JP2014109739A - Stereo camera - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a stereo camera capable of adjusting depth representation with a simple arrangement.SOLUTION: The stereo camera includes: an aperture constituted of a transparent liquid crystal panel which is disposed between an objective lens and a lens at the imaging element side, which are co-axial each other; and an image sensor disposed in front of a focal point of the lens at the imaging element side. The image sensor has an arrangement that right photoelectric conversion parts and left photoelectric conversion parts are disposed alternately; the liquid crystal panel includes a pair of right and left openings formed therein; and a lenticular lens for focusing beams of light passing through the openings on the respective right and left photoelectric conversion parts. By changing the distance between the openings by means of the liquid crystal panel, a pair of right and left symmetrical openings with respect to optical axis of the lens achieves an image with rich depth representation with base line length which is changed by the stereo camera can be obtained. Thus, no right and left lenses are required, and stereoscopic visions can be obtained with a single eye. The stereo cameras can be reduced in size.

Description

  The present invention relates to a stereo camera.

  Conventionally, a stereo camera is known in which an object is photographed simultaneously from a plurality of different directions, and information in the depth direction can be recorded. On the other hand, in endoscopic surgery, by taking a stereoscopic image, treatment can be performed while looking at an image with a sense of depth that cannot be obtained with a two-dimensional endoscope, so a stereoscopic image can be taken with an endoscope. There are various types of such devices (see, for example, Patent Document 1).

JP 2010-128354 A

  In a stereoscopic image, in order to increase a sense of depth and a sense of distance according to the distance of the subject, it is preferable to increase the baseline length that is the distance between the objective lenses. In the above-mentioned Patent Document 1, a pair of optical systems (left and right lenses) constituting two optical paths are provided in the lens front group, and a time division shutter for switching the two optical paths in a time division manner is provided on the lens rear group side. The base length can be changed by providing a prism, but the distance between the pair of apertures corresponding to each optical path of the time-division shutter is fixed, and only the incident angle to the image sensor changes depending on the rear lens group. That's right. This can cause a shift of each image due to the two optical paths, but cannot change the sense of depth or the like by changing the baseline length.

  Also, the lens front group can be exchanged, and a plurality of optical systems with parallax suitable for the observation purpose are prepared, and the lens front group is exchanged according to the size of the parallax and the distance of the subject. You can do that. However, there is a problem that the work for exchanging the front lens group increases and the usability is poor.

  The present invention has been devised to solve such problems of the prior art, and its main object is to provide a stereo camera capable of adjusting the sense of depth with a simple configuration. .

  A stereo camera according to the present invention includes an imaging device including a plurality of photoelectric conversion units, a condensing lens for condensing light from a subject toward the imaging device, and one of light from the subject to the imaging device. An aperture member that allows a portion to pass therethrough, and the aperture member is disposed between a pair of apertures arranged on the left and right with respect to the optical axis of the condenser lens, and between the apertures that can change a distance between the pair of apertures. It is set as the structure which has a distance variable means.

  According to the present invention, since the distance between the pair of left and right openings arranged with respect to the optical axis of the condenser lens can be changed, the distance between the pair of openings becomes the baseline length in the stereo camera, and The base line length can be changed according to the perspective, thereby forming an image that can provide a stereoscopic effect according to the distance to the subject. In addition, it is only necessary to provide a pair of openings on the left and right with respect to the optical axis of the lens, and it is not necessary to provide separate lenses on the left and right. Further, by performing optimal aperture correction according to the position and shape of the opening, it is possible to acquire a high-definition stereo image regardless of the position and shape of the opening. Further, normal imaging and stereo imaging can be easily switched by controlling the aperture and the image sensor.

Schematic principal part sectional drawing which shows the example of the endoscope to which this invention was applied It is a front view which shows an example of an aperture, (a) shows the opening position in case base length is standard length, (b) shows the opening position in case base length is short, (c) shows base length Diagram showing opening position when long Explanatory drawing which shows the optical path by this invention Block diagram showing the configuration of the control circuit (A) is a table | surface which shows the relative relationship of the baseline length and aperture correction amount according to the difference in a feeling of depth, (b) is an example of MTF correction | amendment when the shape of an opening and the circumferential direction of a lens and the radial direction are considered. Table showing (A) shows the relationship between the MTF characteristics and aperture correction in the central part, the peripheral part, and the intermediate part thereof in the radial direction of the lens, and (b) shows the relationship between the radial direction and the circumferential direction of the lens. The difference in MTF characteristics is shown, (c) is a diagram showing the difference in MTF characteristics due to the difference in the shape of the opening. (A) shows the case where the opening is elliptical, (b) shows the case where the opening is substantially rhombus, and (c) shows the case where the opening is formed by a chord and an arc. (A) shows the light quantity distribution when passing through the right opening, and (b) shows the light quantity distribution when passing through the left opening. The figure which shows the relationship between correction | amendment and the light quantity distribution on either side of FIG. (A) shows a second example of an aperture, and (b) shows a third example. The 4th example of an aperture is shown, (a) is a typical side view, (b) is a perspective view showing an important section. The figure which shows the 5th example of an aperture The figure which shows the 6th example of an aperture The figure which shows the 7th example (time division opening) of an aperture Timing chart showing an example of control in a time-division opening The figure which shows the 8th example (time division opening) of an aperture The figure corresponding to FIG. 1 which shows another example of whole structure

  A first invention made to solve the above problems includes an imaging device including a plurality of photoelectric conversion units, a condensing lens for condensing light from a subject toward the imaging device, and the subject. An aperture member that allows a part of the light reaching the image sensor to pass therethrough, and the aperture member is disposed between the pair of openings disposed on the left and right with respect to the optical axis of the condenser lens, and between the pair of openings. It is set as the structure which has an inter-opening distance variable means which can change a base line length.

