JP6422629B2 - Fundus photographing device - Google Patents

Description

It relates to a fundus imaging apparatus of a scanning type for photographing a subject's eye.

  Some scanning fundus photographing apparatuses acquire detailed photographing information of an eye to be examined by designating a local imaging position on a wide-angle photographed image. This type of fundus imaging apparatus has an optical member such as a concave mirror for causing illumination light to enter the eye, and the illumination light scanned by the scanning member is reflected by the optical member and enters the eye. Reflected light from the eye is received by a light receiving element, and is used as a pixel for forming a captured image and various calculations (see Patent Document 1). In order to acquire detailed photographing information with a wide angle of view, it is necessary to arrange a plurality of optical members or increase the size of the optical members so that illumination light is irradiated to the peripheral portion of the eye (Patent Document 2). reference).

JP 2010-259669 A

Ferguson, RD, Zhong, Z, Hammer, DX, Mujat, M, Patel, AH, Deng, C, Zou, W, Burns, SA, "Adaptive optics SLO with integrated wide-field retinal imaging and tracking" "J.Opt Soc. Am. A 27, A265-A277 (2010) "

By the way, the fundus image acquired in the local region may include distortion due to reflection or refraction of at least one optical member (for example, concave mirror or lens) arranged in the optical system. This distortion varies depending on the local imaging position on the fundus. That is, since the incident angle of light to the optical member differs depending on the imaging position on the fundus, the distortion characteristics of the fundus image at each imaging position are different. It may be difficult to obtain a good fundus image suitable for diagnosis or analysis without considering such distortion characteristics of the fundus image.
Depending on the constructed optical system, in particular, when obtaining a fundus image in a local region using a peripheral portion away from the optical axis in at least one optical member (for example, a concave mirror or a lens) arranged in the optical system, In some cases, the entire fundus image may be distorted.

In view of the problems of the prior art, and an object to provide a fundus photographing equipment for scanning the imaging information of the eye can be accurately acquired.

  In order to solve the above problems, the present invention is characterized by having the following configuration.

(1) An optical scanner for scanning a local area on the fundus of the eye to be inspected with light emitted from a light source, and a detector for detecting light from the fundus at each scanning position in the local area. A fundus imaging optical system for imaging the local region;
A light deflector arranged in an optical path of the fundus imaging optical system, for deflecting light from the light source, and the position of the local region imaged by the fundus imaging optical system is set to be larger than the angle of view of the local region. By driving and controlling the light deflecting means over a wide range, an imaging position changing means that changes up and down, left and right, and a fundus image having the local area set by the imaging position changing means as an imaging range are obtained from the detector. Image processing means for processing and obtaining the received light signal, wherein the image processing means obtains positional information of the local region set by the imaging position changing means, and is a distortion of the fundus image. A fundus imaging apparatus that corrects distortions that differ for each position of the local region based on position information of the local region set by the imaging position changing unit.
(2) The image processing means is an image processing means for correcting the distortion of the fundus image by displacing the coordinates of each pixel in the fundus image when correcting the distortion of the fundus image. The coordinate displacement amount of the fundus image when it is assumed that there is no distortion caused by at least one optical member arranged in the fundus imaging optical system, and the coordinates of the fundus image actually acquired The fundus imaging apparatus according to (1), wherein each of the scanning positions of the optical scanner in the local region set by the imaging position changing unit is set based on a displacement amount with respect to the position.
(3) The image processing unit corrects the distortion of the fundus image according to the distance from the optical axis of the fundus imaging optical system to the local region set by the imaging position changing unit (1) or (2) Fundus photographing apparatus.

  According to the present invention, it is possible to acquire eye photographing information with high accuracy.

  An embodiment of the present invention will be described. FIG. 1 is an external view of a fundus imaging apparatus 500. FIG. 2 is an explanatory diagram of the optical system of the fundus imaging apparatus 500. FIG. 3 is a control block diagram of the fundus imaging apparatus 500. FIG. 4 is an example of the display screen of the monitor 70.

  The fundus imaging apparatus 500 includes a base 502, an imaging unit 503, and a face support unit 504. An optical system (to be described later) is housed inside the photographing unit 503 attached on the base 502. The face support unit 504 includes a chin rest 505, and the chin rest 505 moves in a three-dimensional direction with respect to the base of the face support unit 4 by driving of a driving means (not shown).

  The optical system has a first photographing unit 100 and a second photographing unit 200. The first photographing unit 100 is configured by a scanning laser ophthalmoscope using a confocal optical system, and photographs the eye E to obtain a photographed image (first fundus image 70a) with a cell level resolution. The second imaging unit 200 obtains a captured image (second fundus image 70b) having a wider angle of view than the first imaging unit 100 in order to specify the imaging position of the first fundus image 70a.

