WO2012117619A1

WO2012117619A1 - 3d imaging device

Info

Publication number: WO2012117619A1
Application number: PCT/JP2011/075739
Authority: WO
Inventors: 岩崎　洋一
Original assignee: 富士フイルム株式会社
Priority date: 2011-03-02
Filing date: 2011-11-08
Publication date: 2012-09-07
Also published as: JP2014103427A

Abstract

Provided is a low-cost, small-form-factor, single-lens 3D imaging device that can image a subject from four different viewpoints. Said imaging device is provided with: a single imaging lens (12); a beam-splitting means (24), placed in the path of incident light that comes from the subject and passes through the imaging lens (12), that splits said incident light along an arbitrary boundary line perpendicular to the optical axis; first and second solid-state imaging elements (30 and 33) that receive the split incident light beams, respectively; and an image-processing means (42) that generates 3D image data constituting a 3D image of the abovementioned subject by performing image processing on output signals from the solid-stage imaging elements (30 and 33). In each solid-stage imaging element (30 and 33), a plurality of pixels are arranged in a two-dimensional array, each two adjacent pixels are paired together, and the two pixels in each pair exhibit parallax about a boundary line perpendicular to the abovementioned boundary line.

Description

Stereo imaging device

The present invention relates to a stereoscopic image capturing apparatus, and more particularly to a stereoscopic image capturing apparatus capable of performing left and right parallax separation and upper and lower parallax separation by a monocular method.

Television receivers that can display stereoscopic images (3D images) have become widespread, and digital cameras (stereoscopic imaging devices) that can capture stereoscopic images of subjects have also shown signs of widespread use.

A conventional stereoscopic image capturing apparatus has a two-lens system in which two photographing lens systems arranged in the horizontal direction are mounted on the front surface of a camera housing as described in, for example, Patent Document 1 below. The left taking lens system corresponds to a human right eye, and the right taking lens system corresponds to a human left eye. The left and right photographing lens systems are provided with a distance of about 6.5 cm, which is the distance between the left and right eyes of a human.

Since such a binocular stereoscopic image capturing apparatus captures a subject image for the left eye and a subject image for the right eye through separate photographing lens systems separated by 6.5 cm, the degree of parallax separation between the left and right sides It is possible to shoot a subject image with high.

However, since the two-lens stereoscopic image pickup apparatus includes two expensive photographing lens systems, there is a problem that the product cost increases.

Therefore, as described in Patent Document 2 below, a monocular stereoscopic image capturing apparatus has been proposed. This stereoscopic image pickup apparatus is equipped with one photographing lens system, and converts incident light from a subject condensed through the photographing lens system into parallel light by passing through a relay lens.

Then, as shown in FIG. 14, the parallel light 1 obtained through the relay lens is separated into right and left by a light splitting mirror 4 in which two mirrors 2 and 3 are abutted at right angles, and incident light reflected by the mirror 2 is reflected. Is reflected by the mirror 5 to form an image on the image sensor 6. Incident light reflected by the mirror 3 is reflected by the mirror 7 and forms an image on the image sensor 8.

Since the photographing lens system is provided on the light incident side of the relay lens that emits the parallel light 1, the incident light from the object field is reversed left and right in the photographing lens system, and the image sensor 6 is passed through the left eye. The viewed image is formed, and an image viewed through the right eye is formed on the image sensor 8.

Japanese Unexamined Patent Publication No. 2008-187385 Japanese Unexamined Patent Publication No. 2010-81580

However, the monocular stereoscopic image capturing apparatus (camera) described in Patent Document 2 can capture a stereoscopic image when a subject is imaged with the camera placed horizontally, but the camera is used to capture a vertically long subject. If a subject is imaged with the vertical position, a stereoscopic image cannot be captured.

An object of the present invention is to provide a stereoscopic image capturing apparatus that is monocular and that can capture a stereoscopic image of a subject even when the subject is imaged horizontally or the subject is imaged vertically. is there.

