WO2019235091A1 - Optical low-pass filter and image capture device - Google Patents

Optical low-pass filter and image capture device Download PDF

Info

Publication number
WO2019235091A1
WO2019235091A1 PCT/JP2019/017211 JP2019017211W WO2019235091A1 WO 2019235091 A1 WO2019235091 A1 WO 2019235091A1 JP 2019017211 W JP2019017211 W JP 2019017211W WO 2019235091 A1 WO2019235091 A1 WO 2019235091A1
Authority
WO
WIPO (PCT)
Prior art keywords
pass filter
optical low
δdef
image
value
Prior art date
Application number
PCT/JP2019/017211
Other languages
French (fr)
Japanese (ja)
Inventor
和彦 桃木
Original Assignee
キヤノン株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by キヤノン株式会社 filed Critical キヤノン株式会社
Publication of WO2019235091A1 publication Critical patent/WO2019235091A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
    • G02B27/42—Diffraction optics, i.e. systems including a diffractive element being designed for providing a diffractive effect
    • G02B27/46—Systems using spatial filters
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B5/00—Optical elements other than lenses
    • G02B5/30—Polarising elements
    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03B—APPARATUS OR ARRANGEMENTS FOR TAKING PHOTOGRAPHS OR FOR PROJECTING OR VIEWING THEM; APPARATUS OR ARRANGEMENTS EMPLOYING ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ACCESSORIES THEREFOR
    • G03B11/00—Filters or other obturators specially adapted for photographic purposes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof

