JP3891226B2 - Image data generation apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image data generation apparatus and method, and more particularly to an image data generation apparatus and method that can capture an image more efficiently by using a lens having a relatively small angle of view.
[0002]
[Prior art]
In a holographic stereogram printer, when a subject is photographed by a video camera to create a holographic stereogram, it is desired to capture a large number of parallax images of the subject in a short time. FIG. 14 shows a method for inputting a parallax image in such a case.
[0003]
As shown in the figure, the video camera 2 is arranged so that its optical axis 2 a is perpendicular to the viewpoint path 3. The video camera 2 is moved linearly along the viewpoint path 3 while keeping the optical axis 2 a perpendicular to the viewpoint path 3. When the video camera 2 moves along the viewpoint path 3, an image obtained by imaging the subject 1 by the video camera 2 is captured as a parallax image.
[0004]
The image captured in this manner is displayed on an LCD or the like after a viewpoint conversion process for localizing the reproduced image near the hologram surface. And object light is produced | generated from this display image, A holographic stereogram is produced using this.
[0005]
[Problems to be solved by the invention]
By the way, in such a conventional image capturing method, the video camera 2 is moved while holding its optical axis 2 a so as to be always perpendicular to the viewpoint path 3. However, a relatively wide-angle photographic lens is required so that the video camera 2 can capture the subject 1 even when the subject 1 moves left or right from the center. As a result, there is a problem that not only the image is likely to be distorted but also it is difficult to capture a high-resolution image in the parallax direction. Further, since there is no need for an image in a range where the subject 1 has not been photographed, there is a problem that wasteful image data increases.
[0006]
The present invention has been made in view of such a situation, and by using a video camera using a standard lens having a relatively small angle of view, an image can be captured and image distortion is suppressed. In addition, a high-resolution image can be efficiently captured depending on the parallax direction.
[0007]
[Means for Solving the Problems]
  The image data generation device according to claim 1, wherein the image data generation device moves on a straight line, rotates so that the optical axis always faces the subject, and takes in a parallax image of the subject,The parallax image captured by the capturing means is scaled by the scaling means and the scaling means for each pixel column corresponding to the rotation angle of the capturing means and the angle from the optical axis, and the scaling means. Mapping means for mapping the parallax image as an image on a plane parallel to the subject, display means for displaying the image mapped by the mapping means, and light from the image displayed on the display means as object light Creating means to create holographic stereograms,It is characterized by providing.
[0008]
  The image data generation method according to claim 6, wherein the image data generation method is configured to capture a parallax image of a subject by rotating the optical axis so as to always face the subject while moving on a straight line.A scaling step for scaling the parallax image captured by the processing of the capturing step in the pixel column direction corresponding to the rotation angle of the capturing unit and the angle from the optical axis for each pixel column, Displayed by the mapping step for mapping the parallax image scaled by the process as an image on a plane parallel to the subject, the display step for displaying the image mapped by the process of the mapping step, and the process of the display step Creating a holographic stereogram using light from an image as object lightIt is characterized by providing.
[0009]
  In the image data generation device according to claim 1 and the image data generation method according to claim 6, when capturing a parallax image of a subject, the optical axis is always directed to the subject,The captured parallax image is scaled in the pixel column direction corresponding to the rotation angle and angle from the optical axis when the parallax image is captured for each pixel column, and the scaled parallax image is parallel to the subject. Since the mapped image is displayed as an image on a plane in a different direction, and the light from the displayed image is used as object light, a holographic stereogram is created.The keystone distortion of the captured parallax image is corrected.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below, but in order to clarify the correspondence between each means of the invention described in the claims and the following embodiments, in parentheses after each means, The features of the present invention will be described with the corresponding embodiment (however, an example) added. However, of course, this description does not mean that each means is limited to the description.