The stereoscopic image pickup apparatus of the present invention is a monocular photographing lens and is placed on an optical path of incident light from a subject incident through the photographing lens, and uses the incident light as an arbitrary vertical line with respect to the optical axis. The light splitting unit for splitting, the first and second solid-state imaging devices that respectively receive the split incident light of the split incident light, and the output signals of the first and second solid-state imaging devices as images An image processing unit configured to generate stereoscopic image data of the subject to be processed to form a stereoscopic image, and the first and second solid-state imaging elements each include a plurality of pixels arranged in a two-dimensional array. Each of the two adjacent pixels is set as a pair pixel, and one pixel and the other pixel of each pair pixel have a parallax with a boundary line orthogonal to the vertical line as a boundary.

The stereoscopic image pickup apparatus of the present invention divides incident light from a subject incident through a monocular photographing lens into, for example, left and right parts, and receives each divided incident light by each solid-state image pickup element. For example, the phase difference pixel that receives light by separating the parallax in the vertical direction can acquire image data constituting a stereoscopic image regardless of whether the camera is placed horizontally or vertically.

1 is an external perspective view of a stereoscopic image capturing apparatus according to an embodiment of the present invention. It is a functional block block diagram of the stereo image imaging device shown in FIG. It is a surface schematic diagram of the solid-state image sensor shown in FIG. It is a figure which shows the incident angle sensitivity characteristic of the solid-state image sensor shown in FIG. FIG. 4 is an explanatory diagram of parallax separation in the vertical direction in the solid-state imaging device shown in FIG. 3. It is explanatory drawing of the parallax separation in the stereo image imaging device of FIG. It is an effect explanatory view of parallax separation means. It is explanatory drawing of the light non-transmission area | region of a parallax separation means. It is explanatory drawing of another embodiment of a parallax separation means. It is explanatory drawing of another embodiment of a parallax separation means. It is explanatory drawing of the solid-state image sensor replaced with FIG. FIG. 7 is an explanatory diagram of another embodiment of parallax separation instead of FIG. 6. FIG. 4 is a surface schematic diagram of a solid-state imaging device instead of FIG. 3. It is explanatory drawing of the conventional monocular type stereo image imaging device.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is an external perspective view of a digital camera capable of capturing a stereoscopic image according to an embodiment of the present invention. This digital camera 10 is provided with a monocular photographing lens 12 on the front surface of a rectangular casing 11. The photographic lens 12 is disposed in a lens barrel 13 provided in the casing 11 so as to be retractable, and a shutter release button 14 is provided on the right shoulder of the casing 11.

FIG. 2 is a functional block configuration diagram of the digital camera 10 shown in FIG. A lens barrel 13 that houses the taking lens 12 is provided. In addition to the taking lens 12, the lens barrel 13 houses a focus positioning lens, a telephoto lens, and the like.

A relay lens 21 is provided on the back of the lens barrel 13, and incident light collected by the photographing lens 12 or the like is converted into parallel light 22 by passing through the relay lens 21.

A parallax separating means 23 and a light splitting mirror 24 are provided in the optical path of the parallel light 22. The parallax separation means 23 to be described in detail later is configured by a liquid crystal shutter in the present embodiment. The light splitting mirror 24 is configured by abutting the front edges of the two

mirrors

25 and 26. It is preferable to arrange a diaphragm for controlling the F value at the front stage or the rear stage of the parallax separation means 23. The parallax separation means 23 is provided to perform the parallax separation well, and is not necessarily required.

The mirror 25 is provided with an inclination of 45 degrees to the right with respect to the parallel light 22, and the mirror 26 is provided with an inclination of 45 degrees with respect to the parallel light 22. And the mirror 25 and the mirror 26 are joined so that the respective leading edges thereof are abutted, and the leading edge joining edge 27 is provided so as to be perpendicular to the bottom surface of the housing 11 of FIG. As a result, the parallel light 22 when viewed from the optical axis direction is divided into two with the front end joint edge 27 as a boundary line, the left half is reflected by the mirror 26 to the left in the horizontal direction, and the right half is reflected by the mirror 25 in the horizontal direction. Reflected to the right.