Definitions

  • the present invention relates to an optical low-pass filter used in an imaging apparatus such as a digital camera or a video camera.
  • an optical low-pass filter capable of limiting high-frequency image information exceeding the Nyquist frequency is used in order to prevent generation of false colors and moire.
  • the following optical low-pass filters are effective for a conventional imaging apparatus in which the pixel pitch of the imaging element is 5 ⁇ m or more.
  • Patent Document 1 discloses an optical low-pass filter that separates a point image into four using a birefringent plate that separates the point image in the horizontal direction and a birefringent plate that separates the point image in the vertical direction.
  • Patent Document 2 discloses an optical low-pass filter that uses four birefringent plates and generates 16 point images by separating the point images in different directions by 45 ° in each layer.
  • the optical low-pass filters disclosed in these Patent Documents 1 and 2 cut high-frequency image information by separating a point image in the in-plane direction of the imaging surface of the imaging element to form a blurred image.
  • an imaging apparatus having an imaging element with a pixel pitch of about 3 ⁇ m requires a bright imaging optical system with an open F value of 4.0 or less.
  • the optical low-pass filter used for such an image sensor needs to be set so as to obtain the maximum resolution at the open F value. Since the resolution deteriorates due to diffraction when the aperture is narrowed, the low-pass effect is not originally necessary for F values other than the vicinity of the open F value.
  • the optical low-pass filters disclosed in Patent Documents 1 and 2 have a constant low-pass effect regardless of the F value. For this reason, when these optical low-pass filters are used for the above-described image sensor having a fine pixel pitch, the image quality is greatly deteriorated when an F value other than the vicinity of the open F value is set (particularly at a small aperture). Become.
  • the present invention provides an optical low-pass filter capable of obtaining a low-pass effect having an appropriate intensity according to the F value, and an imaging apparatus using the optical low-pass filter.
  • An optical low-pass filter as one aspect of the present invention guides a light beam to an imaging surface of an imaging element.
  • the optical low-pass filter generates a blurred image by separating a condensing point of a light beam into a plurality of light beams in a first direction perpendicular to the imaging surface.
  • the pixel pitch of the image sensor is P [ ⁇ m]
  • the size of the blurred image generated by the separation of the condensing point in the first direction is ⁇ def [ ⁇ m]
  • the light beam is condensed in the second direction orthogonal to the first direction.
  • image sensor unit and the image pickup apparatus including the optical low-pass filter also constitute another aspect of the present invention.
  • an optical low-pass filter capable of obtaining a low-pass effect having an appropriate intensity according to the F value.
  • an imaging device that can acquire a higher quality image.
  • FIG. 1 is a cross-sectional view of an imaging system when an optical low-pass filter that is Embodiment 1 of the present invention is mounted on a single-lens digital camera.
  • FIG. 3 is a diagram illustrating an optical low-pass filter according to the first embodiment. The figure which shows the direction of the optical axis of a uniaxial birefringent element. The figure which shows the focus position shift by a parallel plate. The figure which shows the point image separation in a focus direction.
  • FIG. 3 is a diagram showing a blurred image formed by the birefringent element in Example 1.
  • 5 is a graph showing a blur diameter for each F value in Example 1. The graph which shows the blur diameter for every F value in a comparative example.
  • FIG. 3 is a diagram illustrating an optical low-pass filter according to the first embodiment. The figure which shows the direction of the optical axis of a uniaxial birefringent element. The figure which shows the focus position shift by a parallel plate. The figure which shows the point
  • FIG. 6 is a diagram illustrating an optical low-pass filter according to a second embodiment.
  • FIG. 6 is a diagram showing point image separation in the first birefringent element in Example 2.
  • FIG. 6 is a diagram showing point image separation in the first and second birefringent elements in the second embodiment.
  • 7 is a graph showing the blur diameter for each F value in Example 2.
  • the graph which shows the blur diameter for every F value in a comparative example.
  • the figure which shows the point image separation in a comparative example The figure which shows the point image separation in a comparative example.
  • FIG. 1 shows a configuration of an imaging system when an optical low-pass filter 6 that is Embodiment 1 of the present invention is mounted on a single-lens digital camera 1 as an imaging apparatus.
  • a light beam from a subject (not shown) is collected by an imaging lens (imaging optical system) 5 in the interchangeable lens 2 and forms an image in the vicinity of a two-dimensional imaging device (hereinafter referred to as an image sensor) 7.
  • the imaging lens 5 is provided with a diaphragm 8 having a variable aperture diameter.
  • the image sensor 7 is a CCD sensor, a CMOS sensor, or the like, and is a photoelectric conversion element in which a plurality of pixels are two-dimensionally arranged.
  • An optical low-pass filter 6 is disposed between the imaging lens 5 and the image sensor 7.
  • the optical low-pass filter 6 imparts a low-pass effect to the light flux from the imaging lens 5 by separating its condensing point into a plurality of points.
  • the optical low-pass filter 6 and the image sensor 7 may be integrated as an image sensor unit (imaging element unit).
  • FIG. 2 schematically shows the optical low-pass filter 6.
  • the optical low-pass filter 6 is configured by a birefringent element 1 that is slightly larger than the image sensor 7 and formed as a parallel plate having a rectangular shape similar to the image sensor 7.
  • the birefringent element 1 is formed of a birefringent material made of a uniaxial crystal such as lithium niobate.
  • the direction in which the x-axis extending in parallel to the long side of the optical low-pass filter 6 (and image sensor 7) is referred to as the x-axis direction (horizontal direction), and y parallel to the short side.
  • the direction in which the axis extends is called the y-axis direction (vertical direction).
  • a plane including the x axis and the y axis, that is, a plane parallel to the imaging surface of the image sensor 7 is an xy plane.
  • An axis orthogonal to the xy plane (imaging plane) is referred to as a z-axis, and a direction in which the z-axis extends is referred to as a z-axis direction.
  • the z-axis direction is the optical axis direction in which the optical axis O of the imaging lens 5 extends, in other words, the focus direction as the first direction in which the image sensor 7 is disposed when viewed from the optical low-pass filter 6. . Further, the x-axis direction and the y-axis direction are lateral directions as a second direction along (parallel to) the imaging surface of the image sensor 7.
  • the thickness of the birefringent element 1 in the z-axis direction is shown to be thicker than the actual thickness of several hundreds of ⁇ m.
  • the optical axis 1a of the birefringent element 1 as a uniaxial crystal extends in the x-axis direction and extends in parallel to the incident surface 1b of the birefringent element 1.
  • the azimuth angle ⁇ around the z axis (counterclockwise) with respect to the x axis of the optical axis 1a is 0 °
  • the angle ⁇ of the optical axis with respect to the z-axis is set to about 45 ° ⁇ 20 °.
  • the angle ⁇ of the optical axis 1a with respect to the z axis is set to 90 °, that is, the optical axis 1a is arranged in parallel to the incident surface 1b. Yes. For this reason, point image separation in the horizontal direction does not occur, and only point image separation in the focus direction (z-axis direction) occurs.
  • the position of the light condensing point before the insertion that is, the imaging position (hereinafter referred to as the focus position) is set as the focus direction from P0 before the insertion. It shifts to P1 behind. This is because air having a refractive index of 1 is replaced with a medium having a refractive index n, so that the optical path length d in the air becomes d / n.
  • LN lithium niobate
  • the focus position of the ordinary ray and the focus position of the extraordinary ray are not only shifted backward from P0 before insertion of the LN, but are also different rear positions P1. And shift to P2. That is, the focus position is separated into two in the focus direction.
  • the refractive index no for normal light is 2.3247
  • the refractive index for ordinary light is higher. For this reason, the focus position P1 of the ordinary ray is shifted rearward from the focus position P2 of the extraordinary ray.
  • a shift amount (hereinafter referred to as a focus position separation amount) L in the focus direction between the focus position of the ordinary ray and the focus position of the extraordinary ray is given by the following equation.
  • L d (1 / ne-1 / no)
  • L 0.014 mm (14 ⁇ m).
  • FIG. 6 shows the size of a blurred image (hereinafter referred to as a blurred diameter) caused by the deviation of the focus position between the ordinary ray and the extraordinary ray for each F value of the imaging lens 5. Note that even if the focus position is shifted due to the insertion of the parallel plate, the converging angle of the light flux at each F value does not change. follow the light path.
  • a blurred diameter a blurred image caused by the deviation of the focus position between the ordinary ray and the extraordinary ray for each F value of the imaging lens 5.
  • the blur diameter ⁇ def which is the diameter of the light beam, is minimized at an intermediate point between the focus position P1 of the ordinary ray and the focus position P2 of the extraordinary ray.
  • This blur diameter ⁇ def can be obtained geometrically by the following equation, where F is F and L is the separation amount of the focus position of ordinary and extraordinary rays.
  • ⁇ def L / (2F)
  • the ⁇ def for these F values is 3.89, 2.56 and 1.25 ⁇ m, respectively.
  • the blur diameter decreases as the F value increases.
  • the resolution degradation (image degradation) caused by narrowing the aperture 8 in the imaging lens 5 from the open is expressed as the minimum circle of confusion ⁇ Fno when the imaging lens 5 has no aberration.
  • the image degradation due to diffraction increases as the aperture 8 is narrowed from the open position. On the contrary, the blur diameter generated by the birefringent element 1 is reduced, and the image degradation can be suppressed as a whole.
  • the optical low-pass filter 6 As image degradation, it is necessary to consider a superposition of the blur diameter ⁇ def due to the birefringent element 1 (optical low-pass filter 6) and the blur diameter as the minimum circle of confusion ⁇ Fno due to diffraction.
  • the synthesized blur diameter ⁇ all generated by the synthesis of ⁇ def and ⁇ Fno is obtained empirically by the following equation.
  • ⁇ all ⁇ ( ⁇ def 2 + ⁇ Fno 2 )
  • FIG. 7 shows a blur diameter as a minimum circle of confusion ⁇ Fno (short broken line) due to diffraction for each F value and a blur diameter generated in the focus direction by the optical low-pass filter 6 (hereinafter referred to as a focus direction blur diameter) ⁇ def (long).
  • a broken line) and a synthetic blur diameter ⁇ all (solid line) obtained by combining them are shown.
  • FIG. 7 shows a case where the pixel pitch P of the image sensor 7 is 4.2 ⁇ m.
  • the target blur diameter necessary for removing false colors and moire from the image sensor 7 is about 3.0 ⁇ m. This target blur diameter is indicated by a dotted line in FIG.
  • the focus direction blur diameter ⁇ def is large on the open F value side and small on the small aperture F value side.
  • the synthetic blur diameter ⁇ all is 3.0 ⁇ m which is the target blur diameter when the F value is 2.8 to 4.0. Further, in the F values from 2.0 to 1.8, the synthetic blur diameter ⁇ is 4.0 ⁇ m, which is slightly larger than the target blur diameter, and a strong low-pass effect is generated.
  • FIG. 8 shows a case where a conventional optical low-pass filter that performs only point image separation in the horizontal direction is used as a comparative example with respect to FIG.
  • FIG. 8 shows a blur diameter as a minimum circle of confusion ⁇ Fno (short broken line) due to diffraction for each F value, a horizontal blur diameter ⁇ x (long broken line) which is a blur diameter generated in the horizontal direction, and a composition generated by combining these.
  • the blur diameter ⁇ all (solid line) is shown.
  • the pixel pitch is 4.2 ⁇ m.
  • the conventional optical low-pass filter separates the incident light beam into four point images P11 to P14 in the lateral direction, and forms a constant blurred image regardless of the F value.
  • This optical low-pass filter produces a target blur diameter of about 3.0 ⁇ m at the open F value where the highest resolution is obtained.
  • the composite blur diameter ⁇ all is almost the target blur diameter, but the F value of 2.8 to 4.0 is too large, causing image degradation more than necessary. Further, at the F value of 5.6, a blur diameter ( ⁇ Fno) corresponding to the resolution degradation due to diffraction at the F value of 8.0 occurs. Such excessive blur is added to an F value of about 11.3.
  • optical low-pass filter 6 of the present embodiment image deterioration due to the optical low-pass filter at the F value that enables high-resolution imaging can be reduced.
  • the low-pass effect of the optical low-pass filter is ( ⁇ def 2 + ⁇ x 2 ) / P 2 in Equation (1), that is, the sum of the square of the defocus blur diameter ⁇ def and the square of the lateral blur diameter ⁇ x ( ⁇ def 2 + ⁇ x 2 ) And the square of the pixel pitch P is empirically recognized.
  • the upper limit value of the expression (1) indicates that the optical low-pass filter generally generates a blur diameter that is less than or equal to the pixel pitch of the image sensor. If the value of ( ⁇ def 2 + ⁇ x 2 ) / P 2 exceeds the upper limit value, excessive image deterioration is caused, which is not preferable. Further, the lower limit value of the expression (1) indicates a limit value when the low pass effect is weakened. If a blur diameter that is so small that the value of ( ⁇ def 2 + ⁇ x 2 ) / P 2 falls below this lower limit value, the low-pass effect is too weak, and a false color or moire removal effect cannot be expected.
  • Equation (1) By limiting the total amount of blurring (blurring diameter) using Equation (1), even when blurring is generated by point image separation in the focus direction, it easily corresponds to the blurring amount generated by the conventional optical low-pass filter.
  • the amount of blur to be generated can be generated.
  • the condensed light flux from the imaging lens 5 is separated into an ordinary ray and an extraordinary ray by an optical low-pass filter 6 that is a parallel plate of LN, and the focus position thereof is separated in the focus direction.
  • the focus position separation amount L in the focus direction is 14 ⁇ m
  • the focus direction blur diameter ⁇ def 3.89 ⁇ m.
  • ⁇ x 0.
  • the value of ( ⁇ def 2 + ⁇ x 2 ) / P 2 is 0.85, which satisfies the condition of formula (1).
  • the optical low-pass filter 6 gives an appropriate low-pass effect to the pixel pitch of the image sensor 7. Furthermore, it is more preferable to satisfy the following formula (1a). 0.3 ⁇ ( ⁇ def 2 + ⁇ x 2 ) / P 2 ⁇ 0.88 (1a)
  • the inventor shows that the intensity of the low-pass effect is D ⁇ / (F ⁇ P) in the following expression (4), that is, the standard deviation of the separation amount of the focus position in the focus direction from F which is the open F value of the imaging lens 5. It is also empirically recognized that it is given using D ⁇ [ ⁇ m]. Image degradation due to diffraction occurs mainly when the pixel pitch P is 5.0 ⁇ m or less. Further, as the amount of blur in this case, the amount of blur in the focus direction may be mainly considered, and the amount of blur in the horizontal direction may be ignored. For this reason, it is desirable to satisfy the conditions shown by the following formula (4).
  • the value of D ⁇ / (F ⁇ P) is 0.926, which satisfies the condition of the formula (4).
  • F 1.4 It is desirable to determine the amount of blur by setting.
  • the angle (azimuth angle) ⁇ formed by the optical axis 1a of the birefringent element 1 with respect to the normal line of the incident surface 1b is 90 °.
  • the case where the angle ⁇ is smaller than 90 ° will be described in Example 2 below.
  • FIG. 9 shows a configuration of an optical low-pass filter 6 ′ that is Embodiment 2 of the present invention.
  • the optical low-pass filter 6 ′ of this embodiment is used in place of the optical low-pass filter 6 in the camera 1 shown in FIG. 1.
  • the optical low-pass filter 6 ′ is slightly larger than the image sensor 7 and has a first birefringent element 2 formed as a plurality of (three in this embodiment) parallel plates having a rectangular shape similar to the image sensor 7.
  • Two birefringent elements 4 are stacked in the z-axis direction.
  • a third birefringent element 3 is arranged between the first birefringent element 2 and the second birefringent element 4.
  • the first, second and third birefringent elements 2, 4 and 3 are each formed of a birefringent material made of a uniaxial crystal such as lithium niobate.
  • the first, second and third birefringent elements 2, 4 and 3 are shown separated from each other, but they are actually in contact with each other.
  • the thickness of each birefringent element in the z-axis direction is shown to be thicker than the actual several hundred ⁇ m.
  • the azimuth angles ⁇ 2 and ⁇ 4 around the z axis (counterclockwise) with respect to the x axis of the optical axes 2a and 4a of the first and second birefringent elements 2 and 4 are 90 ° and 0 °, respectively. Further, the angles ⁇ 2 and ⁇ 4 with respect to the z axis (that is, the focus direction) that is the normal line of the incident surfaces 2b and 4b of the first and second birefringent elements 2 and 4 of the optical axes 2a and 4a are expressed by the above-described formula ( 6) is set.
  • the angles ⁇ 2 and ⁇ 4 may be the same as each other or different from each other. It should be noted that the order in which the first and second birefringent elements 2 and 4 are stacked may be in this order from the light incident side as shown or may be reversed.
  • the third birefringent element 3 is a depolarizing phase plate, and the azimuth angle ⁇ of the optical axis 3a is 135 °. In the azimuth angle ⁇ , 135 ° is equivalent to 45 °.
  • the phase difference that the third birefringent element 3 gives to the light having the wavelength ⁇ that passes through the third birefringent element 3 is set to ⁇ / 4, so that two straight lines of the ordinary ray and the extraordinary ray separated by the first birefringent element 2 are obtained.
  • Depolarization is performed by converting each polarized light into circularly polarized light. In this case, an achromatic 1 / 4 ⁇ plate is used.
  • a retardation plate that gives a phase difference of ⁇ or more may be used as the third birefringent element 3.
  • a quartz plate having a thickness of about 0.2 to 0.4 mm is used as the third birefringent element 3 to give a phase difference of ⁇ or more to each of ordinary and extraordinary rays of each wavelength. .
  • FIGS. 10 (a) to 10 (d) show point image separation for a light beam transmitted through the first birefringent element 2.
  • FIGS. 11A to 11D show point image separation for light beams transmitted through the first and second birefringent elements 2 and 4.
  • FIGS. 10A and 10B show the separation of the light beams as seen from the y-axis direction and the x-axis direction, respectively. The same applies to FIGS. 11A and 11B.
  • FIGS. 10C and 10D corresponding to FIGS.
  • FIGS. 10A and 10B show a blurred image on an evaluation surface, which will be described later, as seen from the z-axis direction.
  • FIGS. 11 (c) and 11 (d) corresponding to FIGS. 11 (a) and 11 (b).
  • 10 (a) and 10 (b) and FIGS. 11 (a) and 11 (b) the ordinary light beam is indicated by a solid line
  • the extraordinary light beam is indicated by a broken line.
  • point images formed by ordinary rays are indicated by black circles
  • point images formed by extraordinary rays are indicated by white circles.
  • the light beam that has passed through the first birefringent element 2 passes through the optical axis 2a of the first birefringent element 2 in the same manner as when it passes through a normal optical low-pass filter. They are separated into two in the y-axis direction which is the direction of the azimuth angle ⁇ .
  • the ordinary ray passes straight through the first birefringent element 2, and the extraordinary ray shifts in the y-axis direction.
  • the focus positions P1 and P2 of the ordinary ray and the extraordinary ray that have passed through the first birefringent element 2 are shifted in the y-axis direction.
  • the shift amount in the y-axis direction of the focus positions P1 and P2 is set to 1.0 ⁇ m.
  • the light beam forming these two blurred images is subjected to the depolarization action by the third birefringence element 3 and then subjected to the point image separation action by the second birefringence element 4.
  • the azimuth angle ⁇ of the optical axis 4a of the second birefringent element 4 is 0 °
  • two light beams that have passed through the second birefringent element 4 are two in the x-axis direction that is the direction of the azimuth angle ⁇ .
  • the two light beams from the first birefringent element 2 are separated into an ordinary ray and an extraordinary ray, respectively.
  • the ordinary ray passes straight through the second birefringent element 4, and the extraordinary ray shifts in the x-axis direction.
  • the focus positions P3 and P4 of extraordinary rays that have passed through the second birefringent element 4 are shifted in the x-axis direction with respect to the focus positions P1 and P2, respectively.
  • the shift amount in the x-axis direction of the focus positions P3 and P4 is set to 1.0 ⁇ m.
  • the four point images thus separated from each other are located at the four vertices of the square in the lateral direction as shown in FIGS. 11 (c) and 11 (d). Further, as shown in FIGS. 11 (a) and 11 (b), it is separated into three places in the focus direction, and the positions of the two central point images in the focus direction are the same. Since the size of the blurred image is minimized at the center, this is the evaluation surface.
  • a blurred image is generated by combining two large blurred images and two imaged point images.
  • the size of the synthesized blur image is a combination of a blur diameter due to point image separation in the horizontal direction (horizontal blur diameter) and a blur diameter due to point image separation in the focus direction (focus direction blur diameter).
  • the focus direction blur diameter ⁇ def can also be approximated by Expression (3) in the present embodiment as in the first embodiment.
  • FIG. 12 shows a blur diameter as a minimum circle of confusion ⁇ Fno (short broken line) due to diffraction for each F value in this embodiment, a focus direction blur diameter ⁇ def (long broken line) and a horizontal blur diameter ⁇ x (one point) by the optical low-pass filter 6.
  • the synthetic blur diameter is the same when viewed from both the x-axis direction and the y-axis direction.
  • FIG. 12 shows a case where the pixel pitch P of the image sensor 7 is 3.0 ⁇ m.
  • the target blur diameter necessary for removing false colors and moire from the image sensor 7 is about 2.14 ⁇ m. This target blur diameter is indicated by a dotted line in FIG.
  • the focus direction blur diameter ⁇ def is large on the open F value side and small on the small aperture F value side, as in the first embodiment.
  • the horizontal blur diameter ⁇ x is a constant value (1.0 ⁇ m) regardless of the F value.
  • the synthetic blur diameter ⁇ all is extremely close to the target blur diameter when the F value is 1.4 to 2.8 or less.
  • the minimum confusion circle diameter ⁇ Fno is asymptotic and the minimum necessary blur diameter is obtained. For this reason, image deterioration due to small aperture diffraction can be minimized.
  • FIG. 13 shows a case where a conventional optical low-pass filter that performs only point image separation in the horizontal direction is used as a comparative example with respect to FIG.
  • FIG. 13 shows a blur diameter as a minimum circle of confusion ⁇ Fno (short broken line) by diffraction for each F value, a horizontal blur diameter ⁇ x (long broken line), and a synthesized blur diameter ⁇ all (solid line) generated by combining these.
  • the pixel pitch is 3.0 ⁇ m.
  • the conventional optical low-pass filter generates a blur diameter of 2.14 ⁇ m regardless of the F value, and the image deterioration in the F value range of 2.0 to 8.0 is remarkable.
  • FIG. 14 shows a blur diameter as a minimum circle of confusion ⁇ Fno (short broken line) by diffraction for each F value when an optical low-pass filter that performs point image separation only in the focus direction is used as in the first embodiment.
  • a horizontal blur diameter ⁇ x (long broken line) and a synthetic blur diameter ⁇ all (solid line) are shown.
  • the pixel pitch is 3.0 ⁇ m.
  • the angle ⁇ of the optical axis needs to be set in the range of the above-described formula (7). Moreover, the angle ⁇ needs to be set so that a point image separation amount in the focus direction and the horizontal direction is generated such that the focus direction blur diameter is larger than the horizontal blur diameter.
  • An optical low-pass filter that can reduce image degradation due to small-aperture diffraction while reducing false color and moire by adjusting the ratios of the focus direction blur diameter and lateral blur diameter to satisfy these conditions. Can be obtained.
  • each of the above embodiments can be applied to an image sensor having various pixel pitches.
  • the blur diameter due to small aperture diffraction is small, there is no problem that each embodiment does not solve at a relatively large pixel pitch.
  • This problem occurs mainly when an image sensor with a pixel pitch of 5 ⁇ m or less is used, and particularly when the pixel pitch is 3.0 ⁇ m or less.
  • the optical low-pass filter of each embodiment may be used.
  • each embodiment is mainly effective when using a high-resolution imaging lens having an open F value of about 1.0 to 2.8. Furthermore, when using a high-resolution imaging lens having an open F value of 1.0 to 2.0, there is a high demand for image quality obtained by imaging and aberration correction of the imaging lens is performed at a high level. There is a high need for an optical low-pass filter for reducing color and moire. In addition, each embodiment is effective because there is a high demand for preventing image quality deterioration due to small aperture diffraction.
  • the pixel pitch P [ ⁇ m] is also shown. In both Examples 1 and 2, the conditions of Formula (1) and Formula (4) are satisfied.