[0011]
  The image data generation apparatus according to claim 1 is a capturing unit that captures a parallax image of a subject by moving the optical axis so that the optical axis always faces the subject while moving on a straight line (for example, the CCD video camera of FIG. 61)Scaling means for scaling the parallax image captured by the capturing means for each pixel column in accordance with the rotation angle of the capturing means and the angle from the optical axis in the pixel column direction(For example, the flowchart of FIG.Step S3(Viewpoint conversion processing CPU 72 in FIG. 3)And mapping means for mapping the parallax image scaled by the scaling means as an image on a plane parallel to the subject (for example, the viewpoint conversion processing CPU 72 in FIG. 3 that performs the processing in step S4 in the flowchart in FIG. 9). And display means (for example, LCD 30 in FIG. 3) that displays the image that has been subjected to the keystone distortion correction processing, and creation that creates a holographic stereogram using light from the image displayed on the display means as object light Means (for example, the exposure head control CPU 81 and the exposure head 82 in FIG. 3).
[0013]
The image data generation apparatus according to claim 4 further includes conversion means (for example, a viewpoint conversion processing CPU 72 of FIG. 3 that performs the processing of the flowchart of FIG. 13) that performs viewpoint conversion processing for locating the reproduced image on the hologram surface. It is characterized by providing.
[0014]
FIGS. 1 and 2 respectively show a configuration of the optical system of the holographic stereogram printer to which the image data generation device of the present invention is applied, and a configuration viewed from the side.
[0015]
The light source 21 generates light L1. The shutter 22 is controlled so as to pass or block the light L1 emitted from the light source 21. The light L2 that has passed through the shutter 22 enters the half mirror 23 and is separated into reflected light and transmitted light. The reflected light L3 is incident on the cylindrical lens 24 as reference light. The cylindrical lens 24 converts incident light into divergent light. The collimator lens 25 converts the divergent light incident from the cylindrical lens 24 into parallel light and emits it to the total reflection mirror 26. The light reflected by the total reflection mirror 26 is incident on the hologram recording medium 32 from an oblique direction as reference light.
[0016]
On the other hand, the light L4 as the object light transmitted through the half mirror 23 is reflected by the total reflection mirror 27 and is incident on the spatial filter 28. The spatial filter 28 includes an objective lens and a pinhole, converts incident light into divergent light, and outputs the divergent light to the collimator lens 29. The collimator lens 29 converts the incident divergent light into parallel light and emits it to the LCD 30.
[0017]
The light transmitted through the LCD 30 is converged in the horizontal direction by the cylindrical lens 31 and is incident on the hologram recording medium 32 at the exposure angle θe. The recording medium feeding mechanism 33 moves the hologram recording medium 32 at a predetermined timing.
[0018]
FIG. 3 shows a configuration example of the electrical system of the holographic stereogram printer. As shown in the figure, this electrical system basically includes a parallax image input unit 51, a viewpoint conversion processing unit 52, and a hologram exposure unit 53. The parallax image input unit 51 includes a CCD video camera 61 that captures an image of a subject and captures the parallax image using the NTSC method.
[0019]
The viewpoint conversion processing unit 52 is a JPEG compression board 71 that performs processing for compressing and expanding the image signal supplied from the CCD video camera 61 using the JPEG method, and viewpoint conversion processing that performs keystone distortion correction processing and viewpoint conversion processing. A CPU 72 and a RAM 73 for temporarily storing image data are provided.
[0020]
The hologram exposure unit 53 is driven by the LCD driver 83 to which the image data after the keystone distortion correction process and the viewpoint conversion process are performed from the JPEG compression board 71, and is driven by the LCD driver 83 to display an image. It has an LCD 30. The hologram exposure unit 53 also has an exposure head control CPU 81 to which a timing signal is supplied from the viewpoint conversion processing CPU 72 of the viewpoint conversion processing unit 52 and an exposure head 82 controlled by the exposure head control CPU 81. The exposure head 82 has the light source 21 and the shutter 22 shown in FIG.
[0021]
Next, the basic operation will be described. When creating a holographic stereogram, a subject is imaged by the CCD video camera 61 and the image signal (NTSC image signal) is input to the JPEG compression board 71 as a parallax image signal. The JPEG compression board 71 binarizes the input image signal, compresses the image signal using the JPEG method, and converts the image signal into data in a VGA (Video Graphics Array (trademark)) format. This data is stored in the RAM 73 via the viewpoint conversion processing CPU 72. As will be described later, the CCD video camera 61 is moved at a constant speed along the viewpoint path 111 (FIG. 5). Thereby, the RAM 73 compresses and stores a plurality of images each composed of 480 pixels in the horizontal direction and 640 pixels in the vertical direction.