A mirror 28 is provided on the reflecting surface of the mirror 25 so as to be slightly separated from each other. Incident light reflected by the mirror 28 passes through a condensing lens 29 and forms an image on the light receiving surface of the solid-state imaging device 30.

Similarly, a mirror 31 is provided on the reflecting surface of the mirror 26 so as to be slightly separated from each other. The incident light reflected by the mirror 31 passes through the condenser lens 32 and forms an image on the light receiving surface of the solid-state imaging device 33. The

mirrors

28 and 31 may be omitted, and the reflected light of the

mirrors

25 and 26 may be received by the

condensing lenses

29 and 32 and the solid-

state imaging devices

30 and 33 provided in the direction of the reflected light.

The electric control system of the digital camera 10 includes a central control unit (CPU) 40 that performs overall control of the entire digital camera 10, an operation unit (including the shutter release button 14) 41 that receives operation instructions from a user, and image processing. Unit 42, an encoder 44 that encodes image data processed by the image processing unit 42 into display data, a driver 46 that displays the display data on the display unit 45, a main memory 47, and a write / read of the memory card 48. A media control unit 49 that performs read control and a bus 50 that interconnects them are provided.

Analog signal processing units (AFE) 34 and 35 and analog / digital (A / D)

converters

36 and 37 are connected to the solid-

state imaging devices

30 and 33, respectively. Captured image signals from the solid-

state imaging devices

30 and 33 converted into digital signals by the A /

D converters

36 and 37 are input to the bus 50. Note that the

AFEs

34 and 35 and the A /

D converters

36 and 37 may be integrated into one unit and used by switching.

A device control unit 51 is connected to the CPU 40. The device control unit 51 controls the photographing lens 12 including the focus alignment lens and the telephoto lens based on an instruction from the CPU 40, and the parallax separation unit 23, the solid-

state imaging devices

30 and 33,

AFEs

34 and 35, A /

D converters

36 and 37 are controlled.

FIG. 3 is a schematic view of the surface of the solid-

state imaging device

30, 33 shown in FIG. The solid-

state imaging devices

30 and 33 may be of a CCD type or a CMOS type or other type as a signal readout format. In the solid-state imaging device 30 (33) shown in FIG. 3, pixels (photoelectric conversion elements: photodiodes) are arranged in a square lattice pattern.

In each odd-numbered row of pixels 52, three primary color filters RGB are arranged in a Bayer array, and in each even-numbered row of pixels 52, three primary color filters rgb are arranged in a Bayer array. RGBrgb described in each pixel indicates the color of red (R, r), green (G, g), and blue (B, b) color filters, and R = r, G = g, B = b. It is. Two pixels of the same color arranged in the vertical direction constitute a pair pixel that performs parallax separation.

The upper surface of each pixel 52 is covered with a light shielding film, and an opening is provided in the light shielding film on each pixel 52. The light shielding film of each pixel 52 in the odd-numbered row is provided with an opening 52x that is eccentric in the upward direction in the drawing, and the light shielding film of each pixel 52 in the even-numbered row is provided with an opening 52y that is eccentric in the downward direction. Yes.

FIG. 4B of FIG. 4 is a diagram showing the incident angle sensitivity characteristic in the direction perpendicular to the bottom surface of the housing 11 of the solid-state imaging device 30 (33) shown in FIG. The light receiving sensitivity TD of the odd-numbered pixels of the solid-state imaging device 33 is shifted to the left, and the light receiving sensitivity TU of the even-numbered pixels is shifted to the right.