Abstract

[Problem] To provide an optical low-pass filter that achieves a low-pass effect with an intensity that is appropriate for a given F-number. [Solution] An optical low-pass filter 6 introduces a light flux onto the image capture surface of an image capture element. The optical low-pass filter allows a light flux to form a plurality of separated focal points in a first direction perpendicular to the image capture surface to thereby create a blurred image. The optical low-pass filter satisfies the following conditional expressions: 0.2 ≤ (δdef2 + δx2) / P2 ≤ 0.9 and δx ≤ δdef where P [µm] represents the pixel pitch on the image capture element, δdef [µm] represents the size of a blurred image created by the focal points separated in the first direction, and δx [µm] represents the size of a blurred image created by the focal points separated in a second direction perpendicular to the first direction (wherein δx = 0 unless the focal point is separated in the second direction).

Description

光学ローパスフィルタおよび撮像装置Optical low-pass filter and imaging device
　本発明は、デジタルカメラやビデオカメラ等の撮像装置に用いられる光学ローパスフィルタに関する。 The present invention relates to an optical low-pass filter used in an imaging apparatus such as a digital camera or a video camera.
　ＣＣＤセンサやＣ－ＭＯＳセンサ等の２次元撮像素子を使用する撮像装置では、偽色やモアレの発生を防ぐために、ナイキスト周波数を超える高周波の像情報を制限できる光学ローパスフィルタが用いられる。撮像素子の画素ピッチが５μｍ以上である従来の撮像装置に有効な光学ローパスフィルタとして以下のものがある。 In an image pickup apparatus using a two-dimensional image pickup element such as a CCD sensor or a C-MOS sensor, an optical low-pass filter capable of limiting high-frequency image information exceeding the Nyquist frequency is used in order to prevent generation of false colors and moire. The following optical low-pass filters are effective for a conventional imaging apparatus in which the pixel pitch of the imaging element is 5 μm or more.
　特許文献１には、点像を水平方向に分離する複屈折板と垂直方向に分離する複屈折板とを用いて、点像を４つに分離する光学ローパスフィルタが開示されている。また、特許文献２には、４層の複屈折板を用い、各層で点像を４５°ずつ異なる方向に分離することで１６の点像を生成する光学ローパスフィルタが開示されている。これら特許文献１，２に開示された光学ローパスフィルタは、点像を撮像素子の撮像面の面内方向で分離してボケ像とすることで高周波の像情報をカットする。 Patent Document 1 discloses an optical low-pass filter that separates a point image into four using a birefringent plate that separates the point image in the horizontal direction and a birefringent plate that separates the point image in the vertical direction. Patent Document 2 discloses an optical low-pass filter that uses four birefringent plates and generates 16 point images by separating the point images in different directions by 45 ° in each layer. The optical low-pass filters disclosed in these Patent Documents 1 and 2 cut high-frequency image information by separating a point image in the in-plane direction of the imaging surface of the imaging element to form a blurred image.
特開平１０－０５４９６０号公報Japanese Patent Application Laid-Open No. 10-054960 特開昭６２－００３２０２号公報JP-A 62-003202
　しかしながら、画素ピッチが５μｍより微細になると、撮像光学系の絞りによる回折の影響が無視できなくなる。例えば、画素ピッチが３μｍ程度の撮像素子を有する撮像装置では、開放Ｆ値が４．０以下の明るい撮像光学系が必要とされる。 However, when the pixel pitch is smaller than 5 μm, the influence of diffraction due to the diaphragm of the imaging optical system cannot be ignored. For example, an imaging apparatus having an imaging element with a pixel pitch of about 3 μm requires a bright imaging optical system with an open F value of 4.0 or less.
　このような撮像素子に用いられる光学ローパスフィルタは、開放Ｆ値において最大の解像度が得られるように設定される必要がある。絞りを絞り込むと回折によって解像度が劣化するため、開放Ｆ値付近以外のＦ値ではローパス効果は本来は必要ない。しかし、特許文献１および２にて開示された光学ローパスフィルタは、Ｆ値にかかわらず一定のローパス効果を有する。このため、これらの光学ローパスフィルタを上述した微細な画素ピッチを有する撮像素子に対して用いると、開放Ｆ値付近以外のＦ値が設定されたとき（特に小絞り時）に画質の劣化が大きくなる。 The optical low-pass filter used for such an image sensor needs to be set so as to obtain the maximum resolution at the open F value. Since the resolution deteriorates due to diffraction when the aperture is narrowed, the low-pass effect is not originally necessary for F values other than the vicinity of the open F value. However, the optical low-pass filters disclosed in Patent Documents 1 and 2 have a constant low-pass effect regardless of the F value. For this reason, when these optical low-pass filters are used for the above-described image sensor having a fine pixel pitch, the image quality is greatly deteriorated when an F value other than the vicinity of the open F value is set (particularly at a small aperture). Become.
　本発明は、Ｆ値に応じた適切な強度のローパス効果が得られるようにした光学ローパスフィルタおよびこれを用いた撮像装置を提供する。 The present invention provides an optical low-pass filter capable of obtaining a low-pass effect having an appropriate intensity according to the F value, and an imaging apparatus using the optical low-pass filter.
　本発明の一側面としての光学ローパスフィルタは、撮像素子の撮像面に光束を導く。該光学ローパスフィルタは、撮像面に垂直な第１の方向において光束の集光点を複数に分離することでボケ像を生じさせる。撮像素子の画素ピッチをＰ［μｍ］、第１の方向での集光点の分離により生ずるボケ像のサイズをδｄｅｆ［μｍ］、第１の方向に直交する第２の方向において光束の集光点を複数に分離する場合の該分離により生ずるボケ像のサイズをδｘ［μｍ］（ただし、集光点が第２の方向において分離されない場合はδｘ＝０）とするとき、
０．２≦（δｄｅｆ２＋δｘ２）／Ｐ２≦０．９
δｘ≦δｄｅｆ
なる条件を満足する。 An optical low-pass filter as one aspect of the present invention guides a light beam to an imaging surface of an imaging element. The optical low-pass filter generates a blurred image by separating a condensing point of a light beam into a plurality of light beams in a first direction perpendicular to the imaging surface. The pixel pitch of the image sensor is P [μm], the size of the blurred image generated by the separation of the condensing point in the first direction is δdef [μm], and the light beam is condensed in the second direction orthogonal to the first direction. When the size of a blurred image generated by separating the points into a plurality of points is δx [μm] (however, δx = 0 when the condensing points are not separated in the second direction),
0.2 ≦ (δdef 2 + δx 2 ) / P 2 ≦ 0.9
δx ≦ δdef
Satisfy the following conditions.
　なお、上記光学ローパスフィルタを備えた撮像素子ユニットおよび撮像装置も、本発明の他の一側面を構成する。 Note that the image sensor unit and the image pickup apparatus including the optical low-pass filter also constitute another aspect of the present invention.
　本発明によれば、Ｆ値に応じた適切な強度のローパス効果が得られる光学ローパスフィルタを実現することができる。そして、このような光学ローパスフィルタを用いることで、より高画質の画像を取得可能な撮像装置を実現することができる。 According to the present invention, it is possible to realize an optical low-pass filter capable of obtaining a low-pass effect having an appropriate intensity according to the F value. By using such an optical low-pass filter, it is possible to realize an imaging device that can acquire a higher quality image.
本発明の実施例１である光学ローパスフィルタを一眼デジタルカメラに搭載したときの撮像系の断面図。1 is a cross-sectional view of an imaging system when an optical low-pass filter that is Embodiment 1 of the present invention is mounted on a single-lens digital camera. 実施例１の光学ローパスフィルタを示す図。FIG. 3 is a diagram illustrating an optical low-pass filter according to the first embodiment. 一軸性複屈折素子の光学軸の方向を示す図。The figure which shows the direction of the optical axis of a uniaxial birefringent element. 平行平板によるピント位置ずれを示す図。The figure which shows the focus position shift by a parallel plate. フォーカス方向での点像分離を示す図。The figure which shows the point image separation in a focus direction. 実施例１における複屈折素子により形成されるボケ像を示す図。FIG. 3 is a diagram showing a blurred image formed by the birefringent element in Example 1. 実施例１におけるＦ値ごとのボケ径を示すグラフ。5 is a graph showing a blur diameter for each F value in Example 1. 比較例におけるＦ値ごとのボケ径を示すグラフ。The graph which shows the blur diameter for every F value in a comparative example. 実施例２の光学ローパスフィルタを示す図。FIG. 6 is a diagram illustrating an optical low-pass filter according to a second embodiment. 実施例２における第１の複屈折素子での点像分離を示す図。FIG. 6 is a diagram showing point image separation in the first birefringent element in Example 2. 実施例２における第１および第２の複屈折素子での点像分離を示す図。FIG. 6 is a diagram showing point image separation in the first and second birefringent elements in the second embodiment. 実施例２におけるＦ値ごとのボケ径を示すグラフ。7 is a graph showing the blur diameter for each F value in Example 2. 比較例におけるＦ値ごとのボケ径を示すグラフ。The graph which shows the blur diameter for every F value in a comparative example. １つの複屈折素子を用いた場合のＦ値ごとのボケ径を示すグラフ。The graph which shows the blur diameter for every F value at the time of using one birefringent element. 比較例における点像分離を示す図。The figure which shows the point image separation in a comparative example.
　以下、本発明の実施例について図面を参照しながら説明する。 Hereinafter, embodiments of the present invention will be described with reference to the drawings.
　図１は、本発明の実施例１である光学ローパスフィルタ６を撮像装置としての一眼デジタルカメラ１に搭載したときの撮像系の構成を示す。不図示の被写体からの光束は、交換レンズ２内の撮像レンズ（撮像光学系）５により集光されて２次元撮像素子（以下、イメージセンサという）７付近に結像する。撮像レンズ５には、絞り開口径が可変である絞り８が設けられている。 FIG. 1 shows a configuration of an imaging system when an optical low-pass filter 6 that is Embodiment 1 of the present invention is mounted on a single-lens digital camera 1 as an imaging apparatus. A light beam from a subject (not shown) is collected by an imaging lens (imaging optical system) 5 in the interchangeable lens 2 and forms an image in the vicinity of a two-dimensional imaging device (hereinafter referred to as an image sensor) 7. The imaging lens 5 is provided with a diaphragm 8 having a variable aperture diameter.
　イメージセンサ７は、ＣＣＤセンサやＣＭＯＳセンサ等であり、複数の画素が２次元に配置された光電変換素子である。撮像レンズ５とイメージセンサ７との間には、光学ローパスフィルタ６が配置されている。光学ローパスフィルタ６は、撮像レンズ５から光束に対して、その集光点を複数に分離することでローパス効果を付与する。光学ローパスフィルタ６とイメージセンサ７は、イメージセンサユニット（撮像素子ユニット）として一体化されていてもよい。 The image sensor 7 is a CCD sensor, a CMOS sensor, or the like, and is a photoelectric conversion element in which a plurality of pixels are two-dimensionally arranged. An optical low-pass filter 6 is disposed between the imaging lens 5 and the image sensor 7. The optical low-pass filter 6 imparts a low-pass effect to the light flux from the imaging lens 5 by separating its condensing point into a plurality of points. The optical low-pass filter 6 and the image sensor 7 may be integrated as an image sensor unit (imaging element unit).
　図２は、光学ローパスフィルタ６を模式的に示している。図１からも分かるように、光学ローパスフィルタ６は、イメージセンサ７より一回り大きく、イメージセンサ７と相似な長方形状を有する平行平板として形成された複屈折素子１により構成されている。複屈折素子１は、ニオブ酸リチウム等の一軸性結晶からなる複屈折性材料により形成されている。 FIG. 2 schematically shows the optical low-pass filter 6. As can be seen from FIG. 1, the optical low-pass filter 6 is configured by a birefringent element 1 that is slightly larger than the image sensor 7 and formed as a parallel plate having a rectangular shape similar to the image sensor 7. The birefringent element 1 is formed of a birefringent material made of a uniaxial crystal such as lithium niobate.
　以下の説明では、図２に示すように、光学ローパスフィルタ６（およびイメージセンサ７）の長辺に平行なｘ軸が延びる方向をｘ軸方向（水平方向）といい、短辺に平行なｙ軸が延びる方向をｙ軸方向（垂直方向）という。ｘ軸とｙ軸を含む面、すなわちイメージセンサ７の撮像面に平行な面はｘｙ面である。また、ｘｙ面（撮像面）に直交する軸をｚ軸といい、該ｚ軸が延びる方向をｚ軸方向という。ｚ軸方向は、撮像レンズ５の光軸Ｏが延びる方向である光軸方向であり、言い換えれば、光学ローパスフィルタ６から見てイメージセンサ７が配置された第１の方向としてのフォーカス方向である。また、ｘ軸方向およびｙ軸方向は、イメージセンサ７の撮像面に沿う（平行な）第２の方向としての横方向である。 In the following description, as shown in FIG. 2, the direction in which the x-axis extending in parallel to the long side of the optical low-pass filter 6 (and image sensor 7) is referred to as the x-axis direction (horizontal direction), and y parallel to the short side. The direction in which the axis extends is called the y-axis direction (vertical direction). A plane including the x axis and the y axis, that is, a plane parallel to the imaging surface of the image sensor 7 is an xy plane. An axis orthogonal to the xy plane (imaging plane) is referred to as a z-axis, and a direction in which the z-axis extends is referred to as a z-axis direction. The z-axis direction is the optical axis direction in which the optical axis O of the imaging lens 5 extends, in other words, the focus direction as the first direction in which the image sensor 7 is disposed when viewed from the optical low-pass filter 6. . Further, the x-axis direction and the y-axis direction are lateral directions as a second direction along (parallel to) the imaging surface of the image sensor 7.
　なお、図２では、複屈折素子１のｚ軸方向での厚みを実際の厚みである数１００μｍ程度によりも厚く示している。 In FIG. 2, the thickness of the birefringent element 1 in the z-axis direction is shown to be thicker than the actual thickness of several hundreds of μm.
　図３（ａ），（ｂ）に示すように、一軸性結晶としての複屈折素子１の光学軸１ａは、ｘ軸方向に延びるとともに、該複屈折素子１の入射面１ｂに平行に延びている。すなわち、図３（ａ）に示すように、光学軸１ａのｘ軸に対するｚ軸回り（反時計回り）の方位角θは０°であり、入射面１ｂの法線であるｚ軸に対する角φは９０°である。 As shown in FIGS. 3A and 3B, the optical axis 1a of the birefringent element 1 as a uniaxial crystal extends in the x-axis direction and extends in parallel to the incident surface 1b of the birefringent element 1. Yes. That is, as shown in FIG. 3A, the azimuth angle θ around the z axis (counterclockwise) with respect to the x axis of the optical axis 1a is 0 °, and the angle φ with respect to the z axis that is the normal line of the incident surface 1b. Is 90 °.
