amount of calculation increases. Therefore, the number and interval of the distance planes are preferably determined according to the number of pixels of the element image group G and the like.

As shown in FIG. 3, the light wave image area k _r of an element image corresponding virtual element lens (virtual opening [virtual pinhole]) via the V _L reaches the distance plane with the spread To do. For example, a range w ₁ of the image region k _r of the element image is enlarged at a distance plane distance L ₁ as the light wave is expressed by the following equation (3).

As for the range w _n of the image area k _r of the element image that reaches the distance L _n farthest from the virtual element lens group V, a can be obtained as for formula (3).
That is, the light intensity distribution (volume light intensity distribution) calculated by the planar light intensity distribution calculating unit 30 is determined from the distance L ₁ closest to the virtual element lens group V in the z direction which is the depth direction, from the virtual element lens group V. In the range of the distance L _n farthest from the virtual element lens group V, the range in which the light wave of the element image group G spreads at the distance L _n farthest from the virtual element lens group V, that is, above and below the size of the element image group G It is generated in a range larger by w _n / 2 on the left and right.

Here, the configuration of the planar light intensity distribution calculating means 30 will be described with reference to FIG.
In order to generate the light intensity distribution (volume light intensity distribution) described with reference to FIG. 3, the planar light intensity distribution calculating unit 30 includes a dividing unit 31, an element image converting unit 33, and a combining unit 35.

  The dividing unit 31 divides an element image group input as a video signal into light waves in units of element images based on display device information. That is, the dividing unit 31 divides the element image group for each element image corresponding to the arrangement position of the element lens in the display device information, and sets a pixel value as an amplitude in each pixel of the element image, and a predetermined phase. Is added to make the pixel value of the element image light wave information (hereinafter simply referred to as a light wave) represented by a complex number of amplitude and phase. However, it is not always necessary to add a phase.

Note that the dividing unit 31 divides the element image group by the size of the element image handled by the stereoscopic image display apparatus 200. When the stereoscopic image capturing apparatus 100 and the stereoscopic image display apparatus 200 have different element image sizes (size, resolution), the dividing unit 31 matches the element image size of the stereoscopic image display apparatus 200. Image conversion processing such as down-conversion and up-conversion is performed.
The dividing unit 31 outputs a light wave for each divided element image to the element image converting unit 33.

  The element image converting means 33 converts the light wave for each element image input from the dividing means 31 into a light wave on a preset distance plane by performing wave optical calculation. Here, the element image conversion means 33 includes a light wave calculation means 33a, a phase shift means 33b, and a light wave calculation means 33c.

The light wave calculating means (first light wave calculating means) 33a calculates the light wave propagating to the virtual element lens by approximating the light wave of the element image by Fresnel approximation. That is, as shown in FIG. 5, the light wave calculating unit 33 a determines the position corresponding to the m th element image from the light wave (gs _{, m} (x _{s, m} , y _{s, m} )) of the m th element image. The light wave (R _{i, m} (x _{o, m} , y _{o, m} )) that reaches the virtual element lens V _m virtually arranged at is calculated using the following formula (4) using Fresnel approximation.

Here, x _{s, m} , x _{o, m} are the x coordinate having the origin at the center of the mth element image in the entire image (element image group G), and the mth virtual element lens of the virtual element lens group V, respectively. the optical axis center of the V _m is the x-coordinate of the origin. Y _{s, m} , y _{o, m} are the y coordinate with the center of the m-th element image in the entire image (element image group G) as the origin, and the m-th virtual element lens V of the virtual element lens group V, respectively. _This is the y coordinate with the center of the optical axis of _m as the origin. Further, f is the focal length of the virtual element lenses V _m (in FIG. 5, corresponds to a distance d _c), k is the wave number 2π / λ (λ is a wavelength of a predetermined light). In the formula (4), the element image to be integral range (m) is an image region k _r of the element image m shown in FIG.
The light wave calculating means 33a outputs the light wave (R _{i, m} (x _{o, m} , _{yo, m} )) for each element image to the phase shift means 33b.

The phase shift means (first phase shift means) 33b converts the phase of the light wave (R _{i, m} (x _{o, m} , _{yo, m} )) of the element image input from the light wave calculation means 33a to the virtual element lens. It is shifted by the phase. That is, the phase shifting means 33b, as shown in FIG. 5, the light waves propagating in the virtual element lenses _{V m (R i, m (} x o, m, y o, m)) , and changes in the virtual element lenses _{V m} The light wave (R _{o, m} (x _{o, m} , y _{o, m} )) emitted from the virtual element lens V _m is calculated by shifting the phase by the following equation (5).

The phase shift means 33b outputs the light wave (R _{o, m} (x _{o, m} , _{yo, m} )) for each element image to the light wave calculation means 33c.

The light wave calculating means (second light wave calculating means) 33c performs Fresnel approximation on the light wave (R _{o, m} (x _{o, m} , y _{o, m} )) for each element image input from the phase shift means 33b. The light wave in the preset distance plane is calculated. That is, the light wave calculating means 33c, as shown in FIG. 5, the light waves emitted from the virtual element lenses _{V m (R o, m (} x o, m, y o, m)) from the virtual element lenses _{V m} A light wave (R _{d, m} (x _{d, m} , y _{d, m} )) that reaches a predetermined distance plane T separated by a distance L is calculated using the Fresnel approximation according to the following equation (6).