This left / right shift becomes a parallax in a direction perpendicular to the bottom surface of the housing 11, and a captured image obtained from the odd-numbered pixels of the solid-state image sensor 33 is obtained as the right-eye image from the even-numbered pixels of the solid-state image sensor 33. By using the captured image as the left-eye image, it is possible to reproduce a stereoscopic image of the subject.

FIG. 4A of FIG. 4 combines the incident angle sensitivity characteristic in the direction parallel to the casing 11 of the solid-state image sensor 30 and the incident angle sensitivity characteristic in the direction parallel to the casing 11 of the solid-state image sensor 33 shown in FIG. FIG.

The sensitivity distribution TL with respect to the incident angle of the light of the solid-state imaging device 30 with respect to the incident light reflected by the mirror 25 among the incident light that has become the parallel light 22 by the relay lens 21 is a distribution shifted to the right side as shown in FIG. 4A. It has become. On the contrary, the sensitivity distribution TR with respect to the incident angle of the light of the solid-state imaging device 33 with respect to the incident light reflected by the mirror 26 is a distribution shifted to the left side.

This left / right misalignment becomes a parallax in a direction parallel to the bottom surface of the housing 11, the captured image obtained from the solid-state image sensor 30 is the right-eye image, and the captured image obtained from the solid-state image sensor 33 is the left-eye image. Thus, it is possible to reproduce a stereoscopic image of the subject.

FIG. 5 is an explanatory diagram of the parallax of the solid-state imaging device 30 (33) shown in FIG. The captured image of the odd-numbered row pixels in which the subject is imaged through the light-shielding film opening 52x that is eccentric in the upward direction and the captured image of the even-numbered pixels in which the same subject is imaged through the light-shielding film opening 52y that is eccentric in the downward direction Due to the eccentricity of 52x and 52y, the image has a parallax shift in the vertical direction. In the present embodiment, a stereoscopic image separated in the vertical direction is picked up by using this parallax shift. That is, when the camera is placed vertically, the vertical parallax shift in FIG. 5 becomes a horizontal parallax shift of a person standing on the ground, and a stereoscopic image can be captured.

FIG. 6 is an explanatory diagram of four viewpoints when the left and right parallax separation by the light dividing mirror 24 and the vertical parallax separation by the eccentricity of the light

shielding film openings

52x and 52y are performed. When the subject 53 is viewed through a monocular lens, the center of the subject 53 is seen. However, in the stereoscopic image capturing apparatus 10 of the present embodiment, the incident light from the subject is divided into the left and right when viewed from the optical axis direction by the light dividing mirror 24, and the viewpoint is shown in FIG. Similarly, it is divided into left and right L and R.

Since the incident light divided into the left and right is incident on the solid-state imaging device 30 (33) described with reference to FIG. 3, the incident light viewed from the viewpoint of the left side L is now a light shielding film opening 52x decentered vertically. 52y, and as shown in FIG. 6B, the incident light that is separated into the upper and lower viewpoints A and B and viewed from the right R viewpoint is the upper and lower viewpoints C and D. Will be separated.

That is, in the stereoscopic image capturing apparatus 10 shown in FIG. 2, when incident light from a subject whose left and right and upside down are reversed by the photographic lens 12 is picked up by the solid-state

image pickup devices

30 and 33, the viewpoint A ( An image in which the subject 53 is viewed from the perspective of viewing the subject diagonally from the upper left) is captured, and an image of the subject 53 viewed from the viewpoint B (viewpoint of viewing the subject from the diagonally lower left) is captured by the odd row pixels of the solid-state imaging device 30. Then, an image of the subject 53 viewed from the viewpoint C (viewpoint viewing the subject from diagonally right above) is captured by the even-numbered pixels of the solid-state image sensor 33, and the viewpoint D (subject diagonally rightward by the odd-numbered pixels of the solid-state image sensor 33 is captured. An image in which the subject 53 is viewed from the viewpoint (viewed from below) is captured.