　従来の一般的な光学ローパスフィルタでは、ｚ軸に対する光学軸の角φを４５°±２０°程度を設定する。これにより、入射した光線（点像）が常光線と異常光線とに横方向に分離する。入射した光線を常光線が形成する点像と異常光線が形成する点像に分離することを、以下の説明では点像分離という。 In the conventional general optical low-pass filter, the angle φ of the optical axis with respect to the z-axis is set to about 45 ° ± 20 °. Thereby, the incident light ray (point image) is separated into an ordinary ray and an extraordinary ray in the lateral direction. The separation of the incident light beam into a point image formed by an ordinary light beam and a point image formed by an extraordinary light beam is referred to as point image separation in the following description.
　これに対して、本実施例の光学ローパスフィルタ６（複屈折素子１）では、ｚ軸に対する光学軸１ａの角φを９０°とする、すなわち光学軸１ａを入射面１ｂと平行に配置している。このため、横方向での点像分離は生じず、フォーカス方向（ｚ軸方向）での点像分離のみが生じる。 On the other hand, in the optical low-pass filter 6 (birefringent element 1) of the present embodiment, the angle φ of the optical axis 1a with respect to the z axis is set to 90 °, that is, the optical axis 1a is arranged in parallel to the incident surface 1b. Yes. For this reason, point image separation in the horizontal direction does not occur, and only point image separation in the focus direction (z-axis direction) occurs.
　図４を用いて、フォーカス方向での点像分離が発生する仕組みについて説明する。集光する光束の光路中に厚さｄの平行平板を挿入すると、挿入前における該光束の集光点の位置、すなわち結像位置（以下、ピント位置という）は挿入前のＰ０からフォーカス方向としての後方のＰ１にずれる。これは、屈折率が１である空気が屈折率ｎの媒質に置き換わることで、空気中での光路長ｄがｄ／ｎとなるためである。 The mechanism of point image separation in the focus direction will be described with reference to FIG. When a parallel plate having a thickness d is inserted in the optical path of the condensed light beam, the position of the light condensing point before the insertion, that is, the imaging position (hereinafter referred to as the focus position) is set as the focus direction from P0 before the insertion. It shifts to P1 behind. This is because air having a refractive index of 1 is replaced with a medium having a refractive index n, so that the optical path length d in the air becomes d / n.
　平行平板の屈折率ｎによるピント位置のずれ量Δは次式によって与えられる。
Δ＝ｄ（１－１／ｎ）
　例えば、平行平板の厚さｄが０．８２ｍｍで、屈折率ｎが２．３２４７である場合は、Δ＝４６７μｍとなる。 The shift amount Δ of the focus position due to the refractive index n of the parallel plate is given by the following equation.
Δ = d (1-1 / n)
For example, when the thickness d of the parallel plate is 0.82 mm and the refractive index n is 2.3247, Δ = 467 μm.
　次に、図５を用いて一軸性結晶であるニオブ酸リチウム（ＬＮ）を光路に挿入した場合について説明する。ＬＮでは、常光線と異常光線とで屈折率が異なるため、常光線のピント位置と異常光線のピント位置とがともＬＮ挿入前のＰ０から後方にずれるだけでなく、互いに異なる後方位置であるＰ１とＰ２にずれる。すなわち、ピント位置がフォーカス方向において二つに分離される。 Next, a case where lithium niobate (LN), which is a uniaxial crystal, is inserted into the optical path will be described with reference to FIG. In LN, since the refractive index is different between the ordinary ray and the extraordinary ray, the focus position of the ordinary ray and the focus position of the extraordinary ray are not only shifted backward from P0 before insertion of the LN, but are also different rear positions P1. And shift to P2. That is, the focus position is separated into two in the focus direction.
　ＬＮでは、常光線に対する屈折率ｎｏ＝２．３２４７、異常光線に対する屈折率ｎｅ＝２．２３５５と常光線に対する屈折率の方が高い。このため、常光線のピント位置Ｐ１が異常光線のピント位置Ｐ２よりも後方にずれる。 In LN, the refractive index no for normal light is 2.3247, the refractive index for extraordinary light is ne = 2.2355, and the refractive index for ordinary light is higher. For this reason, the focus position P1 of the ordinary ray is shifted rearward from the focus position P2 of the extraordinary ray.
　常光線のピント位置と異常光線のピント位置のフォーカス方向でのずれ量（以下、ピント位置分離量という）Ｌは、以下の式で与えられる。
Ｌ＝ｄ（１／ｎｅ－１／ｎｏ）
　図４の例と同様にＬＮの厚さｄを０．８２ｍｍとした場合は、Ｌ＝０．０１４ｍｍ（１４μｍ）となる。 A shift amount (hereinafter referred to as a focus position separation amount) L in the focus direction between the focus position of the ordinary ray and the focus position of the extraordinary ray is given by the following equation.
L = d (1 / ne-1 / no)
Similarly to the example of FIG. 4, when the thickness d of LN is 0.82 mm, L = 0.014 mm (14 μm).
　図６は、撮像レンズ５のＦ値ごとの常光線と異常光線のピント位置のずれにより生ずるボケ像のサイズ（以下、ボケ径という）を示している。なお、平行平板の挿入によってピント位置がずれても各Ｆ値での光束の集光角度は変わらないため、異常光線より後方に結像する常光線は、異常光線の光路を後方に移動させた光路を進む。 FIG. 6 shows the size of a blurred image (hereinafter referred to as a blurred diameter) caused by the deviation of the focus position between the ordinary ray and the extraordinary ray for each F value of the imaging lens 5. Note that even if the focus position is shifted due to the insertion of the parallel plate, the converging angle of the light flux at each F value does not change. Follow the light path.
　図６から分かるように、常光線のピント位置Ｐ１と異常光線のピント位置Ｐ２との中間点で、光束の径であるボケ径δｄｅｆが最小になる。このボケ径δｄｅｆは、Ｆ値をＦとし、常光線と異常光線のピント位置の分離量をＬとすると、幾何学的に以下の式で求めることができる。
δｄｅｆ＝Ｌ／（２Ｆ）
　図６では、Ｆ＝１．８、２．８および５．６の場合を示している。これらのＦ値に対するδｄｅｆはそれぞれ、３．８９、２．５６および１．２５μｍである。図６に灰色の丸で示すように、ボケ径はＦ値が大きいほど小さくなる。 As can be seen from FIG. 6, the blur diameter δdef, which is the diameter of the light beam, is minimized at an intermediate point between the focus position P1 of the ordinary ray and the focus position P2 of the extraordinary ray. This blur diameter δdef can be obtained geometrically by the following equation, where F is F and L is the separation amount of the focus position of ordinary and extraordinary rays.
δdef = L / (2F)
FIG. 6 shows the cases of F = 1.8, 2.8, and 5.6. The δdef for these F values is 3.89, 2.56 and 1.25 μm, respectively. As shown by the gray circles in FIG. 6, the blur diameter decreases as the F value increases.
　一方、撮像レンズ５内の絞り８を開放から絞り込むことによる解像度の劣化（像劣化）については、撮像レンズ５に収差がない場合での最小錯乱円径δＦｎｏとして表される。最小錯乱円径δＦｎｏは、例えばレーリーの解像限界を用いると以下の式で表される。
δＦｎｏ＝１．２２Ｆλ
　図６に白丸で示すように、Ｆ＝１．８、２．８および５．６の場合のδＦｎｏはそれぞれ、１．２０７、１．８７８および３．７５７μｍである。Ｆ値が大きいほど最小錯乱円径（つまりはボケ径）δＦｎｏが大きくなる。 On the other hand, the resolution degradation (image degradation) caused by narrowing the aperture 8 in the imaging lens 5 from the open is expressed as the minimum circle of confusion δFno when the imaging lens 5 has no aberration. The minimum circle of confusion δFno is expressed by the following equation using, for example, the Rayleigh resolution limit.
δFno = 1.22Fλ
As indicated by white circles in FIG. 6, δFno for F = 1.8, 2.8, and 5.6 are 1.207, 1.878, and 3.757 μm, respectively. As the F value increases, the minimum circle of confusion (that is, the blur diameter) δFno increases.
　絞り８を開放から絞り込むに従って回折による像劣化が増加するが、それとは逆に複屈折素子１により生じるボケ径は小さくなり、全体として像劣化を抑えることができる。 The image degradation due to diffraction increases as the aperture 8 is narrowed from the open position. On the contrary, the blur diameter generated by the birefringent element 1 is reduced, and the image degradation can be suppressed as a whole.
　像劣化としては、複屈折素子１（光学ローパスフィルタ６）によるボケ径δｄｅｆと、回折による最小錯乱円径δＦｎｏとしてのボケ径とを重畳したものを考える必要がある。ボケ径δｄｅｆと最小錯乱円径δＦｎｏとを円で近似した場合に、これらδｄｅｆとδＦｎｏの合成により生ずる合成ボケ径δａｌｌは、経験的に以下の式で得られる。
δａｌｌ＝√（δｄｅｆ２＋δＦｎｏ２）
　このδａｌｌを撮像光学系５に対する光学ローパスフィルタ６の効き量として扱うことで、光学ローパスフィルタ６を設計することができる。 As image degradation, it is necessary to consider a superposition of the blur diameter δdef due to the birefringent element 1 (optical low-pass filter 6) and the blur diameter as the minimum circle of confusion δFno due to diffraction. When the blur diameter δdef and the minimum circle of confusion δFno are approximated by a circle, the synthesized blur diameter δall generated by the synthesis of δdef and δFno is obtained empirically by the following equation.
δall = √ (δdef 2 + δFno 2 )
By treating this δall as the effect of the optical low-pass filter 6 on the imaging optical system 5, the optical low-pass filter 6 can be designed.
　図７は、Ｆ値ごとの回折による最小錯乱円径δＦｎｏ（短破線）としてのボケ径と、光学ローパスフィルタ６によりフォーカス方向にて発生したボケ径（以下、フォーカス方向ボケ径という）δｄｅｆ（長破線）と、これらが合成されることで得られる合成ボケ径δａｌｌ（実線）とを示す。図７は、イメージセンサ７の画素ピッチＰが４．２μｍである場合を示している。このイメージセンサ７に対して偽色やモアレの除去に必要な目標ボケ径は約３．０μｍである。この目標ボケ径を図７中に点線で示している。 7 shows a blur diameter as a minimum circle of confusion δFno (short broken line) due to diffraction for each F value and a blur diameter generated in the focus direction by the optical low-pass filter 6 (hereinafter referred to as a focus direction blur diameter) δdef (long). A broken line) and a synthetic blur diameter δall (solid line) obtained by combining them are shown. FIG. 7 shows a case where the pixel pitch P of the image sensor 7 is 4.2 μm. The target blur diameter necessary for removing false colors and moire from the image sensor 7 is about 3.0 μm. This target blur diameter is indicated by a dotted line in FIG.
　前述したように、フォーカス方向ボケ径δｄｅｆは、開放Ｆ値側では大きく、小絞りＦ値側では小さくなる。合成ボケ径δａｌｌは、Ｆ値２．８から４．０においては目標ボケ径である３．０μｍとなっている。また、Ｆ値２．０から１．８においては、合成ボケ径δは４．０μｍと目標ボケ径よりやや大きめとなっており、強めのローパス効果が生じている。 As described above, the focus direction blur diameter δdef is large on the open F value side and small on the small aperture F value side. The synthetic blur diameter δall is 3.0 μm which is the target blur diameter when the F value is 2.8 to 4.0. Further, in the F values from 2.0 to 1.8, the synthetic blur diameter δ is 4.0 μm, which is slightly larger than the target blur diameter, and a strong low-pass effect is generated.
　しかし、Ｆ値５．６以上では、合成ボケ径δａｌｌは最小錯乱円径δＦｎｏ（＝１．２２Ｆλ）に漸近している。つまり、回折による像劣化以上の像劣化は生じない。 However, when the F value is 5.6 or more, the synthetic blur diameter δall is asymptotic to the minimum circle of confusion δFno (= 1.22Fλ). In other words, image degradation more than image degradation due to diffraction does not occur.
　図８は、図７に対する比較例として、横方向での点像分離のみを行う従来の光学ローパスフィルタを用いた場合を示す。図８は、Ｆ値ごとの回折による最小錯乱円径δＦｎｏ（短破線）としてのボケ径と、横方向で発生したボケ径である横ボケ径δｘ（長破線）と、これらの合成により生ずる合成ボケ径δａｌｌ（実線）とを示す。画素ピッチは４．２μｍである。 FIG. 8 shows a case where a conventional optical low-pass filter that performs only point image separation in the horizontal direction is used as a comparative example with respect to FIG. FIG. 8 shows a blur diameter as a minimum circle of confusion δFno (short broken line) due to diffraction for each F value, a horizontal blur diameter δx (long broken line) which is a blur diameter generated in the horizontal direction, and a composition generated by combining these. The blur diameter δall (solid line) is shown. The pixel pitch is 4.2 μm.
　従来の光学ローパスフィルタは、図１５に示すように、入射した光束を横方向に４つの点像Ｐ１１～Ｐ１４に分離し、Ｆ値にかかわらず一定のボケ像を形成する。この光学ローパスフィルタは、最も高い解像度が得られる開放Ｆ値において約３．０μｍの目標ボケ径を生じさせる。 As shown in FIG. 15, the conventional optical low-pass filter separates the incident light beam into four point images P11 to P14 in the lateral direction, and forms a constant blurred image regardless of the F value. This optical low-pass filter produces a target blur diameter of about 3.0 μm at the open F value where the highest resolution is obtained.
　開放Ｆ値１．８では合成ボケ径δａｌｌはほぼ目標ボケ径となっているが、Ｆ値２．８から４．０では大き過ぎて必要以上に像劣化を生じさせている。さらにＦ値５．６では、Ｆ値８．０のときの回折による解像度劣化に相当するボケ径（δＦｎｏ）が生じている。このような過剰なボケは、Ｆ値１１．３程度まで付加されている。 When the open F value is 1.8, the composite blur diameter δall is almost the target blur diameter, but the F value of 2.8 to 4.0 is too large, causing image degradation more than necessary. Further, at the F value of 5.6, a blur diameter (δFno) corresponding to the resolution degradation due to diffraction at the F value of 8.0 occurs. Such excessive blur is added to an F value of about 11.3.
　従来、いわゆる小絞り回折による解像度劣化は広く知られており、特に風景を撮像する場合のように高解像度を必要とする撮像シーンにおいては、Ｆ値１１を超えない範囲で撮像を行うことが推奨されている。しかし、図８から分かるように、一般に高解像度の撮像が可能であるＦ値８．０においても光学ローパスフィルタの影響を大きく受けていたことになる。 Conventionally, resolution degradation due to so-called small-aperture diffraction is widely known, and it is recommended to perform imaging within a range that does not exceed F-number 11, particularly in an imaging scene that requires high resolution, such as when capturing a landscape. Has been. However, as can be seen from FIG. 8, even at an F value of 8.0, which is generally capable of high-resolution imaging, it was greatly influenced by the optical low-pass filter.
　これに対して、本実施例の光学ローパスフィルタ６を用いることで、高解像度の撮像が可能であるＦ値での光学ローパスフィルタによる像劣化を軽減する。 On the other hand, by using the optical low-pass filter 6 of the present embodiment, image deterioration due to the optical low-pass filter at the F value that enables high-resolution imaging can be reduced.
　以下、本発明の実施例が満足すべき又は満足することが望ましい条件について説明する。まず、以下の式（１）および式（２）で示す条件を満足すべきである。
０．２≦（δｄｅｆ２＋δｘ２）／Ｐ２≦０．９　　　（１）
δｘ≦δｄｅｆ　　　　　　　　　　　　　　　　　（２）
Ｐはイメージセンサ７の画素ピッチ［μｍ］であり、δｄｅｆはフォーカス方向ボケ径［μｍ］である。また、δｘは横ボケ径である。なお、横方向の点像分離が行われない本実施例では、δｘ＝０である。発明者は、光学ローパスフィルタのローパス効果が式（１）中の（δｄｅｆ２＋δｘ２）／Ｐ２、すなわちデフォーカスボケ径δｄｅｆの二乗と横ボケ径δｘの二乗との合計（δｄｅｆ２＋δｘ２）と画素ピッチＰの二乗との比で与えられると経験的に認識している。 Hereinafter, conditions that should be satisfied or desirable for the embodiments of the present invention will be described. First, the conditions shown by the following formulas (1) and (2) should be satisfied.
0.2 ≦ (δdef 2 + δx 2 ) / P 2 ≦ 0.9 (1)
δx ≦ δdef (2)
P is the pixel pitch [μm] of the image sensor 7, and δdef is the focus direction blur diameter [μm]. Further, δx is a horizontal blur diameter. In the present embodiment in which point image separation in the horizontal direction is not performed, δx = 0. The inventor found that the low-pass effect of the optical low-pass filter is (δdef 2 + δx 2 ) / P 2 in Equation (1), that is, the sum of the square of the defocus blur diameter δdef and the square of the lateral blur diameter δx (δdef 2 + δx 2 ) And the square of the pixel pitch P is empirically recognized.