  According to this, since it is possible to change the distance between the pair of left and right apertures with respect to the optical axis of the condenser lens, the distance between the pair of apertures becomes the base line length in the stereo camera, and the distance to the subject is reduced. Accordingly, the base line length can be changed, whereby an image can be formed in which a stereoscopic effect corresponding to the distance to the subject is obtained. In addition, it is only necessary to provide a pair of openings on the left and right with respect to the optical axis of the lens, and it is not necessary to provide separate lenses on the left and right. Further, by performing optimal aperture correction according to the position and shape of the opening, it is possible to acquire a high-definition stereo image regardless of the position and shape of the opening. Further, normal imaging and stereo imaging can be easily switched by controlling the aperture and the image sensor.

  According to a second aspect of the present invention, in the first aspect of the present invention, the aperture member is formed of a transmissive liquid crystal panel, and the pair of openings are formed by making part of the transmissive liquid crystal panel transparent. To do.

  According to this, the position and shape of the opening can be set freely.

  According to a third invention, in the first or second invention, the aperture member is fixed at a position crossing the optical axis, and is displaceable along the fixed plate member. The movable plate-like member is provided, and the fixed plate-like member and the movable plate-like member are respectively provided with holes for forming the pair of openings when the movable plate-like member is displaced. The configuration is as follows.

  According to this, the distance between a pair of openings can be made variable by a simple configuration of two members, a fixed plate-like member and a movable plate-like member, and can be easily applied to a device that requires downsizing. .

  According to a fourth aspect of the present invention, in the first or second aspect of the present invention, the aperture member is provided with a pair of divided bodies provided so as to be close to and away from each other in a direction orthogonal to the optical axis, and the pair of divided bodies. A pair of split members connected to the pair of divided bodies, and the pair of openings are formed by openings provided in the pair of divided bodies, respectively.

  According to this, a pair of division body and an expansion-contraction member can be arrange | positioned on the surface orthogonal to an optical axis, and the space saving of an optical axis direction is attained.

  According to a fifth invention, in the first or second invention, the aperture members are arranged in a plurality of rows in the optical axis direction and are slidable independently from each other in a direction perpendicular to the optical axis. A plurality of shutter members are provided, and the plurality of shutter members are provided with the pair of openings having different baseline lengths for each of the plurality of rows.

  According to this, since a pair of openings having different distances between the openings can be provided for each independent shutter member, the distance between the pair of openings can be changed more finely by increasing the number of shutter members.

  According to a sixth invention, in any one of the first to fifth inventions, when the pair of openings are located on the outer peripheral side of the aperture member, the sixth invention is orthogonal to a straight line connecting the centers of gravity of the openings. It is set as the structure currently formed in the shape long in a direction.

  According to this, since the centers of gravity of the pair of openings are biased further away from each other, the distance between the openings when the pair of openings is positioned on the outer peripheral side of the aperture member can be further increased.

  According to a seventh invention, in any one of the first to sixth inventions, the imaging element is disposed at a position different from a focal point of the condenser lens, and the plurality of photoelectric conversion units are arranged on the left and right sides. The light passing through one of the pair of openings is imaged on the photoelectric conversion unit for the left image on the light receiving surface of the image sensor, and the other of the pair of openings is A second lens for imaging the passing light on the photoelectric conversion unit for the right image is provided.

  According to this, by using, for example, a lenticular lens as the second lens, an image with a blur on the light receiving surface of the image sensor can be formed on the light receiving surface as it is, and is arranged separately for the left and right images. Since the left and right images of each photoelectric conversion unit can be imaged, the formation of left and right images when using a monocular lens can be realized with a simple configuration.

  According to an eighth aspect of the present invention, in any one of the first to sixth aspects, the pair of openings are alternately opened and closed in a time division manner, and photoelectric conversion by the imaging element is performed in synchronization with the time division. In addition, it is configured to have time-division control means for dividing the left and right video signals.

  According to this, left and right video signals can be alternately extracted using all of the plurality of photoelectric conversion units of the image sensor, and the resolution can be increased. Furthermore, when the left and right openings are at the same position, it can be imaged as a normal 2D camera, and switching between stereo 3D imaging and 2D imaging can be easily realized.

  According to a ninth invention, in any one of the first to eighth inventions, the pair of openings are arranged symmetrically with respect to the optical axis of the condenser lens. This facilitates adjustment of the image sensor.

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

  FIG. 1 is a schematic cross-sectional view of an essential part showing an example of an endoscope to which the present invention is applied. The endoscope 1 in the illustrated example may be a rigid endoscope used for medical or industrial purposes, and mainly includes a main body portion 2 that performs an operation and an insertion portion 3 that extends from the main body portion 2. The insertion portion 3 is a portion having a small diameter (for example, an outer diameter of 8 mm) and high rigidity that does not easily bend, and is inserted into a subject (for example, a patient's body) (not shown).

  The outer shell (cover member) of the insertion portion 3 is composed of a metal protective tube 4 and a cylindrical resin or glass tip cover 5 provided coaxially on the distal end side of the protective tube 4 in the insertion direction. Is done. A dome-shaped convex portion 5 a having a hemispheric surface functioning as an imaging window that transmits light is formed at the distal end portion in the insertion direction of the distal end cover 5. The base end side opposite to the insertion direction of the protective tube 4 is fixed to the main body 2, and the protective tube 4 defines a sealed internal space together with the distal end cover 5.

  An image sensor (imaging device) 6 made of, for example, a CCD or a CMOS, and an optical unit 7 for condensing light incident from the dome-shaped convex portion 5 a toward the image sensor 6 are provided in the distal end cover 5. It has been. The optical unit 7 has a condensing lens 8 and an aperture member 9 coaxially. The image sensor 6 and the aperture member 9 (liquid crystal panel 11) are connected to a control circuit 12 built in the main body 2 via a cable. Note that the aperture member 9 may be the aperture itself or another member having an aperture function.