Captured image data of the solid-

state imaging devices

30 and 33 is taken into the main memory 47, and is subjected to well-known image processing such as offset correction, gamma correction, RGB / YC conversion processing and the like in the image processing unit 42 and data compression in JPEG format. And stored in the memory card 48. At the time of saving, the four pieces of image data on the left, right, upper and lower sides are saved in association with each other. For example, it is stored in the MPO format that is a standard of the Camera and Imaging Products Association (CIPA).

When a stereoscopic image is reproduced from four pieces of image data when the subject 53 is photographed with the camera placed horizontally, an image obtained by adding the image data of the viewpoints A and B in FIG. The image obtained by adding the image data of the viewpoints C and D may be used as the right eye image. Further, when the subject is photographed with the camera placed vertically, the position of the viewpoint ABCD in FIG. 6B is rotated by 90 degrees counterclockwise, and the addition data of the viewpoints C and A and the addition of the viewpoints B and D are obtained. It is possible to generate a stereoscopic image of a standing image of the subject 53 with the data.

Alternatively, in the state of the viewpoint ABCD in FIG. 6B, a stereoscopic image of the subject 53 is generated with the addition data of the viewpoints C and A and the addition data of the viewpoints B and D. In this case, when the user views the stereoscopic image displayed on the television receiver or the personal computer monitor while lying down (a state where the left and right eyes are arranged up and down), stereoscopic viewing is possible.

The principle of separating the vertical and horizontal parallaxes has been described with reference to FIG. However, if the vertical and horizontal misalignment, that is, the parallax is not sufficient, it is difficult to visually recognize the left eye image and the right eye image as a stereoscopic image. In FIG. 4, stereoscopic viewing becomes more difficult as the ratio of the shaded area (overlapping area) where the sensitivity distribution TD and the sensitivity distribution TU overlap each sensitivity distribution TD, TU increases. In particular, in the case of the monocular system, in order to capture a stereoscopically viewable image, it is preferable to reduce the ratio of the overlapping area to each sensitivity distribution.

Therefore, in this embodiment, the parallax separation means 23 is provided. With this parallax separating means 23, as shown in FIG. 4, incident light at an incident angle of 0 degree and the vicinity thereof is cut (shielded) so as not to enter the light splitting mirror 24. As a result, as shown in FIG. 7B in FIG. 7, the incident angle near 0 degrees is cut in the vertical direction, the ratio of the hatched area (overlapping area) to the sensitivity distributions TU and TD is reduced, and the vertical parallax separation is good. It becomes. Further, as shown in FIG. 7A, the vicinity of the incident angle of 0 degree is cut in the left-right direction, the sensitivity becomes zero, and the parallax separation in the left-right direction becomes good.

FIG. 8 is an explanatory diagram of the parallax separation means 23. The parallax separating means 23 placed in the front stage of the light splitting mirror 24 having a V-shaped cross section is composed of, for example, a liquid crystal shutter whose light incident surface is erected perpendicular to the optical axis. In this liquid crystal shutter, the ridgeline which becomes the boundary line between the left and right parallaxes of the light dividing mirror 24 is covered with the required width x, and the upper and lower parallax separation boundaries of the solid-

state imaging devices

30 and 33 are covered with the required width x. A cross-shaped (cross-shaped) light non-transmissive region 61 is electrically formed.

The transparent region of the square liquid crystal shutter is completely divided into four

regions

23a, 23b, 23d, and 23c by the light non-transmissive region 61, and each region 23a to 23d corresponds to the viewpoint ABDC.

The width x of the light non-transmissive region 61 may be a fixed value, but preferably the width x is variably controlled according to the photographing conditions. For example, if the width x is too wide, the light receiving sensitivity is lowered as can be seen from FIG. 7, so the width x is determined in consideration of the brightness of the shooting scene. That is, when the shooting scene is dark, if the light non-transmissive region 61 is wide, only a dark image is captured. Therefore, the width x is narrowed, and when the shooting scene is bright, the width x is wide.