　式（１）の上限値は、一般に光学ローパスフィルタはイメージセンサの画素ピッチ以下のボケ径を生じさせるため、それを示すものである。（δｄｅｆ２＋δｘ２）／Ｐ２の値がこの上限値を上回ると、過度の像劣化を生じさせることになり、好ましくない。また、式（１）の下限値は、ローパス効果を弱めるときの限界値を示す。（δｄｅｆ２＋δｘ２）／Ｐ２の値がこの下限値を下回るほどに小さいボケ径を生じさせるだけでは、ローパス効果が弱すぎて、偽色やモアレの除去効果が期待できない。式（１）を用いてボケ量（ボケ径）の合計を制限することで、フォーカス方向での点像分離によりボケを発生させる場合でも、容易に従来の光学ローパスフィルタにより発生するボケ量に相当するボケ量を生じさせることができる。 The upper limit value of the expression (1) indicates that the optical low-pass filter generally generates a blur diameter that is less than or equal to the pixel pitch of the image sensor. If the value of (δdef 2 + δx 2 ) / P 2 exceeds the upper limit value, excessive image deterioration is caused, which is not preferable. Further, the lower limit value of the expression (1) indicates a limit value when the low pass effect is weakened. If a blur diameter that is so small that the value of (δdef 2 + δx 2 ) / P 2 falls below this lower limit value, the low-pass effect is too weak, and a false color or moire removal effect cannot be expected. By limiting the total amount of blurring (blurring diameter) using Equation (1), even when blurring is generated by point image separation in the focus direction, it easily corresponds to the blurring amount generated by the conventional optical low-pass filter. The amount of blur to be generated can be generated.
　本実施例では、撮像レンズ５からの集光光束は、ＬＮの平行平板である光学ローパスフィルタ６により常光線と異常光線とに分離されるとともに、それらのピント位置がフォーカス方向に分離される。本実施例ではフォーカス方向でのピント位置の分離量Ｌは１４μｍであり、開放Ｆ値はＦ＝１．８であるため、フォーカス方向ボケ径δｄｅｆは３．８９μｍとなる。一方、前述したように横方向には点像分離が行われないので、δｘ＝０となる。そして、（δｄｅｆ２＋δｘ２）／Ｐ２の値は０．８５となり、式（１）の条件を満足している。この条件を満足することで、光学ローパスフィルタ６は、イメージセンサ７の画素ピッチに対して適切なローパス効果を与える。
さらに、以下の式（１ａ）を満足することがより好ましい。
０．３≦（δｄｅｆ２＋δｘ２）／Ｐ２≦０．８８　　　（１ａ）
　なお、フォーカス方向ボケ径δｄｅｆは、以下の式（３）により近似することができる。
δｄｅｆ≒Ｄσ／Ｆ　　　（３）
　つまり、δｄｅｆを、δｄｅｆ＝Ｄσ／Ｆを用いて求めてもよい。 In this embodiment, the condensed light flux from the imaging lens 5 is separated into an ordinary ray and an extraordinary ray by an optical low-pass filter 6 that is a parallel plate of LN, and the focus position thereof is separated in the focus direction. In this embodiment, the focus position separation amount L in the focus direction is 14 μm, and the open F value is F = 1.8, so the focus direction blur diameter δdef is 3.89 μm. On the other hand, since point image separation is not performed in the horizontal direction as described above, δx = 0. The value of (δdef 2 + δx 2 ) / P 2 is 0.85, which satisfies the condition of formula (1). By satisfying this condition, the optical low-pass filter 6 gives an appropriate low-pass effect to the pixel pitch of the image sensor 7.
Furthermore, it is more preferable to satisfy the following formula (1a).
0.3 ≦ (δdef 2 + δx 2 ) / P 2 ≦ 0.88 (1a)
The focus direction blur diameter δdef can be approximated by the following equation (3).
δdef≈Dσ / F (3)
That is, δdef may be obtained using δdef = Dσ / F.
　さらに発明者は、ローパス効果の強度が以下の式（４）中のＤσ／（Ｆ・Ｐ）、すなわち撮像レンズ５の開放Ｆ値であるＦとフォーカス方向でのピント位置の分離量の標準偏差Ｄσ［μｍ］とを用いて与えられることも経験的に認識している。回折による像劣化は、主として画素ピッチＰが５．０μｍ以下の場合に発生する。また、この場合のボケ量としては、フォーカス方向のボケ量を主に考えればよく、横方向のボケ量は無視してもよい。このため、以下の式（４）で示す条件を満足すること望ましい。
０．３≦Ｄσ／（Ｆ・Ｐ）≦１．０　　　（４）
Ｐ≦５．０［μｍ］
　Ｄσ／（Ｆ・Ｐ）の値が式（４）の下限値より小さいと、フォーカス方向のピントの分離量が少なく、高ＦＮｏによる回折影響の補正に対する効果が少なくなりすぎるので好ましくない。また、Ｄσ／（Ｆ・Ｐ）の値が式（４）の上限値を超えると、高ＦＮｏによる回折の補正効果は大きいが、ボケ像自体が大きくなってローパス効果が強すぎて画像が劣化する不具合があるので好ましくない。 Furthermore, the inventor shows that the intensity of the low-pass effect is Dσ / (F · P) in the following expression (4), that is, the standard deviation of the separation amount of the focus position in the focus direction from F which is the open F value of the imaging lens 5. It is also empirically recognized that it is given using Dσ [μm]. Image degradation due to diffraction occurs mainly when the pixel pitch P is 5.0 μm or less. Further, as the amount of blur in this case, the amount of blur in the focus direction may be mainly considered, and the amount of blur in the horizontal direction may be ignored. For this reason, it is desirable to satisfy the conditions shown by the following formula (4).
0.3 ≦ Dσ / (F · P) ≦ 1.0 (4)
P ≦ 5.0 [μm]
If the value of Dσ / (F · P) is smaller than the lower limit value of the equation (4), the amount of focus separation in the focus direction is small, and the effect of correcting the diffraction effect due to high FNo is too small, which is not preferable. Also, if the value of Dσ / (F · P) exceeds the upper limit of equation (4), the diffraction correction effect due to high FNo is large, but the blurred image itself becomes large and the low-pass effect is too strong and the image deteriorates. It is not preferable because there is a problem to do.
　本実施例では、Ｄσ／（Ｆ・Ｐ）の値は０．９２６であり、式（４）の条件を満足している。 In this example, the value of Dσ / (F · P) is 0.926, which satisfies the condition of the formula (4).
　さらに、以下の式（４ａ）を満足することがより好ましい。
０．４≦Ｄσ／（Ｆ・Ｐ）≦０．９５　　　（４ａ）
　また、撮像レンズ５の開放Ｆ値であるＦは、以下の式（５）で示す条件を満足することが望ましい。
１．０≦Ｆ≦２．０　　（５）
　式（５）で示す条件を満足する開放Ｆ値においては、解像度がイメージセンサ７の画素ピッチの２倍よりも高くなり、特に偽色とモアレの発生が懸念される。式（５）の上限値を上回る開放Ｆ値を有する撮像レンズ５では、回折による像劣化のために偽色やモアレの心配はなく、光学ローパスフィルタ６は不要である。 Furthermore, it is more preferable that the following formula (4a) is satisfied.
0.4 ≦ Dσ / (F · P) ≦ 0.95 (4a)
Further, it is desirable that F, which is the open F value of the imaging lens 5, satisfies the condition expressed by the following expression (5).
1.0 ≦ F ≦ 2.0 (5)
In the open F value that satisfies the condition expressed by Expression (5), the resolution is higher than twice the pixel pitch of the image sensor 7, and there is a concern that false colors and moire are generated. In the imaging lens 5 having an open F value that exceeds the upper limit value of Expression (5), there is no concern about false color or moire due to image degradation due to diffraction, and the optical low-pass filter 6 is unnecessary.
　なお、撮像レンズ５が交換レンズに設けられている場合は、
Ｆ＝１．４
と設定してボケ量を決定することが望ましい。 When the imaging lens 5 is provided on the interchangeable lens,
F = 1.4
It is desirable to determine the amount of blur by setting.
　また、前述したように本実施例において複屈折素子１の光学軸１ａが入射面１ｂの法線に対してなす角度（方位角）φは９０°であるが、角φは以下の式（６）で示す範囲の角度であってもよい。
８０°≦φ＜９０°　　（６）
　この角φが９０°より小さい場合については、以下の実施例２で説明する。 As described above, in this embodiment, the angle (azimuth angle) φ formed by the optical axis 1a of the birefringent element 1 with respect to the normal line of the incident surface 1b is 90 °. The angle in the range indicated by
80 ° ≦ φ <90 ° (6)
The case where the angle φ is smaller than 90 ° will be described in Example 2 below.
　図９は、本発明の実施例２である光学ローパスフィルタ６’の構成を示している。本実施例の光学ローパスフィルタ６’は、図１に示したカメラ１に光学ローパスフィルタ６に代えて用いられる。光学ローパスフィルタ６’は、イメージセンサ７より一回り大きく、イメージセンサ７と相似な長方形状を有する複数（本実施例では３つ）の平行平板として形成された第１の複屈折素子２と第２の複屈折素子４がｚ軸方向に積層されて構成されている。第１の複屈折素子２と第２の複屈折素子４との間には、第３の複屈折素子３が配置されている。 FIG. 9 shows a configuration of an optical low-pass filter 6 ′ that is Embodiment 2 of the present invention. The optical low-pass filter 6 ′ of this embodiment is used in place of the optical low-pass filter 6 in the camera 1 shown in FIG. 1. The optical low-pass filter 6 ′ is slightly larger than the image sensor 7 and has a first birefringent element 2 formed as a plurality of (three in this embodiment) parallel plates having a rectangular shape similar to the image sensor 7. Two birefringent elements 4 are stacked in the z-axis direction. A third birefringent element 3 is arranged between the first birefringent element 2 and the second birefringent element 4.
　第１、第２および第３の複屈折素子２，４，３はそれぞれ、ニオブ酸リチウム等の一軸性結晶からなる複屈折性材料により形成されている。なお、図９では第１、第２および第３の複屈折素子２，４，３を互いに離して示しているが、これらは実際には互いに接している。また、図９では、各複屈折素子のｚ軸方向での厚みを実際の数１００μｍ程度よりも厚く示している。 The first, second and third birefringent elements 2, 4 and 3 are each formed of a birefringent material made of a uniaxial crystal such as lithium niobate. In FIG. 9, the first, second and third birefringent elements 2, 4 and 3 are shown separated from each other, but they are actually in contact with each other. In FIG. 9, the thickness of each birefringent element in the z-axis direction is shown to be thicker than the actual several hundred μm.
　第１および第２の複屈折素子２，４の光学軸２ａ，４ａのｘ軸に対するｚ軸回り（反時計回り）の方位角θ２，θ４はそれぞれ９０°と０°である。また、光学軸２ａ，４ａの第１および第２の複屈折素子２，４の入射面２ｂ，４ｂの法線であるｚ軸（つまりはフォーカス方向）に対する角φ２，φ４は、上述した式（６）の範囲に設定されている。角φ２，φ４は互いに同じであってもよいし、互いに異なっていてもよい。なお、第１および第２の複屈折素子２，４の積層順は図示の通り光入射側からこの順であってもよいし、逆であってもよい。 The azimuth angles θ2 and θ4 around the z axis (counterclockwise) with respect to the x axis of the optical axes 2a and 4a of the first and second birefringent elements 2 and 4 are 90 ° and 0 °, respectively. Further, the angles φ2 and φ4 with respect to the z axis (that is, the focus direction) that is the normal line of the incident surfaces 2b and 4b of the first and second birefringent elements 2 and 4 of the optical axes 2a and 4a are expressed by the above-described formula ( 6) is set. The angles φ2 and φ4 may be the same as each other or different from each other. It should be noted that the order in which the first and second birefringent elements 2 and 4 are stacked may be in this order from the light incident side as shown or may be reversed.
　第３の複屈折素子３は、偏光解消用の位相板であり、その光学軸３ａの方位角θは１３５°である。なお、方位角θにおいて１３５°は４５°と等価である。第３の複屈折素子３がこれを通過する波長λの光に与える位相差をλ／４とすることで、該第１の複屈折素子２で分離された常光線と異常光線の二つの直線偏光をそれぞれ円偏光に変換することで偏光解消を行う。この場合、色消しの１／４λ板を用いる。 The third birefringent element 3 is a depolarizing phase plate, and the azimuth angle θ of the optical axis 3a is 135 °. In the azimuth angle θ, 135 ° is equivalent to 45 °. The phase difference that the third birefringent element 3 gives to the light having the wavelength λ that passes through the third birefringent element 3 is set to λ / 4, so that two straight lines of the ordinary ray and the extraordinary ray separated by the first birefringent element 2 are obtained. Depolarization is performed by converting each polarized light into circularly polarized light. In this case, an achromatic 1 / 4λ plate is used.
　また、第３の複屈折素子３として、λ以上の位相差を与える位相差板を用いてもよい。このような位相差板を用いると、波長が異なる光が受ける位相差が互いに大きく異なることになる。このため、直線偏光の偏光方向の回転量が波長ごとに異なり、広い波長域で平均すると偏光解消の効果が得られる。本実施例では、第３の複屈折素子３として、０．２～０．４ｍｍ程度の厚みの水晶板を用いることで、各波長の常光線と異常光線のそれぞれにλ以上の位相差を与える。 Further, as the third birefringent element 3, a retardation plate that gives a phase difference of λ or more may be used. When such a retardation plate is used, the phase differences received by light having different wavelengths are greatly different from each other. For this reason, the amount of rotation of the polarization direction of linearly polarized light differs depending on the wavelength, and if it is averaged over a wide wavelength range, the effect of depolarization can be obtained. In this embodiment, a quartz plate having a thickness of about 0.2 to 0.4 mm is used as the third birefringent element 3 to give a phase difference of λ or more to each of ordinary and extraordinary rays of each wavelength. .
　以上のように構成された３層構造の光学ローパスフィルタ６’での点像分離（ローパス効果）について、図１０（ａ）～（ｄ）および図１１（ａ）～（ｄ）を用いて説明する。図１０（ａ）～（ｄ）は、第１の複屈折素子２を透過した光束についての点像分離を示す。図１１（ａ）～（ｄ）は、第１および第２の複屈折素子２，４を透過した光束についての点像分離を示す。図１０（ａ），（ｂ）はそれぞれ、ｙ軸方向とｘ軸方向から見た光束の分離を示す。図１１（ａ），（ｂ）も同様である。図１０（ａ），（ｂ）に対応する図１０（ｃ），（ｄ）はそれぞれ、後述する評価面でのボケ像をｚ軸方向から見て示す。図１１（ａ），（ｂ）に対応する図１１（ｃ），（ｄ）も同様である。図１０（ａ），（ｂ）と図１１（ａ），（ｂ）では、常光線の光束を実線で、異常光線の光束を破線で示す。図１０（ａ）～（ｄ）と図１１（ａ）～（ｄ）では、常光線が形成する点像を黒丸で、異常光線が形成する点像を白丸で示す。 Point image separation (low-pass effect) in the optical low-pass filter 6 ′ having the three-layer structure configured as described above will be described with reference to FIGS. 10 (a) to 10 (d) and FIGS. 11 (a) to 11 (d). To do. 10 (a) to 10 (d) show point image separation for a light beam transmitted through the first birefringent element 2. FIG. FIGS. 11A to 11D show point image separation for light beams transmitted through the first and second birefringent elements 2 and 4. FIGS. 10A and 10B show the separation of the light beams as seen from the y-axis direction and the x-axis direction, respectively. The same applies to FIGS. 11A and 11B. FIGS. 10C and 10D corresponding to FIGS. 10A and 10B show a blurred image on an evaluation surface, which will be described later, as seen from the z-axis direction. The same applies to FIGS. 11 (c) and 11 (d) corresponding to FIGS. 11 (a) and 11 (b). 10 (a) and 10 (b) and FIGS. 11 (a) and 11 (b), the ordinary light beam is indicated by a solid line, and the extraordinary light beam is indicated by a broken line. In FIGS. 10A to 10D and FIGS. 11A to 11D, point images formed by ordinary rays are indicated by black circles, and point images formed by extraordinary rays are indicated by white circles.