[0022]
Here, the reason for setting 640 [pixel] in the vertical direction and 480 [pixel] in the horizontal direction is that the amount of information tends to be smaller in the holographic stereogram having only horizontal parallax than in the parallax direction. Since the pixels are sequentially arranged in the scanning direction with the upper left corner of the image as the origin, the horizontal line of the image is processed when the viewpoint conversion processing is performed. The processing speed can be increased by exchanging continuous data between parallax images in units of.
[0023]
The viewpoint conversion processing CPU 72 reads the image data stored in the RAM 73, supplies it to the JPEG compression board 71, and decompresses it. Then, the decompressed image data is stored in the RAM 73 again. Then, the decompressed image data is read again, and keystone distortion correction processing and viewpoint conversion processing (details thereof will be described later) are performed. The image after the viewpoint conversion process is output from the JPEG compression board 71 to the LCD 30 via the LCD driver 83 of the hologram exposure unit 53 and displayed.
[0024]
The viewpoint conversion processing CPU 72 generates a timing signal synchronized with the image displayed on the LCD 30 and supplies the timing signal to the exposure head control CPU 81 of the hologram exposure unit 53 via RS-232C or Ethernet. The exposure head CPU 81 controls the exposure head 82 in synchronization with this timing signal to execute an exposure operation.
[0025]
The light L1 generated by the light source 21 of the exposure head 82 is incident on the half mirror 23 through the shutter 22 that is opened during the exposure period. The light L3 as the reference light reflected by the half mirror 23 is incident on the cylindrical lens 24, converted into divergent light, then incident on the collimator lens 25, and converted into parallel light. The light emitted from the collimator lens 25 is reflected by the total reflection mirror 26 and is incident on one surface of the hologram recording medium 32.
[0026]
On the other hand, the light L4 as the object light transmitted through the half mirror 23 is reflected by the total reflection mirror 27, then enters the spatial filter 28, and is converted into divergent light. The divergent light is incident on the collimator lens 29, converted into parallel light, and then irradiated on the LCD 30. As described above, the LCD 30 displays an image captured by the CCD video camera 61 and subjected to keystone distortion correction processing and viewpoint conversion processing. Accordingly, the light corresponding to the image displayed on the LCD 30 is incident on the other surface of the hologram recording medium 32 through the cylindrical lens 31 at an exposure angle θe substantially perpendicularly. As a result, the reference light and the object light interfere on the hologram recording medium 32, and interference fringes generated by this interference are recorded on the hologram recording medium 32 as a change in refractive index.
[0027]
The recording medium feed mechanism 33 is controlled by the exposure head control CPU 81, and each time a strip-shaped element hologram (FIG. 4) corresponding to one image is formed, the hologram recording medium 32 is moved in a predetermined direction. , Transport sequentially. Thereby, the hologram recording medium 32 is formed as a holographic stereogram capable of reproducing a three-dimensional hologram.
[0028]
FIG. 4 shows a schematic configuration of the holographic stereogram 101 formed in this way. As shown in the figure, the holographic stereogram 101 has a large number (m) of strip-shaped element holograms 102 formed therein. Each element hologram 102 is formed with a hologram corresponding to an image having a width of 480 pixels and a length of 640 pixels.
[0029]
As shown in FIG. 5, the CCD video camera 61 is moved linearly along the viewpoint path 111. At this time, in the CCD video camera 61, the optical axis 61a is rotated by an angle θ corresponding to the position on the viewpoint path 111 so that the optical axis 61a always points to the subject 112. Note that the angle of view θv of the camera when the CCD video camera 61 captures the subject 112 need not correspond to the exposure angle θe shown in FIG.