In addition, when the photographic lens 12 has a short focal length, such as a wide-angle lens, it is difficult to separate the parallax. Therefore, it is easy to separate the parallax by widening the width x of the light non-transmissive region 61, and the focal point is like a telephoto lens. Conversely, when the distance is long, the width x is narrowed.

Furthermore, the width x of the light non-transmissive region 61 may be variably controlled in relation to the F value. When the F value is small (when the aperture is open), the scene is often dark, and when the F value is large (when the aperture is narrow), the scene is often bright. The width x of the light non-transmissive region 61 is controlled. That is, when the F value is large, the sensitivity is not lowered even if the width x is wide because the scene is bright, so the width x is widened to increase the degree of parallax separation.

In the embodiment shown in FIG. 2, the parallax separation means 23 is disposed immediately before the light splitting mirror 24 to shield the light splitting boundary portion of the light splitting mirror 24. It is not limited to just before the light splitting mirror 24. For example, a diaphragm is also arranged near the focal position of the photographing lens 12 that collects incident light. However, a parallax separation unit may be provided at the position of the diaphragm. Also by providing the parallax separating means at this position, it becomes possible to equally divide the incident light into four in the horizontal and vertical directions with a small area liquid crystal shutter.

FIG. 9 is a perspective view of the parallax separation means 63 (also used as the light splitting mirror 24) according to another embodiment. Instead of the reflecting mirrors 25 and 26 of the light splitting mirror 24, electrochromic mirrors 65 and 66 whose reflectance can be partially changed by electric control are used, and reflection of the ridge line portion of the tip joint edge 67 of the

mirrors

65 and 66 is reflected. Change the rate with the required width. Preferably, the reflectance is 0%. The portion having the reflectance of 0% becomes the parallax separation means of this embodiment. Similarly, the reflectance of the center line portion 68 orthogonal to the tip joint edge 67 of the

mirrors

65 and 66 is changed to a required width.

It is preferable that this width can be variably controlled. As in the embodiment of FIG. 8, this makes it possible to obtain the same effect as when the region along the boundary line for parallax separation in the incident light is shielded with the required width, and the overlapping of the upper and lower sensitivity distributions TD and TU. It is possible to cut most of the region and most of the boundary between the left and right sensitivity distributions TR and TL.

FIG. 10 is a perspective view of the parallax separation means 71 according to still another embodiment. In the present embodiment, the

mirrors

65 and 66 shown in FIG. 9 are constituted by four

mirrors

69a, 69b, 69c, and 69d from which the reflectance changing portion is removed, and the reflecting mirror is moved between the reflecting mirrors 69a to 69d. It can be expanded by the mechanism. The width of the gap 69 formed by widening the space between the reflecting surfaces 69a to 69d is variably controlled.

By doing in this way, it is possible to prevent the incident light on the boundary line dividing the incident light when viewed from the optical axis direction into the left, right, up and down, from being incident on the solid-

state imaging devices

30 and 33, and the parallax. The left / right separation and the upper / lower separation can be performed satisfactorily. In addition, by providing the third solid-state imaging device 68 that receives incident light transmitted through the gap 69, the solid-state imaging device 68 can capture a two-dimensional image (planar image) of the subject.

As described above, according to the above-described embodiment, it is possible to obtain image data of a subject image viewed from at least four viewpoints. Even if the subject is photographed horizontally, the subject can be photographed vertically. It is also possible to reproduce a stereoscopic image of the subject. In addition, since it is a monocular system and can shoot subject images of four viewpoints only by using two solid-state imaging devices, it is possible to reduce the size and cost of the camera.

In the embodiment shown in FIG. 2, the incident light is separated from the left and right by the mirror 24 and the parallax is separated from the upper and lower by the solid-

state imaging devices

30 and 33, but the reverse is also possible. That is, the arrangement of the mirror 24 is rotated by 90 degrees, and the incident light is vertically separated by the mirror 24, and the solid-state

image pickup devices

30 and 33 of FIG. The viewpoint may be divided into four.