Here, x _{d, m} is the x-coordinate of the distance plane T that the center of the optical axis of the m-th virtual element lenses V _m of the virtual element lens unit V and the origin. Y _{d, m} is the y coordinate of the distance plane T with the origin at the optical axis center of the m-th virtual element lens V _m of the virtual element lens group V. L is the distance from the virtual element lens group V to the distance plane T. In the formula (6), the range w to spread the element image (m) as the integration range is a range w of the image region k _r of the element image shown in FIG. 5 is enlarged at a distance plane T as light waves. This range w is obtained by the same calculation as the above-described equation (3).
The light wave calculating unit 33 c outputs a light wave (R _{d, m} (x _{d, m} , y _{d, m} )) for each element image reaching the distance plane T to the combining unit 35.

The combining unit 35 combines the light waves of the element images on the distance plane converted by the element image converting unit 33 by the element image of the element image group on the distance plane to generate a plane light intensity distribution. That is, as shown in FIG. 5, the combining unit 35 transmits light waves (R _{d, m} (x _{d, m) for} each element image reaching the distance plane T set to a predetermined distance L from the virtual element lens group V. , Y _{d, m} )) are added within the region of the distance plane T.
More specifically, the combining unit 35 converts the light wave (R _{d, m} (x _{d, m} , y _{d, m} )) for each element image into the number of element images (here, −M ≦ m ≦ M). (2M + 1)) are added according to the following equation (7).

Here, x _{p, m} is an x coordinate with the origin of the center of a predetermined distance plane T. Further, yp _{, m} is a y coordinate with the center of a predetermined distance plane T as the origin.
Then, the coupling means 35 squares the light wave reaching the predetermined distance plane T calculated by the expression (7) by the following expression (8), so that the light intensity distribution (planar light intensity in the distance plane T) is obtained. Distribution) t (x _{p, m} , y _{p, m} ) is calculated.

Here, the planar light intensity distribution t will be described with reference to FIG.
In FIG. 6, the x-axis is the horizontal direction, the y-axis is the vertical direction, and the z-axis is the depth direction. Consider a case where two objects, a cylindrical body and a triangular pyramid, are taken as subjects, and the bottom surfaces of these cylindrical bodies and the triangular pyramid are photographed toward the z axis. In this case, in the planar light intensity distribution t, the light intensity distribution range in which the light intensity exceeds zero represents a three-dimensional image of the cylindrical body and a three-dimensional image of the triangular pyramid viewed from the z axis.

Then, the combining unit 35 writes the calculated planar light intensity distribution in the volume light intensity distribution storage unit 40 (see FIG. 2) in association with the distance plane.
The planar light intensity distribution calculating means 30 described above sequentially resets the distance from the virtual element lens group, and calculates the planar light intensity distribution corresponding to a plurality of distance planes as described with reference to FIG. That is, the plane light intensity distribution calculating means 30, as shown in FIG. 3, the distance from the virtual element lens group _{V L 1, L 2, ...} , the light intensity distribution _{t 1, t} 2 corresponding to _{L n,} ..., to calculate the t _n.
The light intensity distributions t ₁ , t ₂ ,..., T _n generated by the planar light intensity distribution calculating means 30 are arranged for each distance as shown in FIG. Can be expressed as

Thereby, the volume light intensity distribution storage means 40 stores a volume light intensity distribution composed of light intensity distributions in a plurality of distance planes. This volumetric light intensity distribution corresponds to a state in which the light intensity is distributed in the stereoscopic image space when the element image group is displayed on the stereoscopic image display device.
Returning to FIG. 2, the description of the configuration of the stereoscopic image processing means 20 will be continued.

  The volume light intensity distribution storage means 40 stores the light intensity distribution for each distance plane calculated by the plane light intensity distribution calculation means 30 as a volume light intensity distribution in association with the distance. Storage device. This volume light intensity distribution is corrected by the depth range correction unit 50 and is referred to by the correction element image group generation unit 60.

  The depth range correction unit 50 corrects the coordinates of the volume light intensity distribution stored in the volume light intensity distribution storage unit 40 in the depth direction using the display depth range information and the reference depth range information.

This volume light intensity distribution is a set of light intensities assigned to each coordinate in the stereoscopic image space before correction. Therefore, in order to calculate the volume light intensity distribution (corrected volume light intensity distribution) in which the depth range is corrected, as shown in FIG. 8A, coordinates before correction of the stereoscopic image space in which the stereoscopic image is generated ( X _r , Y _r , Z ′ _r ) are assigned to corrected coordinates (X _c , Y _c , Z ′ _c ) of the stereoscopic image space in which the stereoscopic image is generated, as shown in FIG. That's fine.
8A illustrates the coordinate axes of the stereoscopic image space before correction as x _r , y _r , and z _r, and FIG. 8B illustrates the coordinate axes of the stereoscopic image space after correction as x _c , y _c , and so on. Illustrated as _zc .

Here, the ratio between the display depth range β and the reference depth range α is Ω (where Ω> 1). In this case, the relationship between the coordinates before correction (X _r , Y _r , Z ′ _r ) and the coordinates after correction (X _c , Y _c , Z ′ _c ) is expressed by the following equations (9) to (11). ). That is, the depth range correction unit 50 compresses the volume light intensity distribution in the depth direction with the ratio Ω between the display depth range β and the reference depth range α, using Expressions (9) to (11).

  Here, the depth range correction unit 50 writes the corrected volume light intensity distribution obtained by correcting the coordinate position of the volume light intensity distribution in the volume light intensity distribution storage unit 40, and notifies the correction element image group generation unit 60 that the correction is completed. Notice.

  The correction element image group generation unit 60 uses the virtual volume of the corrected volume light intensity distribution corrected by the depth range correction unit 50 as the virtual element lens (the same pitch and focal length of the element lens of the stereoscopic image display device 200 indicated by the display device information). An element image group at a position corresponding to the display surface of the stereoscopic image display device 200 is generated by performing wave optical calculation via a virtual element lens group (virtual element optical system group) configured by a virtual element optical system) To do.