　図１０（ａ），（ｂ）において、第１の複屈折素子２を通過した光束は、通常の光学ローパスフィルタを通過した場合と同様に、該第１の複屈折素子２の光学軸２ａの方位角θの方向であるｙ軸方向に二つに分離する。その際、常光線は第１の複屈折素子２内を真っ直ぐに通過し、異常光線はｙ軸方向にシフトする。この結果、第１の複屈折素子２を通過した常光線と異常光線のピント位置Ｐ１，Ｐ２はｙ軸方向にずれる。本実施例では、このピント位置Ｐ１，Ｐ２のｙ軸方向でのずれ量を１．０μｍに設定している。 10 (a) and 10 (b), the light beam that has passed through the first birefringent element 2 passes through the optical axis 2a of the first birefringent element 2 in the same manner as when it passes through a normal optical low-pass filter. They are separated into two in the y-axis direction which is the direction of the azimuth angle θ. At that time, the ordinary ray passes straight through the first birefringent element 2, and the extraordinary ray shifts in the y-axis direction. As a result, the focus positions P1 and P2 of the ordinary ray and the extraordinary ray that have passed through the first birefringent element 2 are shifted in the y-axis direction. In this embodiment, the shift amount in the y-axis direction of the focus positions P1 and P2 is set to 1.0 μm.
　通常の光学ローパスフィルタにおいては、常光線と異常光線のピント位置のｙ軸方向でのずれ量しか考慮していない。これに対して、本実施例では、図１０（ａ），（ｂ）に示すように、ピント位置Ｐ１，Ｐ２はフォーカス方向にもフォーカス方向ずれ量Ｌ＝３．５３５μｍだけ互いにずれる。これは、第１の複屈折素子２の常光線と異常光線に対する屈折率差によるものである。このとき、常光線の光束と異常光線の光束が形成する合成されたボケ像のサイズはピント位置Ｐ１，Ｐ２の中間の位置で最小となる。この中間の位置にある面を評価面とする。評価面では、図１０（ｃ），（ｄ）に示すように、二つの点像からデフォーカスし、かつｘ軸方向およびｙ軸方向にずれた二つのボケ像が形成される。 In a normal optical low-pass filter, only the shift amount in the y-axis direction of the focus positions of ordinary rays and extraordinary rays is considered. On the other hand, in this embodiment, as shown in FIGS. 10A and 10B, the focus positions P1 and P2 are shifted from each other by the focus direction deviation amount L = 3.535 μm in the focus direction. This is due to a difference in refractive index between the first birefringent element 2 with respect to the ordinary ray and the extraordinary ray. At this time, the size of the synthesized blurred image formed by the ordinary light beam and the extraordinary light beam is minimized at a position intermediate between the focus positions P1 and P2. The surface in the middle position is set as the evaluation surface. On the evaluation surface, as shown in FIGS. 10C and 10D, two blurred images defocused from the two point images and shifted in the x-axis direction and the y-axis direction are formed.
　次に、この二つのボケ像を形成した光束は、第３の複屈折素子３によって偏光解消作用を受けた後、第２の複屈折素子４による点像分離作用を受ける。 Next, the light beam forming these two blurred images is subjected to the depolarization action by the third birefringence element 3 and then subjected to the point image separation action by the second birefringence element 4.
　第２の複屈折素子４の光学軸４ａの方位角θは０°であるため、該第２の複屈折素子４を通過した光束は、該方位角θの方向であるｘ軸方向に二つに分離する。つまり、第１の複屈折素子２からの二つの光束がそれぞれ、常光線と異常光線に分離する。その際、常光線は第２の複屈折素子４内を真っ直ぐに通過し、異常光線はｘ軸方向にシフトする。この結果、第２の複屈折素子４を通過した異常光線のピント位置Ｐ３，Ｐ４はそれぞれ、ピント位置Ｐ１，Ｐ２に対してｘ軸方向にずれる。本実施例では、ピント位置Ｐ１，Ｐ２間のｙ軸方向でのずれ量と同様に、ピント位置Ｐ３，Ｐ４のｘ軸方向でのずれ量も１．０μｍに設定している。また、ピント位置Ｐ３，Ｐ４はフォーカス方向にもフォーカス方向ずれ量Ｌ＝３．５３５μｍだけ互いにずれる。 Since the azimuth angle θ of the optical axis 4a of the second birefringent element 4 is 0 °, two light beams that have passed through the second birefringent element 4 are two in the x-axis direction that is the direction of the azimuth angle θ. To separate. That is, the two light beams from the first birefringent element 2 are separated into an ordinary ray and an extraordinary ray, respectively. At that time, the ordinary ray passes straight through the second birefringent element 4, and the extraordinary ray shifts in the x-axis direction. As a result, the focus positions P3 and P4 of extraordinary rays that have passed through the second birefringent element 4 are shifted in the x-axis direction with respect to the focus positions P1 and P2, respectively. In the present embodiment, similarly to the shift amount in the y-axis direction between the focus positions P1 and P2, the shift amount in the x-axis direction of the focus positions P3 and P4 is set to 1.0 μm. The focus positions P3 and P4 are also shifted from each other in the focus direction by a focus direction deviation L = 3.535 μm.
　こうして互いに分離した４つの点像は、横方向においては、図１１（ｃ），（ｄ）に示すように正方形の４つの頂点に位置する。また、図１１（ａ），（ｂ）に示すようにフォーカス方向においては３箇所に分離し、中央の二つの点像のフォーカス方向での位置は互いに同じとなる。この中央にてボケ像のサイズが最小になるため、ここを評価面とする。 The four point images thus separated from each other are located at the four vertices of the square in the lateral direction as shown in FIGS. 11 (c) and 11 (d). Further, as shown in FIGS. 11 (a) and 11 (b), it is separated into three places in the focus direction, and the positions of the two central point images in the focus direction are the same. Since the size of the blurred image is minimized at the center, this is the evaluation surface.
　評価面では、二つの大きなボケ像と二つの結像した点像とが合成されたボケ像が生じる。この合成されたボケ像のサイズは、横方向での点像分離によるボケ径（横ボケ径）とフォーカス方向での点像分離によるボケ径（フォーカス方向ボケ径）とが合成されたものとなる。フォーカス方向ボケ径δｄｅｆは、本実施例でも、実施例１と同様に式（３）で近似することができる。 On the evaluation surface, a blurred image is generated by combining two large blurred images and two imaged point images. The size of the synthesized blur image is a combination of a blur diameter due to point image separation in the horizontal direction (horizontal blur diameter) and a blur diameter due to point image separation in the focus direction (focus direction blur diameter). . The focus direction blur diameter δdef can also be approximated by Expression (3) in the present embodiment as in the first embodiment.
　図１２は、本実施例におけるＦ値ごとの回折による最小錯乱円径δＦｎｏ（短破線）としてのボケ径と、光学ローパスフィルタ６によるフォーカス方向ボケ径δｄｅｆ（長破線）および横ボケ径δｘ（一点鎖線）と、これらが合成されることで得られる合成ボケ径δａｌｌ（実線）とを示す。合成ボケ径は、ｘ軸方向およびｙ軸方向のいずれから見ても同じである。図１２は、イメージセンサ７の画素ピッチＰが３．０μｍである場合を示している。このイメージセンサ７に対して偽色やモアレの除去に必要な目標ボケ径は約２．１４μｍである。この目標ボケ径を図１２中に点線で示している。 FIG. 12 shows a blur diameter as a minimum circle of confusion δFno (short broken line) due to diffraction for each F value in this embodiment, a focus direction blur diameter δdef (long broken line) and a horizontal blur diameter δx (one point) by the optical low-pass filter 6. A chain line) and a synthetic blur diameter δall (solid line) obtained by synthesizing them. The synthetic blur diameter is the same when viewed from both the x-axis direction and the y-axis direction. FIG. 12 shows a case where the pixel pitch P of the image sensor 7 is 3.0 μm. The target blur diameter necessary for removing false colors and moire from the image sensor 7 is about 2.14 μm. This target blur diameter is indicated by a dotted line in FIG.
　光学ローパスフィルタを用いないと、Ｆ＝２．８以下において最小錯乱円径δＦｎｏが目標ボケ径より小さくなるために偽色やモアレが発生する。 If an optical low-pass filter is not used, since the minimum circle of confusion δFno is smaller than the target blur diameter at F = 2.8 or less, false color or moire occurs.
　フォーカス方向ボケ径δｄｅｆは、実施例１と同様に、開放Ｆ値側では大きく、小絞りＦ値側では小さくなる。横ボケ径δｘは、Ｆ値にかかわらず一定の値（１．０μｍ）となっている。合成ボケ径δａｌｌは、Ｆ値が１．４から２．８以下では目標ボケ径に極めて近い。そして、Ｆ値が４．０以上であると最小錯乱円径δＦｎｏに漸近し、必要最小限のボケ径となっている。このため、小絞り回折による像劣化を最小限に抑えることができる。 The focus direction blur diameter δdef is large on the open F value side and small on the small aperture F value side, as in the first embodiment. The horizontal blur diameter δx is a constant value (1.0 μm) regardless of the F value. The synthetic blur diameter δall is extremely close to the target blur diameter when the F value is 1.4 to 2.8 or less. When the F value is 4.0 or more, the minimum confusion circle diameter δFno is asymptotic and the minimum necessary blur diameter is obtained. For this reason, image deterioration due to small aperture diffraction can be minimized.
　図１３は、図１２に対する比較例として、横方向での点像分離のみを行う従来の光学ローパスフィルタを用いた場合を示す。図１３は、Ｆ値ごとの回折による最小錯乱円径δＦｎｏ（短破線）としてのボケ径と、横ボケ径δｘ（長破線）と、これらの合成により生ずる合成ボケ径δａｌｌ（実線）とを示す。画素ピッチは３．０μｍである。 FIG. 13 shows a case where a conventional optical low-pass filter that performs only point image separation in the horizontal direction is used as a comparative example with respect to FIG. FIG. 13 shows a blur diameter as a minimum circle of confusion δFno (short broken line) by diffraction for each F value, a horizontal blur diameter δx (long broken line), and a synthesized blur diameter δall (solid line) generated by combining these. . The pixel pitch is 3.0 μm.
　従来の光学ローパスフィルタは、Ｆ値にかかわらず２．１４μｍのボケ径を生じさせ、Ｆ値２．０から８．０の範囲での像劣化が著しい。 The conventional optical low-pass filter generates a blur diameter of 2.14 μm regardless of the F value, and the image deterioration in the F value range of 2.0 to 8.0 is remarkable.
　一方、図１４は、実施例１と同様にフォーカス方向にのみ点像分離を行う光学ローパスフィルタを用いた場合のＦ値ごとの回折による最小錯乱円径δＦｎｏ（短破線）としてのボケ径と、横ボケ径δｘ（長破線）と、合成ボケ径δａｌｌ（実線）とを示す。画素ピッチは３．０μｍである。Ｆ値２．０から８．０の範囲で像劣化は防げているが、開放Ｆ値１．４での像劣化が実施例２より大きい。 On the other hand, FIG. 14 shows a blur diameter as a minimum circle of confusion δFno (short broken line) by diffraction for each F value when an optical low-pass filter that performs point image separation only in the focus direction is used as in the first embodiment. A horizontal blur diameter δx (long broken line) and a synthetic blur diameter δall (solid line) are shown. The pixel pitch is 3.0 μm. Although image deterioration can be prevented in the range of F value 2.0 to 8.0, image deterioration at open F value 1.4 is larger than that in Example 2.
　開放Ｆ値における像劣化を極限まで抑えるためには、実施例２のように、フォーカス方向の点像分離と横方向の点像分離と両方を用いることが好ましい。この際、光学軸の角φは上述した式（７）の範囲に設定する必要がある。しかも、角φは、フォーカス方向ボケ径が横ボケ径より大きくなるようなフォーカス方向と横方向の点像分離量が生じるように設定する必要がある。 In order to suppress image degradation at the open F value to the limit, it is preferable to use both point image separation in the focus direction and point image separation in the horizontal direction as in the second embodiment. At this time, the angle φ of the optical axis needs to be set in the range of the above-described formula (7). Moreover, the angle φ needs to be set so that a point image separation amount in the focus direction and the horizontal direction is generated such that the focus direction blur diameter is larger than the horizontal blur diameter.
　これらの条件を満足するように、フォーカス方向ボケ径および横ボケ径の割合を調整することで、偽色やモアレを低減しつつ、小絞り回折による像劣化も低減することが可能な光学ローパスフィルタを得ることができる。 An optical low-pass filter that can reduce image degradation due to small-aperture diffraction while reducing false color and moire by adjusting the ratios of the focus direction blur diameter and lateral blur diameter to satisfy these conditions. Can be obtained.
　上記各実施例は様々な画素ピッチを有するイメージセンサに適用可能であるが、小絞り回折によるボケ径は小さいため、比較的大きい画素ピッチでは各実施例が解決せんとする問題は生じない。この問題が生じるのは、主として画素ピッチが５μｍ以下のイメージセンサを用いる場合であり、特に画素ピッチが３．０μｍ以下の場合に顕著となる。ただし、画素ピッチが５μｍ以上の場合でも各実施例の光学ローパスフィルタを用いてもよい。 Each of the above embodiments can be applied to an image sensor having various pixel pitches. However, since the blur diameter due to small aperture diffraction is small, there is no problem that each embodiment does not solve at a relatively large pixel pitch. This problem occurs mainly when an image sensor with a pixel pitch of 5 μm or less is used, and particularly when the pixel pitch is 3.0 μm or less. However, even when the pixel pitch is 5 μm or more, the optical low-pass filter of each embodiment may be used.
　また、開放Ｆ値が４．０から５．６程度以上と大きい場合は、小画素ピッチのイメージセンサに対しては十分な結像性能を有しておらず、偽色やモアレの問題が生じない。この場合、光学ローパスフィルタ自体が不要となる。このため、各実施例は、開放Ｆ値が１．０から２．８程度の高解像度の撮像レンズを用いる場合に主として有効である。さらに、開放Ｆ値が１．０から２．０の高解像度の撮像レンズを用いる場合は、撮像により得られる画質に対する要求が高く、撮像レンズの収差補正も高いレベルで行われているため、偽色やモアレを低減するための光学ローパスフィルタの必要性が高い。また、小絞り回折による画質劣化を防ぐことに対する要求も高くなるため、各実施例が有効である。 In addition, when the open F-number is as large as about 4.0 to 5.6 or more, it does not have sufficient imaging performance for an image sensor with a small pixel pitch, and a problem of false color and moire occurs. Absent. In this case, the optical low-pass filter itself becomes unnecessary. Therefore, each embodiment is mainly effective when using a high-resolution imaging lens having an open F value of about 1.0 to 2.8. Furthermore, when using a high-resolution imaging lens having an open F value of 1.0 to 2.0, there is a high demand for image quality obtained by imaging and aberration correction of the imaging lens is performed at a high level. There is a high need for an optical low-pass filter for reducing color and moire. In addition, each embodiment is effective because there is a high demand for preventing image quality deterioration due to small aperture diffraction.
　表１は、実施例１におけるＦ＝１．８および実施例２におけるＦ＝１．４でのδｄｅｆ［μｍ］、δｘ［μｍ］、式（１）の値、Ｄσ［μｍ］および式（４）の値を示す。画素ピッチＰ［μｍ］も併せて示す。実施例１，２のいずれにおいても、式（１）および式（４）の条件を満足している。 Table 1 shows δdef [μm], δx [μm], the value of equation (1), Dσ [μm], and equation (4) at F = 1.8 in Example 1 and F = 1.4 in Example 2. ) Value. The pixel pitch P [μm] is also shown. In both Examples 1 and 2, the conditions of Formula (1) and Formula (4) are satisfied.
Figure JPOXMLDOC01-appb-T000001
Figure JPOXMLDOC01-appb-T000001
　以上説明した各実施例は代表的な例にすぎず、本発明の実施に際しては、各実施例に対して種々の変形や変更が可能である。
  Each embodiment described above is only a representative example, and various modifications and changes can be made to each embodiment in carrying out the present invention.