In the embodiment of FIG. 3, the parallax separation is performed by decentering the position of the light shielding film opening. However, as shown in FIG. 11, one microlens 83 is provided for each pair of

pixels

81 a and 81 b that perform two parallax separations. May be used to perform parallax separation (pupil division). Furthermore, the position of the 4-viewpoint ABCD can be set to the vertical and horizontal positions with respect to the subject as shown in FIG. In this case, this is possible by tilting the tip joint edge 27 of the mirror 24 of FIG. Furthermore, the pixel arrangement of the solid-

state imaging devices

30 and 33 is not limited to a square arrangement, and as shown in FIG. 13, even-numbered pixel rows are shifted by ½ pixel pitch with respect to odd-numbered pixel rows. A so-called honeycomb pixel arrangement may be used. In the case of FIG. 13, each pixel is indicated by a circle,

symbols

84x and 84y are light shielding film openings, and pixels adjacent obliquely are paired pixels (parallax separation pixels: phase difference pixels).

As described above, the stereoscopic image capturing apparatus according to the present embodiment is placed on the optical path of incident light from a monocular photographing lens and a subject incident through the photographing lens, and forms an arbitrary vertical line with respect to the optical axis. A light dividing unit that divides the incident light as a boundary line, first and second solid-state imaging devices that respectively receive the divided incident lights of the divided incident light, and the first and second solids An image processing unit configured to perform image processing on an output signal of the imaging device to generate stereoscopic image data of the subject that forms a stereoscopic image, and the first and second solid-state imaging devices include a plurality of pixels in a two-dimensional array Are formed as a pair pixel, and one pixel and the other pixel of each pair pixel have a parallax with a boundary line orthogonal to the vertical line as a boundary. It is characterized by.

In addition, the stereoscopic image capturing apparatus according to the embodiment includes a parallax separation unit that blocks incidence of incident light on the two boundary lines having a cross shape into the first and second solid-state imaging elements. To do.

In addition, the stereoscopic image capturing apparatus according to the embodiment includes a control unit that controls the blocking width on the boundary line.

Further, the control unit of the stereoscopic image capturing apparatus according to the embodiment is characterized in that the blocking width is adjusted according to a shooting condition.

Further, the control unit of the stereoscopic image capturing apparatus of the embodiment widens the width as the F value is smaller, the photographing scene is brighter, the focal length of the photographing lens is shorter, or the distance to the main subject is shorter, The width is narrowed as the F value is larger, the photographing scene is darker, the focal length of the photographing lens is longer, or the distance to the main subject is longer.

Further, the parallax separation unit of the stereoscopic image pickup device of the embodiment is configured by a liquid crystal shutter placed in front of the light splitting unit, and the light non-transmission region formed in the cross shape in the center of the liquid crystal shutter. The incident light on the boundary line is cut.

Further, the light splitting unit of the stereoscopic image capturing apparatus according to the embodiment is configured by abutting the leading edges of two mirrors opened at 90 degrees.

Further, the light dividing unit and the parallax separating unit of the stereoscopic image capturing apparatus of the embodiment are integrally formed, and the light dividing unit is configured by abutting the front end edges of two mirrors opened at 90 degrees, The parallax separation unit is configured by controlling the reflectance of the front edge of the two mirrors and the center line of each mirror perpendicular to the front edge over a required width.

In addition, the light dividing unit and the parallax separating unit of the stereoscopic image capturing apparatus of the embodiment are integrally formed, and the light dividing unit is configured by arranging two mirrors opened at 90 degrees in two stages. The parallax separation unit is configured by a gap formed between the mirrors.

According to the embodiment described above, it is possible to capture a stereoscopic image of a subject even when the subject is imaged with the camera placed horizontally or the subject is imaged with the camera placed vertically. In addition, since the incident light is divided into two by using the pixels of the solid-state imaging device as phase difference pixels, the camera can be reduced in size and weight.