  That is, the correction element image group generation means 60 corresponds to the light intensity data (for example, data of the point a) in the volume light intensity distribution after correction via the virtual element lens group V as shown in FIG. The data is assigned as point data on the image (for example, data of the point a ′). In FIG. 9, only data at a certain point is assigned on the image, but assignment is performed over the entire volumetric light intensity distribution after correction. As a result, a corrected element image group (corrected element image group G ′) is generated.

  Here, the configuration of the correction element image group generation means 60 will be described with reference to FIG. 10 (refer to FIG. 2 as appropriate). As shown in FIG. 10, the correction element image group generation unit 60 includes an element image reverse conversion unit 61 and a connection unit 63.

The element image reverse conversion means 61 generates an element image corresponding to the arrangement position of the element lens of the stereoscopic image display device 200 indicated by the display device information from the corrected volume light intensity distribution.
That is, the element image inverting unit 61, the light intensity distribution of distances for each plane having spread through the virtual element lens V _L corresponding to the image area k _r of the element images described in FIG. 5, the elemental image group by inverse transform on the image plane that constitutes, generates an element image corresponding to the image area k _r. In the plane light intensity distribution calculation means 30, the distances L ₁ , L ₂ ,..., L _n are predetermined, but the element image inverse conversion means 61 uses the corrected volume light intensity distribution as follows: The plane (xy plane) at the distance (z coordinate) to which the light intensity is assigned becomes the target distance plane for performing the inverse transformation.
Here, the element image reverse conversion unit 61 includes a light wave calculation unit 61a, a phase shift unit 61b, a light wave calculation unit 61c, and an addition unit 61d.

The light wave calculation means (third light wave calculation means) 61a propagates the light wave having the light intensity distribution to the virtual element lens group by Fresnel approximation for each distance (distance plane) indicated by the corrected volume light intensity distribution. The light wave is calculated.
In other words, the light wave calculating means 61a, from the light wave (R _p (x _p , y _p )) corresponding to the element image in a certain distance plane, reaches the virtual element lens group separated by the distance L (R _o (x _o , y _o )) is calculated using the Fresnel approximation according to the following equation (12).

Here, x _p and x _o are x coordinates with the origin at the center of a predetermined distance plane. Further, y _p and _yo are y coordinates with the center of a predetermined distance plane as the origin. In Expression (12), the “distance plane region” serving as the integration range is the horizontal (x direction) of the plane light intensity distribution constituting the volume light intensity distribution stored in the volume light intensity distribution storage means 40. This is a vertical range (y direction).
Note that the light wave (R _p (x _p , y _p )) for the element image is obtained by multiplying the light intensity of the volume light intensity distribution stored in the volume light intensity distribution storage means 40 by a factor of 1/2 (see above). Inverse calculation of equation (8)).
The light wave calculation means 61a outputs the light wave (R _o (x _o , y _o )) reaching the virtual element lens group for each calculated distance plane to the phase shift means 61b.

The phase shift means (second phase shift means) 61b shifts the phase of the light wave (R _o (x _o , y _o )) for each distance plane input from the light wave calculation means 61a by the phase of the virtual element lens. Is. That is, the phase shifting means 61b is the light wave _{(R o (x o, y} o)) and only the phase component that varies in the virtual element lens, by shifting the equation (13) below, the light wave _(R i (x _o , y _o )) is calculated.

Here, f is the focal length of the virtual element lens.
The phase shift means 61b outputs a light wave (R _i (x _o , y _o )) for each distance plane to the light wave calculation means 61c.

The light wave calculating means (fourth light wave calculating means) 61c performs a virtual Fresnel approximation for each virtual element lens with the light wave (R _i (x _o , y _o )) for each distance plane input from the phase shift means 61b. The light wave for each element image corresponding to the distance plane reaching the focal length of the element lens is calculated. In other words, the light wave calculation means 61c is configured to display the image plane (rear side) of the element image group G separated from the light wave (R _i (x _o , y _o )) propagated in the opposite direction from the virtual element lens by the focal length of the virtual element lens. The light wave (g ′ _{s, m} (x _{s, m} , y _{s, m} )) of the element image m for each distance plane reaching the side focal plane) is calculated using the Fresnel approximation by the following equation (14). To do.

Here, x _{s, m} , x _{i, m} are respectively the x coordinate with the center of the mth element image in the entire image (element image group) as the origin, and the light of the mth virtual element lens of the virtual element lens group. This is the x coordinate with the axis center as the origin. Y _{s, m} , y _{i, m} are respectively the y coordinate with the origin of the center of the mth element image in the entire image (element image group), and the optical axis of the mth virtual element lens of the virtual element lens group The y coordinate with the center as the origin. F is the focal length of the virtual element lens, and k is the wave number 2π / λ (λ is the wavelength). In the formula (14), the integral range "element image (m)" is an image region k _r of the element image m shown in FIG.
The light wave calculating means 61c outputs the light wave (g ′ _{s, m} (x _{s, m} , y _{s, m} )) of the element image corresponding to the calculated distance plane to the adding means 61d.

The adding means 61d calculates the light wave of the corrected element image by adding the light waves of the element image for each virtual element lens input from the light wave calculating means 61c by the distance plane. That is, the adding means 61d uses the following equation (15) to calculate the light wave (g ′ _{s, m} (x _{s, m} , y _{s, m} ) of the element image for each distance plane calculated as the light wave from each distance plane. ) Are added for the distance plane, and the light wave (gs _{, m} ( _{xs, m} , ys _{, m} )) of the element image is calculated.

Here, the “distance plane” serving as the addition range is the number of distance planes of the volume light intensity distribution corrected by the depth range correction unit 50.
The adding means 61 d outputs the light wave (gs _{, m} (x _{s, m} , y _{s, m} )) of the element image to the connecting means 63.