  FIG. 2 is a front view showing an example of the aperture member 9. As shown in FIG. 2 (a), the aperture member 9 is constituted by a liquid crystal panel 11 as a base line length adjusting means provided on a circular wall 9a made of a light-shielding body having a disk shape. The circular wall 9a is provided in, for example, a cylindrical case of the optical unit 7, and blocks incident light from the dome-shaped convex portion 5a. The liquid crystal panel 11 is formed in a horizontally long rectangular shape extending in the diameter direction through the center of the circular wall 9a. The liquid crystal panel 11 is composed of a transparent liquid crystal that makes an arbitrary portion transparent in the entire rectangular panel. In accordance with a control signal from the control circuit 12, a transparent opening 13 having an arbitrary shape is formed in a part of the liquid crystal panel 11.

  As shown in FIG. 2A, a pair of circular openings 13 are formed at symmetrical positions with respect to the center of the circular wall 9a. The distance between the pair of openings 13 (the distance between the centers of gravity of the openings) is the base line length d in the stereo camera. The stereo camera in this embodiment includes a so-called “stereo 3D2D variable camera” that can switch between stereo 3D (three-dimensional) imaging and 2D (two-dimensional) imaging. The opening 13 can be formed at an arbitrary position by a control signal from the control circuit 12. For example, as shown in FIG. 2B, the base line length d is shortened by being positioned closer to the center of the circular wall 9a ( d1) and the base line length d can be increased (d2) by being positioned closer to the outer periphery of the circular wall 9a as shown in FIG. 2 (c). Note that the pair of openings 13 do not necessarily need to be bilaterally symmetric, and can be made asymmetric by adjusting the arrangement of the image sensor (imaging device) 6.

  Further, as shown in FIG. 1, a micro lens (second lens) 14 is disposed on the front surface (condenser lens 8 side) side of the image sensor 6. The microlens 14 may be, for example, a lenticular lens or a digital microlens by the applicant.

  FIG. 3 is an explanatory view showing an optical path according to the present invention. As shown in the figure, the condenser lens 8 is composed of an objective lens 8a on the subject 15 side and an imaging element side lens 8b on the image sensor 6 side. In the figure, each lens 8a and 8b is shown as a single unit, but may be composed of a lens group. The aperture member 9 is disposed at the stop position of the optical system including both lenses 8a and 8b.

  In the illustrated example, the microlens 14 is described as a lenticular lens. On the front surface of the image sensor 6, a left photoelectric conversion unit 6L and a right photoelectric conversion unit 6R corresponding to each pixel in the horizontal direction shown in the figure are paired, and a number of pairs corresponding to the number of lines of the lenticular lens is shown in the figure. It is arranged in the horizontal direction shown. The microlens 14 is formed in a shape in which a plurality of microlenses 14a having a semicircular cross-sectional shape that forms one line for the pair of photoelectric conversion units 6R and 6L are arranged in the left-right direction shown in the drawing.

  The light receiving surfaces of the plurality of photoelectric conversion units 6R and 6L that are the front surfaces of the image sensor 6 are positioned at a position F where light emitted from an arbitrary point of the subject 15 is focused through the optical unit 7 as shown in FIG. On the other hand, it is located closer to the imaging element side lens 8b. As a result, the light in the right optical path 16R passing through the right opening 13 in the figure from any one point of the subject 15 (for example, the intersection with the lens optical axis CL) is directed to the right photoelectric conversion unit 6R and passes through the left opening 13. The light in the left optical path 16L travels toward the left photoelectric conversion unit 6L. Then, images are formed on the corresponding photoelectric conversion units 6R and 6L by the minimum lens 14a disposed on the front surface of the image sensor 6, respectively. By setting the position of the image sensor 6 and the refractive index of the microlens 14a, the imaging positions of the optical paths 16R and 16L can be separated like the photoelectric conversion units 6R and 6L, and light from other arbitrary points can be obtained. The same processing can be performed on the lenticular lens, and a pair of the left photoelectric conversion unit 6L and the right photoelectric conversion unit 6R can correspond to a lenticular lens.

  In this way, each light emitted from the same point and passing through the separate optical paths 16R and 16L forms an image on the separate photoelectric conversion units 6R and 6L. These images have a parallax with the distance between the two openings 13 of the aperture member 9 as the base line length. Then, by capturing left and right images having parallax with all the photoelectric conversion units 6R and 6L, a video with a three-dimensional effect (a sense of depth) can be obtained.

  FIG. 4 is a block diagram showing a configuration of the control circuit 12 built in the main body 2. An output signal from the image sensor 6 is input to the left / right separation unit 12a. In the left / right separation unit 12a, pixel signals composed of electrical signals photoelectrically converted by the photoelectric conversion units 6R and 6L of the image sensor 6 are divided into left and right parts and sent to the camera signal processing unit 12b. The camera signal processing unit 12b creates left and right image signals based on the left and right pixel signals, and outputs both image signals to the 3D output combining unit 12c. The 3D output composition unit 12 c creates a 3D image based on the left and right image signals, and outputs the 3D image signal from the external terminal 17. As shown in FIG. 1, a monitor 40 is connected to the external terminal 17, and an image with a stereoscopic effect is displayed on the monitor 40.

  The control circuit 12 is provided with a drive timing generation unit 12 d for driving the image sensor 6. The main body 2 is provided with a depth sensation input switch 18 for selecting a sense of depth in, for example, three levels of large, medium, and small. The depth sensation input switch 18 is a three-dimensional aperture provided in the control circuit 12. The control unit 12e, the aperture correction control unit 12f, and the shading correction control unit 12g are connected to each other.

  The three-dimensional aperture controller 12e outputs a signal for setting the shape and position of the opening 13 of the aperture member 9, and also outputs an information signal for each setting to the left and right separation unit 12a. The aperture correction control unit 12f controls the correction amount (boost amount) and boost frequency of the aperture correction processing unit in the camera signal processing unit 12b. Examples of the correction amount and boost frequency are shown in FIGS. The shading correction control unit 12g outputs a shading correction control signal for performing shading correction to the camera processing unit 12b. An example of the shading correction amount is shown in FIG.