Claims (10)

  1. 　撮像素子の撮像面に光束を導く光学ローパスフィルタであって、
    　前記撮像面に垂直な第１の方向において光束の集光点を複数に分離することでボケ像を生じさせ、
    　前記撮像素子の画素ピッチをＰ［μｍ］、前記第１の方向での前記集光点の分離により生ずる前記ボケ像のサイズをδｄｅｆ［μｍ］、前記第１の方向に垂直な第２の方向において光束の集光点を複数に分離する場合の該分離により生ずるボケ像のサイズをδｘ［μｍ］（ただし、前記集光点が前記第２の方向において分離されない場合はδｘ＝０）とするとき、
    ０．２≦（δｄｅｆ２＋δｘ２）／Ｐ２≦０．９
    δｘ≦δｄｅｆ
    なる条件を満足することを特徴とする光学ローパスフィルタ。     An optical low-pass filter that guides a light beam to an imaging surface of an imaging element,
    A blur image is generated by separating a light collecting point in a first direction perpendicular to the imaging surface into a plurality of points,
    The pixel pitch of the image sensor is P [μm], the size of the blurred image generated by the separation of the condensing points in the first direction is δdef [μm], and the second direction perpendicular to the first direction When the condensing point of the light beam is separated into a plurality, the size of the blurred image generated by the separation is δx [μm] (where δx = 0 when the condensing point is not separated in the second direction). When
    0.2 ≦ (δdef 2 + δx 2 ) / P 2 ≦ 0.9
    δx ≦ δdef
    An optical low-pass filter characterized by satisfying the following condition.
  2. 　前記光束を集光させる光学系の開放Ｆ値をＦ、前記第１の方向での前記集光点の分離量の標準偏差［μｍ］をＤσとするとき、
    δｄｅｆ＝Ｄσ／Ｆ
    であることを特徴とする請求項１に記載の光学ローパスフィルタ。     When the open F value of the optical system for condensing the luminous flux is F, and the standard deviation [μm] of the separation amount of the condensing point in the first direction is Dσ,
    δdef = Dσ / F
    The optical low-pass filter according to claim 1, wherein
  3. ０．３≦Ｄσ／（Ｆ・Ｐ）≦１．０
    Ｐ≦５．０［μｍ］
    なる条件を満足することを特徴とする請求項２に記載の光学ローパスフィルタ。     0.3 ≦ Dσ / (F · P) ≦ 1.0
    P ≦ 5.0 [μm]
    The optical low-pass filter according to claim 2, wherein the following condition is satisfied.
  4. １．０≦Ｆ≦２．０
    なる条件をさらに満足することを特徴とする請求項２または３に記載の光学ローパスフィルタ。     1.0 ≦ F ≦ 2.0
    The optical low-pass filter according to claim 2 or 3, further satisfying the following condition.
  5. 　前記光学ローパスフィルタが、複屈折性材料からなる少なくとも１つの複屈折素子により構成されていることを特徴とする請求項１から４のいずれか一項に記載の光学ローパスフィルタ。 The optical low-pass filter according to any one of claims 1 to 4, wherein the optical low-pass filter includes at least one birefringent element made of a birefringent material.
  6. 　前記複屈折素子の光学軸が前記第１の方向に対してなす角が９０°であることを特徴とする請求項５に記載の光学ローパスフィルタ。 The optical low-pass filter according to claim 5, wherein an angle formed by the optical axis of the birefringent element with respect to the first direction is 90 °.
  7. 　前記複屈折素子の光学軸が前記第１の方向に対してなす角をφとするとき、
    ８０°≦φ＜９０°
    なる条件を満足することを特徴とする請求項５に記載の光学ローパスフィルタ。     When the angle formed by the optical axis of the birefringent element with respect to the first direction is φ,
    80 ° ≦ φ <90 °
    The optical low-pass filter according to claim 5, wherein the following condition is satisfied.
  8. 　前記複屈折性材料は、一軸性結晶の材料であることを特徴とする特徴とする請求項４から６のいずれか一項に記載の光学ローパスフィルタ。 The optical low-pass filter according to any one of claims 4 to 6, wherein the birefringent material is a uniaxial crystal material.
  9. 　前記一軸性結晶は、ニオブ酸リチウムの結晶であることを特徴とする請求項８に記載の光学ローパスフィルタ。 The optical low-pass filter according to claim 8, wherein the uniaxial crystal is a crystal of lithium niobate.
  10. 　請求項１から９のいずれか一項に記載の光学ローパスフィルタと、前記撮像素子とを有することを特徴とする撮像装置。 An image pickup apparatus comprising the optical low-pass filter according to any one of claims 1 to 9 and the image pickup element.
PCT/JP2019/017211 2018-06-06 2019-04-23 Optical low-pass filter and image capture device WO2019235091A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2018-108685 2018-06-06
JP2018108685A JP2019211668A (en) 2018-06-06 2018-06-06 Optical low pass filter and imaging device