[0030]
In this way, the CCD video camera 61 is controlled so that its optical axis 61a is always directed to the subject 112 regardless of the position on the viewpoint path 111, so that the photographing lens of the CCD video camera 61 has a wide angle. This is unnecessary, and a parallax image of the subject can always be taken with a relatively small angle of view. As a result, a general-purpose camera such as a home video camera can be used as the CCD video camera 61. Since a lens having a relatively small angle of view can be used, image distortion is also suppressed, and the resolution in the parallax direction can be improved as compared with the case where a wide-angle lens is used.
[0031]
Moving distance d of the CCD video camera 61_CIs represented by the following equation.
d_C= 2d_V・ Tan (θe / 2) + W_H  ... (1)
[0032]
Where d_VRepresents the photographing distance (shortest distance) between the CCD video camera 61 and the subject 112, and this distance corresponds to the viewpoint distance during hologram observation. That is, when the hologram surface 113 is disposed at the position of the subject 112, the viewpoint for observing the image on the hologram surface 113 is the position (viewpoint position) where the CCD video camera 61 is disposed. Accordingly, the viewpoint is positioned on the straight line (viewpoint path 111) on which the CCD video camera 61 moves.
[0033]
In addition, W in the above equation (1)_HRepresents the lateral length of the holographic stereogram to be created.
[0034]
The angle of view θv of the CCD video camera 61 is expressed by the following equation.
θv = θc−θe (2)
[0035]
Here, θe represents the exposure angle of the hologram, and θc represents the camera movement distance d._CWhen viewed from the subject 112, that is, in FIG.₁And point P_ThreeAnd a line connecting the point P₂And points_ThreeIs an angle formed by a straight line connecting the two and is expressed by the following equation.
θc = 2tan^-1(Tan (θe / 2) + W_H/ 2) (3)
[0036]
By taking a picture using the angle of view determined by these equations, a parallax image can be taken with the smallest angle of view, and an image for exposure is reconstructed from the taken parallax image by viewpoint conversion. Is compensated.
[0037]
As shown in FIG. 5, when the optical axis 61a of the CCD video camera 61 is always directed toward the subject 112, keystone distortion occurs. For simplicity, a rectangular object having no thickness in the depth direction is considered here. That is, as shown in FIG. 6A, when the CCD video camera 61 (the CCD 61A) and the subject 112 are parallel to each other, as shown in FIG. The length in the direction is represented by the same length. However, when the CCD video camera 61 (the CCD 61A) and the subject 112 are not parallel as shown in FIG. 7A, the corresponding vertical directions on the left and right sides of the subject 112 are shown in FIG. 7B. Will not match. Therefore, before reconstructing the image for exposure (before displaying the image captured by the CCD video camera 61 on the LCD 30), it is necessary to correct this keystone distortion.
[0038]
FIG. 8 shows the principle of correcting this keystone distortion. As shown in the figure, the CCD video camera 61 now has a point P.₂Located at the top and its rotation angle (point P_FourAnd point P_ThreeAnd the point P₂And point P_ThreeThe angle between the straight line connecting the two and θ) is expressed by the following equation.
θ = tan^-1(Δd_C/ D_V(4)
[0039]
Where Δd_CIs the camera movement distance d_CUpper center point P_FourAnd the point P where the CCD video camera 61 is located at that time₂Represents the distance between
[0040]
And point P₂When the angle from the optical axis 61a representing the position of the pixel in the horizontal direction of the image captured by the CCD video camera 61 located above is α, the enlargement / reduction ratio ratio for keystone distortion correction is expressed by the following equation.
ratio = 1 + tan α · tan θ (5)
However, −θv / 2 ≦ α ≦ θv / 2 (θv: angle of view of the camera)
[0041]
Therefore, the CCD video camera 61 can perform the enlargement / reduction (scaling) processing on the pixels in the pixel row defined by the angle α in the vertical direction (pixel row direction), so that the point P_FourIt is possible to correct the difference between the corresponding vertical lengths of the left and right of the image that occurs when moving from right to left or moving left.