  A procedure for image processing in the control circuit 12 configured as described above will be described below. As described above, the light passing through the left and right openings 13 of the aperture member 9 is focused on the left and right photoelectric change elements 6R and 6L by the microlenses 14, respectively. Pixel signals from the photoelectric conversion units 6R and 6L are transferred to the left / right separation unit 12a by timing pulses generated by the drive timing generation unit 12d. This transfer can be performed according to a known transfer procedure in an image sensor in which photoelectric conversion units are arranged in a grid pattern. In the transfer signal of the present invention, the left and right pixel signals are transmitted in units of one image field alternately. It is done.

  In order to form the left and right images, the left / right separation unit 12a separates the transfer signal in which the left and right pixel signals are mixed into a collection of left pixel signals and right pixel signals, and outputs one field of the left and right video signals. Form minute by minute.

  The three-dimensional aperture control unit 12e sets the shape and position of the opening 13 as specified by the three-dimensional effect input unit 18. The distance between the openings 13 is determined by designating the stereoscopic effect, and the position of the opening 13 of the aperture member 9 is controlled. As with the conventional camera, the aperture size is set according to the size of the illuminance. When “large” of the depth sensation is selected by the depth sensation input switch 18, the distance between the openings 13 can be made larger than the standard (“medium” selection), and an image in which the depth sensation is emphasized can be captured. To. If “Small” is selected, the distance between the openings 13 is reduced so that an image with a reduced depth can be captured.

  The left and right video signals separated and formed by the left and right separation unit 12a are output to the camera signal processing unit 12b. As described above, the camera signal processing unit 12b receives the correction control signals from the aperture correction control unit 12f and the shading correction control unit 12g. The camera signal processing unit 12b creates left and right image signals by adding each correction control signal to the left and right pixel signals. As the aperture correction, the light beam that has passed through the peripheral portion of the lens is corrected to be wider and blurred than the light beam that has passed through the central portion. That is, the difference in the MTF of the lens depending on the position of the opening 13 is corrected. The blur around the lens periphery can be corrected using a known PSF function.

  Here, a case where aperture correction corresponding to a change in MTF depending on the position of the opening 13 is performed will be described. The aperture correction described below is for the change in MTF, and the frequency characteristic of the video signal that is requested in advance is corrected in accordance with the change in the MTF of the lens.

  FIG. 5A is a table showing a relative relationship between the base line length and the aperture correction amount (boost amount) with respect to “medium (standard)”, “small”, and “large” selected by the depth sensation input switch 18. As shown in the figure, when “medium (standard)” is selected (FIG. 2A), the base line length is an intermediate length (medium) d compared to the others, and the aperture correction amount is also intermediate. The amount of boost (medium). When “small” is selected (FIG. 2B), the base line length is short (small) d1, and the aperture correction amount is a boost amount (small) smaller than “medium”. When “large” is selected (FIG. 2C), the base length is the longest (large) d2, and the aperture correction amount is set to a boost amount (large) larger than “medium”. The frequency for performing the aperture correction is increased or decreased according to the characteristics of the lens 8 used.

  FIG. 6A is a diagram illustrating the relationship between the MTF characteristics and the aperture correction in the respective lens portions of the central portion, the peripheral portion, and the intermediate portion in the radial direction of the lens. As for the MTF characteristics, the center portion of the lens is indicated by a one-dot chain line, the peripheral portion near the outer periphery of the lens is indicated by a two-dot chain line, and an intermediate portion between these two portions is indicated by a broken line. become. In the figure, the target MTF characteristic is indicated by a solid line, and the aperture correction amount is a difference between the above-described lines and the solid line. The aperture correction amount is large when it is indicated by a two-dot chain line in the figure that is the peripheral part of the lens, and the aperture correction amount is small when it is indicated by the one-dot chain line in the figure that is the central part of the lens.

  Note that, as shown in the figure, the higher the spatial frequency, the more the MTF characteristic is lowered. Therefore, constant aperture correction is not performed over the entire spatial frequency, but the high frequency of the spatial frequency at which the MTF characteristic is greatly lowered. Aperture correction is performed with emphasis on the portion, and correction is performed so as to minimize the difference between the portions of the lens. Further, when the MTF characteristic is greatly deteriorated, correction is performed in a range where the decrease in the S / N ratio is not noticeable. For example, the intermediate frequency portion of the spatial frequency may be corrected so as to coincide with the target MTF characteristic, and the high frequency portion may be corrected so as to ensure a minimum S / N ratio.

  FIG. 6B is a diagram illustrating the difference in each MTF characteristic between the radial direction and the circumferential direction of the lens. As shown in the figure, the magnitude of the decrease in the spatial frequency of the MTF characteristics on the high frequency side is larger in the circumferential direction (two-dot chain line in the figure) than in the radial direction (solid line in the figure). . In the correction of the MTF characteristic, a difference in correction amount in both the radial direction and the circumferential direction is taken into consideration.

  FIG. 6C is a diagram showing a difference in MTF characteristics due to a difference in the shape of the opening. In the case of the illustrated example, a circular opening such as the opening 13 shown in FIG. 2 (solid line in the figure) and a vertically long case such as the elliptical opening 19 shown in FIG. This is the difference from the two-dot chain line. The pair of openings 19 are arranged so that the major axis of the ellipse, which is in the longitudinal direction, is orthogonal to the straight line connecting the centroids of the ellipse. As shown in FIG. 6C, the magnitude of the decrease in the spatial frequency of the MTF characteristic on the high frequency side is larger in the case of the vertically long opening (19) than in the case of the circular opening (13). FIG. 5B shows an example of MTF correction when the shape of the aperture and the circumferential direction / radial direction of the lens are considered. By separately controlling the amount of aperture correction in the horizontal direction (H) and vertical direction (V) of the image, an optimum image quality when monocular stereo imaging is performed is realized. Further, in the example of FIG. 5B, “large” that makes the effect larger than “large” in FIG. 5A can be selected as the selection target by the depth feeling input switch 18. Thereby, imaging with a greater sense of depth can be realized.