The connecting unit 63 connects the light waves of the individual element images generated by the element image inverse converting unit 61 and constitutes an element image group.
The connecting means 63 generates a corrected element image group (corrected element image group) by connecting the intensity values for each pixel, which are light waves of individual element images, as pixel values and connecting the element images.

As described above, since the stereoscopic image correction apparatus 1 corrects the depth range of the element image group, a stereoscopic image is not displayed at an excessively close position of the eyes, and eye strain can be suppressed. .
The stereoscopic image correction apparatus 1 can be operated by a program (stereoscopic image correction program) that causes a general computer to function as each of the above-described means.

In addition, although the Fresnel approximation formula is used as the wave optical calculation for calculating the propagation of the light wave, the calculation may be performed using the integration based on the Huygens-Fresnel principle, the integration based on the Francofa approximation, or the like.
The element optical system group has been described as a lens array in which element lenses are arranged, but may be an aperture array (spatial filter) in which minute pinholes are arranged.
In addition, since the stereoscopic image correction apparatus 1 performs the operation of executing steps S2 and S4 to S11 described in the second embodiment (FIG. 15), description thereof is omitted.

(Second Embodiment)
[Configuration of stereoscopic image correction apparatus]
With reference to FIG. 11, a difference from the first embodiment will be described for the stereoscopic image correction apparatus 1 </ b> B according to the second embodiment of the present invention (see FIG. 2 as appropriate).
The stereoscopic image correction apparatus 1B is different from the first embodiment in that in addition to the depth range, the stereoscopic image correction apparatus 1B also corrects the depth time change. Therefore, the stereoscopic image correction apparatus 1B includes a display state calculation unit 10B, a stereoscopic image processing unit 20B, and a delay unit 70.

  The stereoscopic image correction apparatus 1B receives standard depth time change information and reference depth time change information as information necessary for correcting the delay element image group from the stereoscopic image capturing apparatus 100.

The standard depth time change information is information indicating a preset depth time change (standard depth time change) when a stereoscopic image is displayed on a standard display device.
The reference depth time change information is information indicating preset depth time change (reference depth time change) appropriate for the observer. In this embodiment, the reference depth time change information is assumed to be a depth time change common to all observers (common depth time change).

Here, the delay means 70 will be described first.
The delay means 70 receives an element image group from the stereoscopic image capturing apparatus 100, delays the input element image group by a delay time, and outputs a delay element image group. The delay means 70 is composed of, for example, a frame memory (not shown) that stores element image groups and a control means (not shown) that controls the frame memory.

In the following description, the time at which the element images is input from the three-dimensional image photographing apparatus 100 and T _1, the time after the delay processing in the delay means 70 and T _2. That is, the delay time Δ = T ₂ −T ₁ and the element image group delayed by the delay time Δ is referred to as a delay element image group.
The delay time Δ is set in advance as an arbitrary value, for example, indicating a minute time.

  The display state calculation unit 10B calculates a state in which the element image group and the delay element image group are displayed in the stereoscopic image display apparatus 200, and includes a depth range calculation unit 11B, a depth range determination unit 13, and a depth time. A change calculating means 15 and a depth time change determining means 17 are provided.

  The depth range calculation unit 11B calculates a delay display depth range when the delay element image group is displayed on the stereoscopic image display device 200 in the same manner as the depth range calculation unit 11 of FIG. 1 in addition to the display depth range. is there.

The depth time change calculation unit 15 calculates a display depth time change which is a change amount between the display depth range and the delay display depth range calculated by the display state calculation unit 10B. That is, the depth time change calculation means 15 calculates the difference between the display depth range and the delay display depth range as the display depth time change.
This display depth time change is information indicating a change in depth in the delay time Δ (minute time).

The depth time change determination means 17 determines whether or not the display depth time change calculated by the depth time change calculation means 15 exceeds the reference depth time change.
When the display depth range exceeds the reference depth range, the depth time change determination unit 17 generates a command signal for correcting the delay element image group.
On the other hand, when the display depth range does not exceed the reference depth range, the depth time change determination unit 17 generates a command signal indicating that the delay element image group is not corrected.
The command signal generated by the depth time change determination unit 17 is output to the stereoscopic image processing unit 20B together with the standard depth time change information and the reference depth time change information.

As shown in FIG. 12, the display depth time change eta, it shows a three-dimensional image space in element image group of time T _1, the amount of change in the depth range of the three-dimensional image space in element image group of time T _2. In this example, the depth time change determination unit 17 generates a command signal for correcting the delay element image group because the display depth time change η exceeds the reference depth time change γ.
In FIG. 12, illustrating the three-dimensional image space in element image group of time T ₁ by a two-dot chain line. The description of the symbol ε in FIG. 12 will be given later.

Returning to FIG. 11, the description of the configuration of the stereoscopic image correction apparatus 1B will be continued.
The stereoscopic image processing means 20B has the same configuration as that of the stereoscopic image processing means 20 in FIG. 2, and corrects the delay element image group in addition to the element image group based on the command signal.
When a command signal for correcting the delay element image group is input, the stereoscopic image processing unit 20B corrects the delay element image group.
On the other hand, when a command signal indicating that the delay element image group is not corrected is input, the stereoscopic image processing unit 20B outputs the signal to the stereoscopic image display apparatus 200 as it is without correcting the delay element image group.

The planar light intensity distribution calculating means 30 (FIG. 2) calculates the planar light intensity distribution for each distance plane from the delayed element image group by the same method as the element image group, and stores the volume light intensity distribution in association with the distance. It is written in the means 40.
Thus, the volume light intensity distribution storage means 40 (FIG. 2) stores the volume light intensity distribution in the delayed element image group in addition to the volume light intensity distribution in the element image group.