[0042]
In order to correct the keystone distortion, the direction of the projection surface of the CCD video camera 61 (the surface of the CCD 61A) and the direction of the subject 112 (the hologram surface 113) are not parallel at the rotation angle θ. Mapping process is required. Therefore, as shown in FIG. 8, the x-axis is taken in a direction parallel to the hologram surface 113, and the x-axis and the point P_ThreeThe x′-axis is taken in the direction parallel to the projection plane of the CCD video camera 61 so as to intersect with each other, and the pixel column data of the CCD video camera 61 represented by the x′-axis coordinate is converted to the x-axis coordinate. To the pixel column data. In this case, the relationship between x, x ', α, and θ is as follows.
α = tan^-1[X / {d_V(Sinθ · tanθ + cosθ) −x · tanθ}]
... (6)
x ′ = d_V・ Tanα (7)
[0043]
That is, the coordinate conversion process can be performed using the above equations (6) and (7).
[0044]
The keystone distortion correction processing executed by the viewpoint conversion processing CPU 72 in accordance with the above principle will be described with reference to the flowchart of FIG. First, in step S1, an angle α corresponding to a pixel column to be processed is obtained. Next, in step S2, the enlargement / reduction ratio ratio shown in the above equation (5) is calculated.
[0045]
Next, the process proceeds to step S3, and scaling processing of the processing target pixel row is executed using the enlargement / reduction ratio ratio obtained in step S2.
[0046]
Further, the process proceeds to step S4, and the process (mapping process) for converting the pixel column data represented by the x ′ coordinate into the coordinate represented by the x coordinate using the above-described expressions (6) and (7) Done.
[0047]
Next, in step S5, it is determined whether or not the processing for all the pixel columns has been completed. If there is a pixel row that has not yet been processed, the process returns to step S1 and the same processing is repeatedly executed.
[0048]
If it is determined in step S5 that the processing for all pixel ratios has been completed, the process proceeds to step S6, where it is determined whether or not the processing for all input images has been completed. If there is an input image that has not been processed yet, the process returns to step S1 and the same processing is repeatedly executed. When the processing for all the input images is completed, the keystone distortion correction processing is completed.
[0049]
By performing the above processing, a planar holographic stereogram can be created, and the viewpoint at the time of observation can be parallel to the hologram surface.
[0050]
By the way, when white light is used as the light source 21 in FIG. 1, the reproduced image is blurred as the reproduced image is further away from the vicinity of the hologram surface. Therefore, it is necessary to perform viewpoint conversion processing so that the reproduced image is localized near the hologram surface. Since the positional relationship between the camera viewpoint and the subject at the time of shooting is preserved for the reconstructed image after exposure, when the viewpoint conversion processing is not performed, the observer's eyes are placed on the hologram surface during observation. Unless this is the case, a reproduced image without distortion cannot be obtained. Theoretically, the reproduced image is formed deeper than the hologram surface by a distance corresponding to the distance between the video camera and the subject, so that the reproduced image becomes a blurred image based on the above-described properties. . Therefore, it is necessary to perform a viewpoint conversion process in which the reproduced image is localized in the vicinity of the hologram surface and the viewpoint at the time of hologram observation is set in front of the hologram surface. Next, this viewpoint conversion process will be described.
[0051]
FIG. 10 shows the positional relationship between the input image and the hologram exposure image (image reconstructed by viewpoint conversion). The hologram exposure image 131 needs to have its angle of view coincident with the hologram exposure angle θe. The viewpoint conversion process is realized by newly generating (reconstructing) one hologram exposure image 131 from a plurality of input images.
[0052]
Here, for the sake of simplicity, when the input image is rotated around the viewpoint path 111 shown in FIG. 10, the input image and the hologram exposure image are in a positional relationship as shown in FIG.
[0053]
As shown in FIG. 11, on the hologram surface 113, the exposure point ep corresponds to the n element holograms 102.₁Thru ep_nExists. Each exposure point ep_iEach of the hologram exposure images 131-i includes 480 pixel columns (parallax information) 131-i-1 to 131-i-480. Each pixel column (parallax information) is composed of one horizontal pixel and 640 vertical pixels.
[0054]
Of these, for example, the exposure point ep₁When attention is paid to this exposure point ep, as shown in FIG.₁This image is composed of a total of 480 pixel columns of the pixel columns 131-1-1 to 131-1-480 of the hologram exposure image.