  The length of the base line length (d3 in the figure) in the case of the opening 19 in FIG. 7A is the distance between both centroids (black circles in the figure) of the ellipse that is the shape of the opening 19 in the illustrated example. The center of gravity is the midpoint of the minor axis of the ellipse (in the radial direction of the circular wall 9a), whereas the center of gravity when the shape of the opening 13 is circular as shown in FIG. If the opening area is the same, the minor axis of the ellipse is shorter than the diameter of the circle. Therefore, when the openings 13 and 19 are positioned so as to be in contact with the outer periphery of the circular wall 9a, the base line length d3 in the case of the elliptical opening 19 as a representative example of the vertically long shape is the farthest in the circular opening 13. 2 becomes longer than the base line length d2 in FIG.

  Thus, by changing the shape of the opening from a circular shape to a vertically long shape to increase the baseline length, it is possible to increase the sense of depth in stereoscopic viewing. This makes it possible to further increase the baseline length within a limited range of external dimensions, and to easily see the distance of the subject by increasing the sense of depth, which is effective for endoscopes with a reduced diameter. It is.

  FIG. 7B is a diagram corresponding to FIG. 7A illustrating another example of the vertically long opening. The shape of the opening 20 shown in the drawing is formed in a substantially rhombus shape that is long in a direction perpendicular to the diameter of the circular wall 9a. The outer peripheral side of the circular wall 9a of the opening 20 may have a shape (arc) along the outer periphery of the circular wall 9a. When the opening 20 having such a shape is bisected by a line connecting two vertices in a direction orthogonal to the diameter of the circular wall 9a, the area on the outer peripheral side of the circular wall 9a is larger than the area on the center side. growing. Thereby, the center of gravity of the substantially rhombic opening 20 is biased toward the outer peripheral side of the circular wall 9a as compared with the case of the ellipse of FIG. 7A, and the distance (baseline length) d4 between the centers of gravity of both the openings 20 is elliptic. It further increases than the distance between the centers of gravity of the openings 19 (base line length) d3. By adopting such a vertically long shape, it becomes possible to obtain an image with a further sense of depth within a range in which the opening is limited.

  The shape of the opening 41 shown in FIG. 7C is formed by an arc along the outer periphery of the circular wall 9a and a straight line (string) connecting both ends of the arc. The center of gravity of the opening 41 having such a shape is more biased toward the outer peripheral side of the circular wall 9a than in the case of the approximately rhombus shown in FIG. It further increases than the distance between the centers of gravity (base line length) d4 of the rhombus opening 20. By adopting such a vertically long shape, it becomes possible to obtain an image with a further sense of depth within a range in which the opening is limited.

  The elliptical shape and the approximate rhombus shown in FIGS. 7A and 7B can be formed in the liquid crystal panel 11 described above. Therefore, for example, the pattern of the elliptical opening 19 in FIG. 7A is added to the change in the base length due to the circular opening 13 in FIGS. 2A to 2C, and the base length is changed in four steps. Good. For such a change in the baseline length, aperture correction is performed according to the decrease in MTF. Aperture correction is performed by changing the horizontal (H) and vertical (V) correction amounts, and is shown in a table in FIG. When the correction amount of the horizontal (H) component is “medium” and the correction amount of the vertical (V) component is “large” when FIG. When the baseline length in (b) is “small”, (H) is “small” and (V) is “medium”, and when the baseline length in FIG. 2C is “large”, (H) is “large”, ( Let V) be “large”. When the baseline length is “maximum” as shown in FIG. 7A, (H) is “large” and (V) is “maximum (> large)”. Considering the characteristics of the lens in the circumferential direction and the radial direction, the best image quality (such as resolution) is realized by correcting the aperture correction independently in the horizontal (H) direction and in the vertical (V direction).

  Next, shading correction will be described. The shading correction is for correcting the luminance unevenness in the entire image. FIG. 8 is a diagram showing a light amount distribution in the image sensor 6 by the optical paths 16R and 16L passing through the left and right openings 13. As shown in FIG. 8A shows the imaging procedure and light amount distribution on the image sensor 6 in the right optical path 16R, and FIG. 8B shows the imaging procedure and light amount distribution on the image sensor 6 in the left optical path 16L. Each light quantity distribution corresponds to the left and right signal levels by the photoelectric conversion units 6R and 6L.

  As shown in FIG. 8A, the optical path passing through the opening 13 on the right side of the aperture member 9 is an image as indicated by an optical path 16Rt corresponding to one end of the subject 15 and an optical path 16Rb corresponding to the other end. An image is formed on the sensor 6. The light quantity distribution in this case is as shown in a graph (broken line in the drawing) in which the X axis of the image sensor 6 corresponding to the horizontal direction of the subject 15 is the horizontal axis and the light quantity (signal level) of the subject 15 is the vertical axis. . As shown in FIG. 8B, the optical path passing through the left opening 13 of the aperture member 9 forms an image on the image sensor 6 as indicated by the optical paths 16Lt and 16Lb similar to the above. The amount of light in this case is as shown in the graph (solid line in the figure) similar to the above.

  When the left and right light quantities are indicated by the same coordinates, the result is as shown in FIG. As shown in the figure, the position where the light quantity is maximum in each light quantity distribution corresponding to the left and right optical paths 16L and 16R is biased to the left and right. This is because a part of the light that can be imaged on the image sensor 6 by the lens 8 when there is no opening 13 is blocked by the portion of the aperture member 9 where the opening 13 is not provided. When the opening 13 is on the right side, the light traveling toward the side corresponding to the left side of the image sensor 6 is reduced, and when the opening 13 is on the left side, the right side is reduced.