  The depth range correction means 50 (FIG. 2) corrects the coordinates of the volume light intensity distribution in the delay element image group in the depth direction using the display depth time change information and the reference depth time change information. That is, the depth range correction unit 50 corrects the volume light intensity distribution in the delay element image group with the difference between the display depth time change and the reference depth time change, similarly to the volume light intensity distribution in the element image group.

  The correction element image group generation means 60 (FIG. 2) generates a correction delay element image group from the corrected volume light intensity distribution in the delay element image group by the same method as the correction element image group.

Here, the correction element image group generation unit 60 presets the distance from the corrected volume light intensity in the delay element image group to the virtual element lens group based on the display device information. This position of the virtual element lens group is referred to as “reference optical element array position”. Then, as shown in FIG. 13, the correction element image group generation means 60 corrects the delay element image group with the virtual element lens group V disposed at a position shifted from the reference optical element array position S by the distance correction value ε. G ″ is generated.
This distance correction value ε indicates the difference between the display depth time change η and the reference depth time change γ, as shown in FIG.

Accordingly, the correction element image group generation unit 60 generates the correction delay element image group so that the display depth time change η in FIG. 14A does not exceed the reference depth time change γ in FIG. 14B. be able to.
In FIG. 14, the stereoscopic image space in the element image group at time T ₁ is illustrated by a two-dot chain line, and the stereoscopic image space in the delay element image group at time T ₂ is illustrated by a solid line.

[Operation of stereoscopic image correction device]
The operation of the stereoscopic image correction apparatus 1B will be described with reference to FIGS.
The stereoscopic image correction apparatus 1B outputs a delayed element image group obtained by delaying the element image group by the delay time by the delay unit 70 (step S1).

In the stereoscopic image correction apparatus 1B, the depth range calculation unit 11B calculates the display depth range from the element image group, and calculates the delay display depth range from the delay element image group (step S2).
In the stereoscopic image correction apparatus 1B, the depth time change calculation unit 15 calculates a display depth time change that is a change amount between the display depth range and the delayed display depth range (step S3).

In the stereoscopic image correction apparatus 1B, the depth range determination unit 13 determines whether or not the display depth range exceeds the reference depth range (step S4).
When the display depth range exceeds the reference depth range (Yes in step S4), the stereoscopic image correction apparatus 1B proceeds to the process of step S5.
In the stereoscopic image correction device 1B, the plane light intensity distribution calculating unit 30 sets distance planes having different distances from the virtual element lens group for the element image group (step S5).

  The stereoscopic image correction apparatus 1B uses the planar light intensity distribution calculating unit 30 to cause the planar light intensity when the light wave of the element image group reaches the distance plane corresponding to the distance set in step S5 via the virtual element lens group. The distribution is calculated by wave optics calculation.

That is, the stereoscopic image correction apparatus 1B divides the element image group for each element image specified by the display device information by the dividing unit 31 of the planar light intensity distribution calculating unit 30.
In the stereoscopic image correction apparatus 1B, the light wave calculation unit 33a of the planar light intensity distribution calculation unit 30 performs a Fresnel approximation of the light wave of each element image, and calculates a light wave that reaches the virtual element lens.
In the stereoscopic image correction apparatus 1B, the phase shift unit 33b of the planar light intensity distribution calculation unit 30 shifts the phase of the light wave reaching the virtual element lens by the phase of the virtual element lens.
The stereoscopic image correction apparatus 1B calculates the light wave for each element image in the set distance plane by Fresnel approximation of the phase-shifted light wave by the light wave calculation unit 33c of the plane light intensity distribution calculation unit 30.
The stereoscopic image correction apparatus 1B combines the light waves of the element images that have reached the set distance plane by an amount corresponding to the element image by the combining unit 35 of the plane light intensity distribution calculating unit 30, and the light intensity distribution (plane light in the distance plane). (Intensity distribution) is calculated (step S6).

  The stereoscopic image correction apparatus 1B writes the planar light intensity distribution generated in step S6 into the volume light intensity distribution storage unit 40 by the planar light intensity distribution calculating unit 30 (combining unit 35) (step S7). This plane light intensity distribution is generated for each different distance plane, and thus becomes volume light intensity.

The stereoscopic image correction apparatus 1B uses the depth range correction unit 50 to display the volume light intensity distribution coordinates stored in the volume light intensity distribution storage unit 40 in the depth direction using the display depth range information and the reference depth range information. to correct. That is, the depth range correction unit 50 compresses the volume light intensity distribution in the element image group in the depth direction at a ratio between the display depth range and the reference depth range (step S8).
The stereoscopic image correction apparatus 1B sets a distance plane in which the corrected volume light intensity distribution exists by the correction element image group generation unit 60 (step S9).

  The stereoscopic image correction apparatus 1B reaches the image plane of the element image group using the volume light intensity distribution corrected in step S8 as a light wave by the wave optical calculation by the element image inverse conversion unit 61 of the correction element image group generation unit 60. The light wave of each element image to be calculated is calculated.

That is, the stereoscopic image correction apparatus 1B uses the light wave calculation unit 61a of the element image reverse conversion unit 61 to generate a light wave having a light intensity distribution by Fresnel approximation for each distance (distance plane) indicated by the volume light intensity distribution after correction. Calculates the light wave reaching the virtual element lens group.
The stereoscopic image correction apparatus 1B shifts the phase of the light wave reaching the virtual element lens by the phase of the virtual element lens by the phase shift means 61b of the element image inverse conversion means 61.
The stereoscopic image correction apparatus 1B calculates the light wave for each element image corresponding to the distance plane in the image plane of the element image group by Fresnel approximation of the light wave that has been phase-shifted by the light wave calculation means 61c of the element image inverse conversion means 61. (Step S10).