[0055]
Of these, for example, the pixel column 131-1-1 of the image for hologram exposure has a leftmost viewpoint vP in the drawing on the viewpoint path 111.₁One pixel column is selected from the 480 pixel columns of the input image 141-1 captured in step S 1, and is associated. However, as described above, the input image 141-1 is subjected to the processing for correcting the keystone distortion and is mapped in parallel with the hologram surface 113, so that the N constituting the input image 141 A- 1 after the distortion correction is configured.₁Book (in this embodiment, N₁= 480), one pixel column (in the case of the example of FIG. 12, the rightmost pixel column PC in the figure)₁) Is selected and responded.
[0056]
The pixel row 131-1-2 of the image for hologram exposure has a viewpoint vP₂Is the pixel column data of the input image 141-2 captured in step N, and N after distortion correction₂(In this embodiment, N₂= 480) One predetermined pixel column from among the pixel columns (in this embodiment, N from the left side)_{twenty one}The pixel row PC_{twenty one}) Is selected and responded.
[0057]
As described above, the exposure point ep₁The processing for obtaining the 480 hologram exposure image pixel columns 131-1-1 to 131-1-480 is performed.
[0058]
Next, the details of the viewpoint conversion processing performed by the viewpoint conversion processing CPU 72 will be further described with reference to the flowchart of FIG.
[0059]
First, in steps S21 and S22, variables j and i are initially set to 1, and in step S23, n exposure points ep on the hologram surface 113 are set.₁Thru ep_nOf these, one exposure point is selected. For example, the exposure point ep_j(J = 1 to n) of exposure points ep₁Is selected. And this exposure point ep₁Among the 480 holographic exposure image pixel rows 131-1-1 to 131-1-480 that constitute the image, one pixel row is a sampling point r._i(I = 1 to 480) is selected. For example, the pixel row 131-1-1 is a sampling point r.₁Selected as.
[0060]
The exposure point ep selected in this way₁And sampling point r₁A virtual line LN1 that connects (pixel column 131-1-1) is defined. In the same manner, the sampling point r₂Thru r₄₈₀As a result, the pixel columns 131-1-2 through 131-1-480 are selected and their sampling points r are selected._iAnd exposure point ep₁An imaginary line LN1 is defined. As a result, a total of 480 virtual lines LN1 are defined.
[0061]
Next, proceeding to step S24, the intersection of the virtual line LN1 defined at step S23 and the viewpoint path 111 is calculated. As a result, n intersection points are obtained on the viewpoint path 111.
[0062]
In step S25, m viewpoints vP on the viewpoint path 111 are displayed.₁To vP_mThe viewpoint vP closest to the intersection obtained in step S24 from (the point where the subject 112 was photographed by the CCD video camera 61)._kIs searched. Thus, m viewpoints vP₁To vP_mA maximum of n viewpoints are searched from the list. When two or more viewpoints overlap and are searched corresponding to different sampling points, the number of viewpoints is n or less.
[0063]
In step S26, the viewpoint vP searched in step S25._kThe input image 141-k captured in is selected. As a result, a maximum of n input images 141 are selected.
[0064]
Next, the process proceeds to step S27, and the viewpoint vP obtained in step S26._kAnd exposure point ep₁Is defined. Thereby, a maximum of n virtual lines LN2 are defined.
[0065]
Further, in step S28, an intersection between the virtual line LN2 obtained in step S27 and the input image after the correction of the keystone distortion is obtained. In step S29, the pixel row of the input image located at the intersection obtained in step S28 is mapped as the pixel row of the hologram exposure image at the corresponding sampling point.
[0066]
For example, as shown in FIG. 12, the exposure point ep₁Sampling point r₁Corresponding to the pixel row 131-1-1 of the hologram exposure image as₁And exposure point ep₁When the straight line passing through is one of the virtual lines LN2, the pixel column PC of the input image 141A-1 after distortion correction located on this straight line₁Is selected and mapped on the RAM 73 as a pixel column corresponding to the pixel column 131-1-1.