Publications (1)

Publication Number Publication Date
WO2019235091A1 true WO2019235091A1 (en) 2019-12-12

Family

ID=68770717

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2019/017211 WO2019235091A1 (en) 2018-06-06 2019-04-23 Optical low-pass filter and image capture device

Country Status (2)

Country Link
JP (1) JP2019211668A (en)
WO (1) WO2019235091A1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09211393A (en) * 1996-02-06 1997-08-15 Olympus Optical Co Ltd Optical low-pass filter
JP2007097652A (en) * 2005-09-30 2007-04-19 Olympus Medical Systems Corp Endoscope

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09211393A (en) * 1996-02-06 1997-08-15 Olympus Optical Co Ltd Optical low-pass filter
JP2007097652A (en) * 2005-09-30 2007-04-19 Olympus Medical Systems Corp Endoscope

Also Published As

Publication number Publication date
JP2019211668A (en) 2019-12-12

Similar Documents

Publication Publication Date Title
US7110034B2 (en) Image pickup apparatus containing light adjustment portion with reflection of a portion of light onto adjacent pixels
JP4558804B2 (en) Imaging device
JP5829365B2 (en) Endoscopic imaging device
JP2744323B2 (en) Optical low-pass filter
JP5496794B2 (en) Solid-state imaging device
US11310448B2 (en) Imaging apparatus and optical low-pass filter
WO2019235091A1 (en) Optical low-pass filter and image capture device
JP6789696B2 (en) Optical low-pass filter and imaging device with it, imaging unit
JP2009217030A (en) Optical system, and imaging device and digital apparatus using the optical system
CN109387950B (en) Optical low-pass filter and imaging device
US20060262208A1 (en) Optical low pass filter and image pickup apparatus having the same
JP7207883B2 (en) Optical low-pass filter and imager
JPH0718985B2 (en) Solid-state imaging device
JP4926471B2 (en) Optical element and imaging apparatus having the same
JP2011232650A (en) Optical low pass filter and digital camera
WO2014097507A1 (en) Solid-state image pickup element
JP4269386B2 (en) Imaging optical system
JP3989862B2 (en) Optical low-pass filter
JP3792992B2 (en) Optical low-pass filter and optical apparatus using the same
JP2020187153A (en) Optical low-pass filter and imaging device
JP2840619B2 (en) Optical low-pass filter
JP2020187154A (en) Optical low-pass filter and imaging device
US20040080826A1 (en) Optical low pass filter
JP2020187155A (en) Optical low pass filter, and imaging apparatus having the same
CN117693814A (en) Solid-state image pickup device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19815348

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19815348

Country of ref document: EP

Kind code of ref document: A1