  In the shading correction control unit 12g, depending on the characteristics of the lens 8 and the position selected by the depth sensation input switch 18, the two-dot chain line (light amount) shown in FIG. As shown by the correction curve), a correction control signal for correcting the corrected signal level so that the shape of the signal level becomes substantially flat is output. In the figure, the light amount correction curve is slightly reduced at the both ends of the subject range with respect to the central portion, but this is to prevent the above-described decrease in the S / N ratio at the end portion from being noticeable. is there.

  In this way, the image displayed on the external monitor 40 by the video signal whose MTF and shading are corrected according to the difference in the baseline length has a stereoscopic effect and a sense of depth, and the work target is accessed with sufficient perspective. Thus, the work performed while observing the monitor 40 in an endoscope or the like can be easily performed. Such a stereoscopic view can be achieved with a common lens (monocular) without providing separate left and right lenses, and the structure of the stereo camera can be simplified and miniaturized. It is suitable for those that are desired to be small.

  The variable structure of the opening position of the aperture member 9 (inter-opening distance varying means) is not limited to that using the liquid crystal panel 11, and other embodiments will be described.

  FIG. 10A is a diagram illustrating a second example of the aperture member. In addition, about the part similar to the above, the same code | symbol is attached | subjected and the detailed description is abbreviate | omitted (hereinafter the same).

  In this second example, a circular wall 9a as a fixed plate member is provided with a horizontally-long rectangular window 21 as a hole in the drawing extending in the diameter direction, and the window 21 is completely shielded and opened. A plate-like slider 22 is provided as a movable plate-like member that can move in a direction orthogonal to the longitudinal direction of 21 (arrow A in the figure). In the slider 22, openings 22 a, 22 b, and 22 c as three pairs of holes that can be selectively positioned in the window 21 as the slider 22 moves are arranged at predetermined intervals in the moving direction of the slider 22. Has been. Each of the pair of openings 22a, 22b, and 22c corresponds to each of the openings 13 in FIGS. 2 (b), (a), and (c).

  The slider 22 is driven by the actuator 23 according to the selection of the depth sensation switch 18 by the control circuit 12. For example, a guide groove for guiding the slider 22 may be provided on the inner wall of the insertion portion 3, a motor-driven actuator 23 may be provided in the main body portion 2, and the slider 22 and the actuator 23 may be connected via a push-pull cable. . In addition, various known slide mechanisms can be applied. In this way, the slider 22 can be moved in the direction of arrow A so that any one of the pair of openings 22a, 22b, and 22c can overlap the window 21, and the distance between the openings that changes the base line length with a simple configuration. Variable means can be constructed.

  FIG. 10B is a diagram showing a third example of the aperture member, and is another example of the slide method similar to the second example. In this third example, a total of six openings 24a, 24b, and 24c as a pair of left and right holes with different base lengths, large, medium, and small, are arranged in a row in the diameter direction on the circular wall body 9a. . The slider 25, which may have the same outer shape and slide structure as the slider 22, has a rectangular shape as each pair of holes so as to selectively overlap each pair of openings 24 a, 24 b, and 24 c as the slider 25 moves. Shape windows 25 a, 25 b, and 25 c are arranged at predetermined intervals in the moving direction of the slider 25.

  In the third example, similarly to the second example, the slider 25 is driven by the actuator 23 so that the base line length by the pair of openings 24a overlapping the window 25a is “small” and the pair of windows 25b overlap each other. Either a state in which the base line length by the pair of openings 24b is “medium” or a state in which the base line length by the pair of openings 24c overlapping the pair of windows 25c is “large” can be selected. In this case as well, the inter-opening distance variable means for changing the base line length can be configured with the same simple configuration as in the second example.

  FIG. 11 is a view showing a fourth example of the aperture member, where (a) shows a schematic side view, and (b) shows a main part in a perspective view. This fourth example is a shutter system in which three sets of plate-like shutter members 26, 27, and 28 each made up of a pair of light shielding members separated in the vertical direction shown in the figure are provided. Each pair of shutter members 26, 27, and 28 constitutes a pair with the optical axis CL of the condenser lens 8 interposed therebetween, and each pair is perpendicular to the optical axis CL and is contacted and separated as indicated by an arrow B in the figure. It is provided so as to be movable in the direction to be.

  The pair of shutter members 26 are respectively provided with a pair of semicircular recesses 26a that form a pair of circular openings in a closed state. The pair of holes formed in each semicircular recess 26a in this closed state corresponds to the pair of openings 13 (baseline length is “small”) in FIG. Similarly, the pair of shutter members 27 is provided with a pair of semicircular recesses 27a corresponding to the pair of openings 13 (base length is “medium”) in FIG. A pair of semicircular recesses 28a corresponding to the pair of openings 13 (base length is “large”) in FIG.

  The plurality of shutter members 26, 27, and 28 are also driven by the actuator 23 similar to the above. For example, by bringing the pair of shutter members 26 into contact with each other as shown by a two-dot chain line in FIG. 11B, the base line length is made “small” by the pair of semicircular recesses 26a at the top and bottom. A corresponding pair of left and right openings (26a, 26a) is formed. The same applies to each of the other shutter members 27 and 28, and it is possible to configure the inter-opening distance varying means for changing the base line length with the same simple configuration as in the second example. In the illustrated example, there are three differences in the length of the base line length. However, according to this shutter method, the number of shutter members arranged in the direction of the optical axis CL can be increased as long as space permits, thereby increasing the length of the base line length. The difference can be set in detail. Further, as another embodiment of this shutter system, a plurality of movable plate-like bodies that are provided with a fixed plate-like body at a position that avoids a range where a pair of openings are formed, and are opposed to and separated from the fixed plate-like body. Each of the movable plate-like bodies may be provided with a pair of openings having different baseline lengths so that any movable plate-like body can be opened and closed like a shutter.