[0067]
Further, as a viewpoint corresponding to the pixel column 131-1-2 of the hologram exposure image, the viewpoint vP is used.₂Is selected and the viewpoint vP₂And exposure point ep₁Is defined. Then, N pixels from the left end of the input image 141A-2 after distortion correction are set as pixel columns located on the virtual line LN2._{twenty one}Th pixel column PC₂Is required. And this pixel column PC₂Are mapped as pixel columns of the pixel column 131-1-2 of the hologram image.
[0068]
As described above, the exposure point ep₁The pixel columns 131-1-1 to 131-1-480 of the hologram exposure image are mapped.
[0069]
In step S30, it is determined whether mapping at all sampling points of the exposure image is complete (i = 480). In this case, exposure point ep₁The mapping of is only completed. Therefore, the process proceeds to step S31, i is incremented by 1, and the process returns to step S23. And exposure point ep₁Next exposure point ep₂The same processing is executed for. The above processing is performed at the exposure point ep.₁Thru ep_nIn step S30, it is determined that mapping at all sampling points of the exposure image has been completed, and the process proceeds to step S32.
[0070]
In step S32, it is determined whether or not mapping has been completed for all exposure points (whether j = n). If NO, the process proceeds to step S33, and j is incremented by 1, and then step S32 is performed. Returning to S22, the subsequent processing is executed. When a determination of YES is made in step S32, the process ends.
[0071]
Various parameters such as the position of the subject, the angle of view, and the position of the viewpoint necessary for performing the viewpoint conversion process need not be changed if set once. In the viewpoint conversion process, since the parallax image to be processed is an image having only horizontal parallax, if the corresponding parallax information (a slit-like image composed of one horizontal pixel and 640 vertical pixels) can be calculated, the slit conversion process is finally performed. It can be executed by exchanging the images. Therefore, at the time of the first calculation, the correspondence relationship between the slit-like images of all the exposure images and the slit-like images of the original parallax images is stored in a storage medium such as a hard disk in advance as a table. Then, the next viewpoint conversion process can be performed at high speed with reference to this table.
[0072]
In the above embodiment, the NTSC system is used as the CCD video camera 61, but a PAL system or other system can also be used. Further, the format processed by the viewpoint conversion processing CPU 72 is not limited to the VGA, but can be an image of another format. Furthermore, it is possible to use a laser light source instead of a white light source as the light source.
[0073]
In the above-described embodiment, the holographic stereogram printer that creates a planar holographic stereogram is basically integrated with the parallax image input unit 51, the viewpoint conversion processing unit 52, and the hologram exposure unit 53. However, it is also possible to separate them appropriately to form a separate structure.
[0074]
【The invention's effect】
  As described above, according to the image data generation device according to claim 1 and the image data generation method according to claim 6, the optical axis always moves toward the subject while moving linearly,The captured parallax image is scaled in the pixel column direction corresponding to the rotation angle and angle from the optical axis when capturing the parallax image for each pixel column, and the scaled parallax image is parallel to the subject. Mapping as an image on the surface of the image, displaying the mapped image, and creating a holographic stereogram using the light from the displayed image as object lightSince it did in this way, the angle of view at the time of imaging | photography can be made comparatively small, and it becomes possible to utilize a home video camera. In addition, by reducing the angle of view, it is possible to suppress image distortion and improve resolution compared to the case of using a wide-angle lens. Furthermore, since a parallax image of the subject can always be obtained, it is possible to suppress the useless image capture. In addition, a planar holographic stereogram having an observation viewpoint parallel to the planar hologram surface can be created.
[Brief description of the drawings]
FIG. 1 is a diagram showing a plane configuration of an optical system of a holographic stereogram printer to which an image data generation apparatus of the present invention is applied.
FIG. 2 is a diagram showing a configuration of a part of a side surface of the optical system shown in FIG.
FIG. 3 is a block diagram illustrating a configuration example of an electrical system of a holographic stereogram printer to which the image data generation apparatus of the present invention is applied.
FIG. 4 is a perspective view illustrating a holographic stereogram.
FIG. 5 is a diagram illustrating a parallax image input method.
FIG. 6 is a diagram illustrating keystone distortion.
FIG. 7 is a diagram for correcting keystone distortion.
FIG. 8 is a diagram illustrating a keystone distortion correction process.