  The stereoscopic image correction apparatus 1B calculates the light wave of the element image after correction by adding the light wave of the element image corresponding to the distance plane by the addition unit 61d of the element image reverse conversion unit 61 by the distance plane, By connecting by the connecting means 63, a correction element image group is generated (step S11).

When the display depth range does not exceed the reference depth range (No in step S4), or after the process of step S11, the stereoscopic image correction apparatus 1B proceeds to the process of step S12.
In the stereoscopic image correction apparatus 1B, the depth time change determination unit 17 determines whether or not the display depth time change exceeds the reference depth time change (step S12).

When the display depth time change exceeds the reference depth time change (Yes in step S12), the stereoscopic image correction apparatus 1B proceeds to the process of step S13.
In the stereoscopic image correction apparatus 1B, the plane light intensity distribution calculating unit 30 sets distance planes having different distances from the virtual element lens group for the delay element image group (step S13).

  The stereoscopic image correction apparatus 1B uses the planar light intensity distribution calculating unit 30 to generate planar light when the light wave of the delay element image group reaches the distance plane corresponding to the distance set in step S13 via the virtual element lens group. The intensity distribution is calculated by wave optics calculation.

That is, the stereoscopic image correction apparatus 1B divides the delayed element image group for each element image specified by the display device information by the dividing unit 31 of the planar light intensity distribution calculating unit 30.
In the stereoscopic image correction apparatus 1B, the light wave calculation unit 33a of the planar light intensity distribution calculation unit 30 performs a Fresnel approximation of the light wave of each delay element image, and calculates a light wave that reaches the virtual element lens.
In the stereoscopic image correction apparatus 1B, the phase shift unit 33b of the planar light intensity distribution calculation unit 30 shifts the phase of the light wave reaching the virtual element lens by the phase of the virtual element lens.
The stereoscopic image correction apparatus 1B calculates the light wave for each element image in the set distance plane by Fresnel approximation of the phase-shifted light wave by the light wave calculation unit 33c of the plane light intensity distribution calculation unit 30.
The stereoscopic image correction apparatus 1B combines the light waves of the element images that have reached the set distance plane by an amount corresponding to the element image by the combining unit 35 of the plane light intensity distribution calculating unit 30, and the light intensity distribution (plane light in the distance plane). (Intensity distribution) is calculated (step S14).

  The stereoscopic image correction apparatus 1B writes the planar light intensity distribution generated in step S14 to the volume light intensity distribution storage unit 40 by the planar light intensity distribution calculating unit 30 (combining unit 35) (step S15).

The stereoscopic image correction apparatus 1B uses the depth range correction unit 50 to display the coordinates of the volume light intensity distribution stored in the volume light intensity distribution storage unit 40 using the display depth time change information and the reference depth time change information. Correct in the direction. That is, the depth range correction unit 50 compresses the volume light intensity distribution in the delay element image group in the depth direction by the difference between the display depth time change and the reference depth time change (step S16).
The stereoscopic image correction apparatus 1B sets a distance plane in which the corrected volume light intensity distribution exists by the correction element image group generation unit 60 (step S17).

  The stereoscopic image correction apparatus 1B uses the volume image intensity distribution corrected in step S15 by the wave optical calculation by the element image inverse conversion unit 61 of the correction element image group generation unit 60 as a light wave on the image surface of the delay element image group. The light waves of the individual element images that arrive are calculated.

That is, the stereoscopic image correction apparatus 1B uses the light wave calculation unit 61a of the element image reverse conversion unit 61 to generate a light wave having a light intensity distribution by Fresnel approximation for each distance (distance plane) indicated by the volume light intensity distribution after correction. Calculates the light wave reaching the virtual element lens group.
The stereoscopic image correction apparatus 1B shifts the phase of the light wave reaching the virtual element lens by the phase of the virtual element lens by the phase shift means 61b of the element image inverse conversion means 61.
The stereoscopic image correction apparatus 1B calculates the light wave for each element image corresponding to the distance plane in the image plane of the delay element image group by performing Fresnel approximation of the phase-shifted light wave by the light wave calculation means 61c of the element image inverse conversion means 61. (Step S18).

The stereoscopic image correction apparatus 1B calculates the light wave of the element image after correction by adding the light wave of the element image corresponding to the distance plane by the addition unit 61d of the element image reverse conversion unit 61 by the distance plane, By connecting by the connecting means 63, a corrected delay element image group is generated (step S19).
When the display depth time change does not exceed the reference depth time change (No in step S12), or after step S19, the stereoscopic image correction apparatus 1B ends the process.

  As described above, since the stereoscopic image correction apparatus 1B corrects the depth range of the element image group and the depth time change of the delay element image group, the stereoscopic image is displayed at an excessively close position of the eye, Since the position of the stereoscopic image does not change frequently in the depth direction, it is possible to suppress eye strain.

  In the second embodiment, the stereoscopic image correction apparatus 1B has been described as correcting both the depth range and the depth time change. However, only the depth time change may be corrected. In this case, the stereoscopic image correction apparatus 1B does not need to include the depth range determination unit 13.

(Third embodiment)
[Configuration of stereoscopic image correction apparatus]
With reference to FIG. 17, a difference from the second embodiment will be described regarding a stereoscopic image correction apparatus 1 </ b> C according to the third embodiment of the present invention.
The stereoscopic image correction apparatus 1C is different from the second embodiment in that a warning is issued under a predetermined condition. Therefore, the stereoscopic image correction apparatus 1C includes a display state calculation unit 10C, a stereoscopic image processing unit 20B, a delay unit 70, a parameter setting unit 80, and a warning unit 90.