  FIG. 12 is a diagram illustrating a fifth example of the aperture member. In this fifth example, as shown in FIG. 12A, a pair of divided bodies 29 having a shape obtained by dividing the circular wall body 9a into two in the left-right direction shown in the figure are provided. An opening 29 a is provided in the vicinity of the line side, and an accordion curtain-shaped light shielding body 30 is provided as an expansion / contraction member between the two divided bodies 29. The two divided bodies 29 have the same actuator 23 as described above between a state in which they are most separated from each other as shown in FIG. 12A and a state in which they are closest to each other as shown in FIG. Is driven in the direction of arrow C shown in FIG.

  Thereby, both the openings 29a can be selectively positioned in a state in which the base length is changed in the large, medium, and small as in the above examples. Further, since the accordion curtain-like light shielding body 30 is provided between the two divided bodies 29, both the divided bodies 29 can be stopped at an arbitrary position (separation distance), and the distance between the two openings 29a, that is, the baseline length Can be set to any length. Therefore, the base line length can be finely adjusted, and an image with a further three-dimensional feeling and depth feeling can be created. The light shielding body 30 is not limited to the accordion curtain shape, and a structure in which an elongated plate-like body is provided in a telescopic shape in the width direction, or an elastic body (elastomer material) that is easily stretchable is provided. May be. Also in the fifth example, it is possible to configure the inter-opening distance varying means for changing the base line length with the same simple configuration as in the second example. In addition, the pair of divided bodies 29 and the light shielding body 30 can be arranged on the same plane orthogonal to the optical axis CL, and space saving in the optical axis direction can be promoted.

  FIGS. 13A and 13B are views showing a sixth example of the aperture member, where FIG. 13A is a front view showing a usage pattern, and FIG. 13B is a perspective view showing a main part configuration. In the sixth example, as shown in FIG. 13B, the disk is composed of two disk bodies 31 and 32 that are stacked coaxially with the optical axis CL. One disc body 31 as a fixed plate member is provided with a pair of horizontally elongated windows 31a extending in series in the radial direction, and the other disc 32 as a movable plate member has a part. A pair of symmetrical arc-shaped openings 32a that overlap with the horizontally long window 31a and that extend from the central portion to the outer periphery are provided. Further, for example, one disk body 31 is fixed, and the other disk body 32 is provided coaxially and rotatable as indicated by an arrow D in the figure, and the other disk body 32 is rotated by the actuator 23. Configure.

  Thus, as shown in FIG. 13A, the outermost peripheral side portion of the arc-shaped opening 32a is overlapped with the horizontally long window 31a, thereby forming a pair of openings (31a and 32a) having the longest baseline length. be able to. Then, by rotating the other disk body 32, the portion (opening) overlapping the arc-shaped opening 32a and the horizontally elongated window 31a moves to the center (optical axis CL) side, and the length of the base line length is the same as above. The length can be set to an arbitrary length between “small” and “large”, and fine adjustment is also possible. The sixth example is the same as the second example and the third example in that two members are used, and it is also possible to configure an aperture distance varying means that changes the base line length with the same simple configuration. it can. Furthermore, by rotating the disc body 32, a portion that protrudes from the outer shape of the disc bodies 31 and 32 does not occur, so that space saving can be promoted.

  FIG. 14 is a diagram illustrating a seventh example of the aperture member, and corresponds to FIG. 3. In the seventh example, the pair of left and right openings 13 formed by the liquid crystal panel 11 constituting the aperture member are alternately opened and closed in a time division manner. In this case, light that reaches the image sensor 6 through any one of the openings 13 is received by all the photoelectric conversion units 6 a of the image sensor 6. Therefore, the microlens 14 shown in the example of FIG. 3 is not necessary. Since the liquid crystal panel 11 is used, the position of the opening 13 can be arbitrarily set, and the effect of stereoscopic viewing by changing the baseline length is the same as described above.

  FIG. 15 is a timing chart showing an example of control by the control circuit 2 as time division control means. The three-dimensional aperture control unit 12e alternately opens and closes the left opening 13 and the right opening 13 of the liquid crystal panel 11 at an opening time t1 and a switching interval t2. Further, the left / right separation unit 12a alternately reads out the received light amount accumulated by the image sensor 6 according to the left and right openings according to the switching timing from the three-dimensional aperture control unit 12e. The left and right accumulation time t3 is a time obtained by adding ½ of the switching interval t2 before and after the opening time t1, and the switching may be instantaneous. The left and right video signals read at the switching timing are output to the camera signal processing unit 12b. The processing after the camera signal processing unit 12b is the same as described above. In this way, when the left and right signals are separated by time division control, the microlens 14 used on the front surface of the image sensor 6 is not necessary. For this reason, a monocular 3D camera with high horizontal resolution (resolution) can be realized. If the aperture (opening 13) is the same on the left and right (aperture distance d = 0), it can be used as a normal 2D camera and can be realized simultaneously from 2D to a 3D camera of any base length. You can freely adjust the depth characteristics according to the doctor's need for surgery. In this case, the depth sensation input switch 18 performs continuous depth sensation input as a depth sensation input volume that enables continuous adjustment.

  FIG. 16 is a diagram illustrating another example (eighth example) of the aperture member of the time division opening. In the eighth example, as in the sixth example, two disk bodies 33 and 34 are arranged coaxially with the optical axis CL, one disk body 33 is fixed, and the other disk body. 34 is rotatably provided. One disc body 33 is provided with a pair of left and right openings 33a, and the other disc body 34 is provided with a transparent portion 34a and a light shielding portion 34b in a semicircular region. The other disk 34 is driven to rotate in one direction as indicated by an arrow E shown in the figure by an actuator 23 similar to the above in accordance with a drive control signal at a predetermined rotational speed from the control circuit 12. Is done. Thereby, when the other rotating disk 34 rotates, the pair of openings 33a are alternately opened and closed, and the same effect as the time-division shutter method similar to the seventh example can be obtained. In the illustrated example, only one pair of openings 33 a is shown in one disk body 33, so that the base line length is fixed as it is, but this eighth example is replaced with the disk body 33 and the above-mentioned By applying any one of the second to sixth examples, the same effect of changing the base line length as those examples can be obtained.