The stereoscopic image capturing apparatus of the present invention can capture a subject image from four viewpoints with a monocular system, and can capture a stereoscopic image regardless of whether the camera is placed vertically or horizontally, at low cost. It is useful when applied to a small stereoscopic image capturing apparatus.
Although the present invention has been described in detail and with reference to specific embodiments, it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the spirit and scope of the invention.
This application is based on a Japanese application (Japanese Patent Application No. 2011-45546) filed on Mar. 2, 2011, the contents of which are incorporated herein by reference.

10 Digital camera (stereoscopic imaging device)
DESCRIPTION OF SYMBOLS 11 Case 12 Shooting lens 21 Relay lens 22

Parallel light

23, 63 Parallax separation means 24 Light splitting mirror 27 Front

joint edge

30, 33 Solid-

state image sensor

52x, 52y Light shielding film opening 40 of paired pixel CPU
42 Image Processing Unit 51 Device Control Unit 61 Light Non-Transparent Area 72 Gap

Claims

A monocular photographic lens,
A light dividing unit that is placed on the optical path of incident light from the subject incident through the photographing lens and divides the incident light with an arbitrary vertical line as a boundary line with respect to the optical axis;
First and second solid-state imaging devices that respectively receive the split incident lights of the split incident light;
An image processing unit that performs image processing on output signals of the first and second solid-state imaging devices to generate stereoscopic image data of the subject that forms a stereoscopic image;
In the first and second solid-state imaging devices, a plurality of pixels are arrayed in a two-dimensional array, and two adjacent pixels among the pixels are paired, and one pixel of each paired pixel and the other A stereoscopic image capturing apparatus configured to have parallax with respect to a boundary line perpendicular to the vertical line.
The stereoscopic image capturing apparatus according to claim 1,
A stereoscopic image capturing apparatus including a parallax separation unit that blocks incidence of incident light on two boundary lines having a cross shape into the first and second solid-state imaging elements.
The stereoscopic image capturing apparatus according to claim 2,
A stereoscopic image capturing apparatus including a control unit that controls the blocking width on the boundary line.
The stereoscopic image capturing apparatus according to claim 3,
The control unit is a stereoscopic image capturing device that adjusts the blocking width according to a shooting condition.
The stereoscopic image capturing apparatus according to claim 4,
The controller increases the width as the F value is smaller or the photographing scene is brighter or the focal length of the photographing lens is shorter, and as the F value is larger or the photographing scene is darker or the focal length of the photographing lens is longer. A stereoscopic image capturing apparatus that narrows the width as much as possible.
The stereoscopic image capturing apparatus according to any one of claims 2 to 5,
The parallax separation unit is composed of a liquid crystal shutter placed in front of the light splitting unit, and cuts the incident light on the boundary line in a light non-transmissive region formed in the cross shape in the center of the liquid crystal shutter. Imaging device.
The stereoscopic image capturing apparatus according to any one of claims 1 to 6,
The light splitting unit is a stereoscopic image pickup device configured by abutting the leading edges of two mirrors opened at 90 degrees.
The stereoscopic image capturing apparatus according to any one of claims 2 to 5,
The light splitting part and the parallax separating part are integrally formed, and the light splitting part is configured by abutting the front end edges of two mirrors opened at 90 degrees, and the front end edges of the two mirrors and the front end A stereoscopic image capturing apparatus in which the parallax separation unit is configured by performing variable reflectance control of a center line of each mirror orthogonal to an edge over a required width.
The stereoscopic image capturing apparatus according to any one of claims 2 to 5,
The light splitting unit and the parallax separation unit are integrally formed, and the light splitting unit is configured by arranging two mirrors opened at 90 degrees in two stages, and a gap formed between a total of four mirrors. A stereoscopic image capturing apparatus in which the parallax separation unit is configured.