Here, the parameter setting means 80 will be described first.
The parameter setting means 80 is for the observer to set various parameters (individual depth range information, individual depth time change information, operation mode information).
The parameters set in the parameter setting unit 80 are referred to by the display state calculation unit 10C, the stereoscopic image processing unit 20B, and the warning unit 90.

The individual depth range information is information indicating a preset depth range (individual depth range) for each observer.
The individual depth time change information is information indicating a preset depth time change (individual depth time change) for each observer, which is set in advance.
The operation mode information is information indicating whether the stereoscopic image correction apparatus 1C operates in the reference setting mode or the viewer setting mode.

  The reference setting mode is that the stereoscopic image correction apparatus 1C operates using the reference depth range information and the reference depth time change information. That is, in the case of the reference setting mode, the stereoscopic image correction apparatus 1C uses a common depth range and a common depth time change common to all viewers.

  The viewer setting mode is the operation of the stereoscopic image correction apparatus 1C using the individual depth range information and the individual depth time change information. That is, in the case of the viewer setting mode, the stereoscopic image correction apparatus 1C uses the individual depth range and the individual depth time change unique to each viewer. Furthermore, the stereoscopic image correction apparatus 1C issues a warning when the correction element image group and the correction delay element image group may cause eye strain.

The display state calculation unit 10C includes a depth range calculation unit 11C, a depth range determination unit 13C, a depth time change calculation unit 15C, and a depth time change determination unit 17C.
Further, the display state calculation unit 10C receives the correction element image group and the correction delay element image group from the stereoscopic image processing unit 20B.

In the viewer setting mode, the depth range calculation unit 11C determines the depth range (correction depth range) of the correction element image group and the depth range of the correction delay element image group in the same manner as the depth range calculation unit 11B of FIG. Is to be calculated.
The corrected depth range calculated by the depth range calculation unit 11C is output to the warning unit 90.

  The depth range determination unit 13C selects one of the individual depth range information and the reference depth range information based on the operation mode information, and determines the same as the depth range determination unit 13 of FIG.

In the viewer setting mode, the depth time change calculation unit 15C changes the depth time change (corrected depth time change) between the correction element image group and the correction delay element image group in the same manner as the depth time change calculation unit 15 in FIG. Is to be calculated.
The corrected depth time change calculated by the depth time change calculation means 15C is output to the warning means 90.

  Depth time change determination means 17C selects either individual depth time change information or reference depth time change information based on the operation mode information, and makes the same determination as depth time change determination means 17 in FIG. It is.

  In the viewer setting mode, the warning unit 90 issues a warning according to a predetermined condition, and includes a depth range warning unit 91 and a depth time change warning unit 93.

The depth range warning unit 91 determines whether or not the corrected depth range input from the depth range calculation unit 11C exceeds the common depth range, and warns when the corrected depth range exceeds the common depth range.
Here, the depth range warning means 91 may add a warning message for the depth range to the video signal without any particular limitation on the warning method. Further, the depth range warning means 91 may transmit this warning message to a preset mail address or may sound a predetermined warning sound.

The depth time change warning means 93 determines whether or not the corrected depth time change input from the depth time change calculation means 15C exceeds the common depth time change, and warns when the corrected depth time change exceeds the common depth time change. To do.
Here, the depth time change warning means 93 can give a warning in the same manner as the depth range warning means 91.

  As described above, the stereoscopic image correction apparatus 1 </ b> C can reproduce a stereoscopic image so as to be within a depth range and depth time change that are optimal for each observer, and may cause eye strain during reproduction of the stereoscopic image. If there is, it can alert the observer.

  Note that the stereoscopic image correcting apparatus 1C may be configured to use only one of the individual depth range information and the individual depth time change information, or the reference depth range information and the reference depth time change information without including the warning unit 90. Good.

1, 1B, 1C Stereoscopic image correction apparatus 10, 10B, 10C Display state calculation means 11, 11B, 11C Depth range calculation means 13, 13C Depth range determination means 15, 15C Depth time change calculation means 17, 17C Depth time change determination means 20, 20B Stereoscopic image processing means 30 Planar light intensity distribution calculating means 31 Dividing means 33 Element image converting means 33a Light wave calculating means 33b Phase shift means 33c Light wave calculating means 35 Combining means 40 Volume light intensity distribution storing means 50 Depth range correcting means 60 Correction element image group generation means 61 Element image inverse conversion means 61a Light wave calculation means 61b Phase shift means 61c Light wave calculation means 61d Addition means 63 Connection means 70 Delay means 80 Parameter setting means 90 Warning means 91 Depth range warning means 93 Depth time change warning means

Claims (6)