  FIG. 17 is a diagram corresponding to FIG. 1 illustrating another example of the overall configuration. In addition, about the part similar to the above, the same code | symbol is attached | subjected and the detailed description is abbreviate | omitted.

  In the example of FIG. 17, the optical unit 7 disposed in the tip cover 5 in FIG. 1 is disposed in the main body 2, and an intermediate case 35 is integrally formed between the main body 2 and the insertion portion 4. Is provided. An objective optical system 36a is disposed in the distal end cover 5 of the insertion section 4, a relay optical system 36b is disposed in the insertion section 4, and an imaging optical system 36c is disposed in the intermediate case 35. Yes. Incident light to the objective optical system 36a reaches the imaging optical system 36c via the relay optical system 36b, and is emitted from the imaging optical system 36c to the optical unit 7. Each optical system 36a to 36c may be a combination of lenses, but an optical fiber may be used for the relay optical system 36b.

  By setting the optical characteristics of the objective optical system 36a, the relay optical system 36b, and the imaging optical system 36c, the emitted light from the imaging optical system 36c is made to have the same form as the incident light to the optical unit 7 in FIG. Thus, the same operational effects as those of the above-described embodiment can be obtained. Furthermore, since the aperture member 9 is provided in the main body 2 that is not limited in diameter reduction like the insertion portion 4, the restriction on the space is eased when the movable portion is provided in the structure in which the base line length is variable. Therefore, the degree of freedom in designing the base length variable structure is increased, and it is possible to easily improve the stereoscopic effect and the depth effect.

  The present invention has been described with reference to the preferred embodiments. However, as those skilled in the art can easily understand, the present invention is not limited to such embodiments, and the gist of the present invention. As long as it does not deviate from the above, it can be appropriately changed. For example, in the above second to sixth examples, the shape of the opening with the maximum baseline length is also shown as a circle, but the shape of the opening when the baseline length is the maximum is an oval or a substantially rhombus shown in FIG. Can be formed. This is effective when it is desired to further increase the sense of depth when the baseline length is maximum. In addition, all the components shown in the above embodiment are not necessarily essential, and can be appropriately selected without departing from the gist of the present invention.

  Since the stereo camera according to the present invention can create a 3D image having a three-dimensional effect and a sense of depth with a simple configuration, the stereo camera can be miniaturized, and the size and size of endoscopes and the like that are required to be small or large. It is useful as an imaging device or the like having restrictions on the image quality. In addition, the baseline length (three-dimensional effect) can be freely set from maximum to zero (from 3D to 2D) according to the needs of the surgeon, minimizing fatigue during surgery, and operator preference Therefore, it is possible to realize a 3D-2D switching ultra-small endoscope that can accurately respond, and is useful as an endoscope that can reduce the burden on doctors and patients.

DESCRIPTION OF SYMBOLS 1 Endoscope 6 Image sensor 8 Condensing lens 9 Aperture 11 Liquid crystal panel 12 Control circuit 13 Aperture 14 Micro lens (2nd lens)
19 Opening (Oval)
20 opening (substantially diamond-shaped)

Claims (9)

An imaging device including a plurality of photoelectric conversion units, a condensing lens for condensing light from a subject toward the imaging device, and an aperture member that allows a part of the light from the subject to the imaging device to pass therethrough Have
The aperture member includes a pair of apertures arranged on the left and right with respect to the optical axis of the condenser lens, and an aperture distance variable means for changing a distance between the pair of apertures. Stereo camera. The aperture member comprises a transmissive liquid crystal panel,
The stereo camera according to claim 1, wherein the pair of openings are formed by making a part of the transmissive liquid crystal panel transparent. The aperture member has a fixed plate-like member fixed at a position crossing the optical axis, and a movable plate-like member provided to be displaceable along the fixed plate-like member,
2. The fixed plate member and the movable plate member, respectively, are provided with holes for forming the pair of openings when the movable plate member is displaced. Item 3. The stereo camera according to Item 2. A pair of divided bodies provided so that the aperture member can be moved close to and away from each other in a direction perpendicular to the optical axis; Have
The stereo camera according to claim 1, wherein the pair of openings are formed by openings provided in the pair of divided bodies, respectively. The aperture member has a plurality of shutter members arranged in a plurality of rows in the optical axis direction and slidable independently from each other in a direction perpendicular to the optical axis,
3. The stereo camera according to claim 1, wherein the plurality of shutter members are provided with the pair of openings having different baseline lengths for each of the plurality of rows.   The said pair of opening is formed in the long shape in the direction orthogonal to the straight line which connects the gravity centers of the said opening, when located in the outer peripheral side of the said aperture member. Stereo camera in any one of. The imaging element is disposed at a position different from the focal point of the condenser lens, and the plurality of photoelectric conversion units are divided for left and right images;
Light passing through one of the pair of openings is imaged on the photoelectric conversion unit for the left image on the light receiving surface of the imaging element, and light passing through the other of the pair of openings is used for the right image The stereo camera according to claim 1, wherein a second lens for forming an image on the photoelectric conversion unit is provided.   2. A time-division control unit that alternately opens and closes the pair of openings in a time-division manner and performs photoelectric conversion by the image sensor in synchronization with the time-division and divides the left and right video signals. The stereo camera according to claim 6.   The stereo camera according to any one of claims 1 to 8, wherein the pair of openings are arranged symmetrically with respect to an optical axis of the condenser lens.

Images