FIG. 9 is a flowchart illustrating a keystone distortion correction process.
FIG. 10 is a diagram illustrating a positional relationship between an input image and a hologram exposure image.
FIG. 11 is a diagram illustrating viewpoint conversion processing.
FIG. 12 is another diagram for explaining viewpoint conversion processing;
FIG. 13 is a flowchart illustrating viewpoint conversion processing.
FIG. 14 is a diagram illustrating a conventional parallax image input method.
[Explanation of symbols]
21 light source, 22 shutter, 30 LCD, 24, 31 cylindrical lens, 32 hologram recording medium, 33 recording medium feeding mechanism, 51 parallax image input unit, 52 viewpoint conversion processing unit, 53 hologram exposure unit, 61 CCD video camera, 71 JPEG compression board, 72 viewpoint conversion processing CPU, 73 RAM, 81 exposure head control CPU, 82 exposure head, 83 LCD driver

Claims (10)

In an image data generation device for generating image data for creating a planar holographic stereogram,
A taking-in means for taking in a parallax image of the subject by rotating so that the optical axis always faces the subject while moving on a straight line;
Scaling means for scaling the parallax image captured by the capturing unit for each pixel column in a pixel column direction corresponding to the rotation angle of the capturing unit and the angle from the optical axis;
Mapping means for mapping the parallax image scaled by the scaling means as an image on a plane parallel to the subject;
Display means for displaying an image mapped by the mapping means;
An image data generation apparatus comprising: a creation unit that creates a holographic stereogram using light from an image displayed on the display unit as object light . The scaling means includes
An enlargement / reduction ratio calculation means for obtaining an enlargement / reduction ratio corresponding to a rotation angle of the acquisition means and an angle from the optical axis, for each pixel column, the parallax image captured by the capture means;
  Enlarging / reducing means for enlarging / reducing the parallax image corresponding to the rotation angle and the angle from the optical axis for each pixel column using the enlargement / reduction ratio calculated by the enlargement / reduction ratio calculating means; Further comprising The image data generation device according to claim 1. The mapping means includes
By mapping the coordinate system of the pixel column data of the parallax image scaled by the scaling means, it is mapped as an image on a plane parallel to the subject. The image data generation device according to claim 1.   The image data generation device according to claim 1, further comprising conversion means for performing viewpoint conversion processing for localizing a reproduced image on a hologram surface. The image data generation apparatus according to claim 4 , wherein the viewpoint conversion process is performed by generating a new image from a pixel array of a plurality of captured parallax images. In an image data generation method for generating image data for creating a planar holographic stereogram,
A capturing step of capturing a parallax image of the subject by rotating so that the optical axis always faces the subject while moving on a straight line;
A scaling step for scaling the parallax image captured by the processing of the capturing step in the pixel column direction corresponding to the rotation angle of the capturing unit and the angle from the optical axis for each pixel column;
A mapping step for mapping the parallax image scaled by the processing of the scaling step as an image on a plane parallel to the subject;
A display step of displaying an image mapped by the mapping step;
And a creation step of creating a holographic stereogram using light from the image displayed by the processing of the display step as object light . The processing of the scaling step is
An enlargement / reduction ratio calculation step for obtaining an enlargement / reduction ratio corresponding to the rotation angle of the acquisition means and the angle from the optical axis for each pixel column of the parallax image acquired by the acquisition means;
  Using the enlargement / reduction ratio calculated by the processing of the enlargement / reduction ratio calculation step, enlargement / reduction of the parallax image corresponding to the rotation angle and the angle from the optical axis is performed for each pixel column. And further comprising a reduction step The image data generation method according to claim 6. The mapping step process is as follows:
By mapping the coordinate system of the pixel column data of the parallax image scaled by the scaling means, it is mapped as an image on a plane parallel to the subject. The image data generation method according to claim 6.   The image data generation method according to claim 6, further comprising a conversion step of performing a viewpoint conversion process for localizing the subject in the vicinity of the hologram surface. The image data generation method according to claim 9, wherein the viewpoint conversion process is performed by generating a new image from a pixel array of a plurality of captured parallax images.

Images