Corrects a depth range of a stereoscopic image when an elemental image group captured by a stereoscopic image capturing device using the integral photography method is displayed on a stereoscopic image display device, and a change in depth time that is a change amount of the depth range. A stereoscopic image correction device,
A delay means for inputting the element image group, delaying the input element image group by a preset delay time, and outputting a delay element image group;
A depth range calculating means for calculating a display depth range and a delayed display depth range when the element image group and the delay element image group are respectively displayed on the stereoscopic image display device by a predetermined depth range calculation method;
Depth range determining means for determining whether or not the display depth range exceeds a preset reference depth range;
A depth time change calculating means for calculating a display depth time change which is a change amount between the display depth range and the delay display depth range;
Depth time change determination means for determining whether or not the display depth time change exceeds a preset reference depth time change;
When the display depth range exceeds the reference depth range, virtual element optics in which virtual element optical systems having the same pitch and focal length of the element optical system of the stereoscopic image display device are arranged in a two-dimensional manner for the element image group A plane light intensity distribution, which is a light intensity distribution for each preset distance plane having a different distance from the virtual element optical system group, is calculated from the element image group by performing wave optical calculation via the system group. When the display depth time change exceeds the reference depth time change, wave optical computation is performed on the delay element image group via the virtual element optical system group, so that the plane from the delay element image group is obtained. Planar light intensity distribution calculating means for calculating the light intensity distribution;
Volume light intensity distribution storage for storing the plane light intensity distribution for each distance plane calculated by the plane light intensity distribution calculating means as a volume light intensity distribution on a stereoscopic image space for displaying the stereoscopic image in association with the distance. Means,
When the display depth range exceeds the reference depth range, the coordinates of the volume light intensity distribution in the element image group are corrected in the depth direction at a ratio between the display depth range and the reference depth range, and the display depth time change Depth range correction means for correcting the coordinates of the volume light intensity distribution in the delay element image group in the depth direction by the difference between the display depth time change and the reference depth time change,
By performing wave optical calculation via the virtual element optical system group for each distance plane of the volume light intensity distribution corrected by the depth range correction unit, at least of the corrected element image group and the delayed element image group Correction element image group generation means for generating one of them,
A three-dimensional image correction apparatus comprising: A stereoscopic image correction apparatus that corrects a depth range of a stereoscopic image when an elemental image group captured by a stereoscopic image capturing apparatus using an integral photography method is displayed on a stereoscopic image display apparatus,
Depth range calculation means for calculating a display depth range when the element image group is displayed on the stereoscopic image display device by a predetermined depth range calculation method;
Depth range determining means for determining whether or not the display depth range exceeds a preset reference depth range;
When the display depth range exceeds the reference depth range, virtual element optics in which virtual element optical systems having the same pitch and focal length of the element optical system of the stereoscopic image display device are arranged in a two-dimensional manner for the element image group Planar light intensity distribution calculation that calculates a planar light intensity distribution, which is a light intensity distribution for each preset distance plane with a different distance from the virtual element optical system group, by performing wave optical calculation via the system group Means,
Volume light intensity distribution storage for storing the plane light intensity distribution for each distance plane calculated by the plane light intensity distribution calculating means as a volume light intensity distribution on a stereoscopic image space for displaying the stereoscopic image in association with the distance. Means,
When the display depth range exceeds the reference depth range, a depth range correction unit that corrects the coordinates of the volume light intensity distribution in the element image group in the depth direction at a ratio between the display depth range and the reference depth range;
Correction element image group for generating a corrected element image group by performing wave optical calculation via the virtual element optical system group for each distance plane of the volumetric light intensity distribution corrected by the depth range correction unit. Generating means;
A three-dimensional image correction apparatus comprising: A stereoscopic image correction device that corrects a depth time change, which is a change amount of a depth range of a stereoscopic image when an element image group captured by a stereoscopic image capturing device by an integral photography method is displayed on a stereoscopic image display device. And
A delay means for inputting the element image group, delaying the input element image group by a preset delay time, and outputting a delay element image group;
A depth range calculating means for calculating a display depth range and a delayed display depth range when the element image group and the delay element image group are respectively displayed on the stereoscopic image display device by a predetermined depth range calculation method;
A depth time change calculating means for calculating a display depth time change which is a change amount between the display depth range and the delay display depth range;
Depth time change determination means for determining whether or not the display depth time change exceeds a preset reference depth time change;
When the display depth time change exceeds the reference depth time change, virtual element optical systems having the same pitch and focal length of the element optical system of the stereoscopic image display device are two-dimensionally arranged for the delay element image group. Plane light that calculates a planar light intensity distribution, which is a light intensity distribution for each preset distance plane that has a different distance from the virtual element optical system group, by performing wave optical calculation via the virtual element optical system group Intensity distribution calculating means;
Volume light intensity distribution storage for storing the plane light intensity distribution for each distance plane calculated by the plane light intensity distribution calculating means as a volume light intensity distribution on a stereoscopic image space for displaying the stereoscopic image in association with the distance. Means,
When the display depth time change exceeds the reference depth time change, the depth for correcting the coordinate of the volume light intensity distribution in the delay element image group in the depth direction by the difference between the display depth time change and the reference depth time change Range correction means;
A correction element image for generating a corrected delay element image group by performing wave optical calculation via the virtual element optical system group for each distance plane of the volumetric light intensity distribution corrected by the depth range correction unit. Group generation means;
A three-dimensional image correction apparatus comprising: The depth range determination means determines whether or not an individual depth range unique to each observer is preset as the reference depth range, and determines whether the display depth range exceeds the individual depth range;
The depth range calculation means further calculates a depth range when the corrected elemental image group is displayed on the stereoscopic image display device by the depth range calculation method,
A common depth range common to the observer is set in advance, and it is determined whether or not a depth range calculated from the corrected element image group exceeds the common depth range, and the depth range corresponds to the common depth range. Depth range warning means to warn when exceeding,
The stereoscopic image correction apparatus according to claim 1, further comprising: The depth time change determination means determines whether or not the individual depth time change unique to each observer is preset as the reference depth time change, and the display depth time change exceeds the individual depth time change. ,
The depth range calculation means further calculates a depth range when the corrected delay element image group is displayed on the stereoscopic image display device by the depth range calculation method,
The depth time change calculating means further calculates a depth time change which is a change amount between the display depth range and the corrected depth element image group,
A common depth time change common to the observer is preset, it is determined whether the depth time change calculated from the corrected delay element image group exceeds the common depth time change, and the depth time change Depth time change warning means for warning when the common depth time change is exceeded,
The stereoscopic image correction apparatus according to claim 1, further comprising:   A stereoscopic image correction program for causing a computer to function as the stereoscopic image correction apparatus according to claim 1.

Images