  The first photographing unit 100 includes a first illumination optical system 100a, a first photographing optical system 100b, and a wavefront compensation unit 110. The first illumination optical system 100a illuminates the eye (fundus) E two-dimensionally by scanning the illumination light beam. The first imaging optical system 100b receives reflected light (reflected light flux) from the fundus and obtains a first fundus image 70a. The wavefront compensation unit 110 detects the wavefront aberration of the eye E and removes the low-order aberration and the high-order aberration.

(First shooting unit)
The first illumination optical system 100a includes a light source 1 (first light source), a lens 2, a polarization beam splitter 4, a beam splitter 71, a mirror 6, a concave mirror 7, a plane mirror 8, and a wavefront compensation on an optical path (optical axis) L1. Device 72, beam splitter 75, concave mirror 11, concave mirror 12, resonant scanner 15, concave mirror 16, concave mirror 17, flat mirror 21, lens 22, flat mirror 23, diopter correction unit 10, flat mirror 25, concave mirror 26, galvano scanner 40 A dichroic mirror 90, a plane mirror 32, a plane mirror 33, and concave mirrors 31 and 35 are arranged.

  The light source 1 is a well-known infrared light source that illuminates the fundus from the near infrared region to the infrared region. For example, an SLD (Super Luminescent Diode) light source having a wavelength of 840 nm, a semiconductor laser that emits highly converged spot light, or the like. Used. The polarization beam splitter 4 passes the light beam of the S polarization component of the irradiation light from the light source 1 and shields the other light beam (such as the P deflection component). The beam splitter 71 has a characteristic of transmitting light having the wavelength of the light source 1 and reflecting light having the wavelength of the aberration detection light source 76 described later.

The wavefront compensation device 72 removes aberrations contained in fundus reflected light detected by a wavefront sensor 73 (described later). For example, the wavefront compensation device 72 uses a reflective LCOS (Liquid Crystal On Silicon). In addition to this, in addition to a reflection type wavefront compensation device of MEMS (Micro Electro Mechanical Systems), a transmission type wavefront compensation device that transmits reflected light from the fundus and compensates for wavefront aberration can be used.

The wavefront compensation device 72 modulates the s-polarized light component of the incident light to illuminate the illumination light (
It is possible to compensate for aberrations with respect to predetermined linearly polarized light (S-polarized light) such as S-polarized light), reflected light from the fundus of illumination light (S-polarized light), and reflected light (S-polarized light component) of wavefront aberration detection light. Placed in the right direction.

  The diopter correction unit 10 is used to change the optical path length for diopter correction of the eye E. The diopter correction unit 10 includes, for example, two plane mirrors, two lenses (not shown in the figure), and a drive unit 10a. The plane mirror and the lens are moved in the direction of arrow A by the drive of the drive unit 10a, and the diopter is corrected by changing the optical path length. The diopter correction unit 10 may be a prism that can move in the optical axis direction.

  The resonant scanner 15 has a mirror 15a and a drive unit 15b. By driving the drive unit 15b, the mirror 15a vibrates in the main scanning direction (X direction) at a required deflection angle (view angle). The galvano scanner 40 includes a galvano mirror 41a, a galvano mirror 41b, and a drive unit. The galvanometer mirror 41a is inclined by the drive unit 42 in the main scanning direction (X direction). The galvanometer mirror 41b is inclined in the sub-scanning direction (Y direction) by the drive unit. The fundus is illuminated two-dimensionally by a combination of scanning by the resonant scanner 15 and the galvanometer mirror 41a.

  Further, the illumination light irradiation position of the light source 1 is moved to the imaging position of the first fundus image 70a specified on the second fundus image 70b by the inclination (deflection) of the galvanometer mirrors 41a and 41b. The reflected light from the fundus is received and photoelectrically converted by the light receiving element 56 which is a detector. A signal output from the light receiving element 56 is quantized by a control unit 80 (described later) and used for formation of a two-dimensional fundus image and various calculations.

  In addition, the light for scanning the local region in the Y direction and obtaining the first fundus image 70a as in the galvano mirror 41b described above is the light for changing the imaging position in the Y direction of the first fundus image 70a. A deflection member may also be used. Of course, the optical scanner for scanning the local region in the X direction to obtain the first fundus image 70a may also serve as the light deflection member for changing the imaging position of the first fundus image 70a in the X direction. Of course, the optical scanner for obtaining the first fundus image 70a and the optical deflection member for changing the imaging position may have different configurations with respect to the X direction and the Y direction, respectively.

  The dichroic mirror 90 has a characteristic of transmitting a light beam from the second photographing unit 200 and reflecting a light beam from the light source 1 and a light source 76 described later, and an optical path between the second photographing unit 200 and the first illumination optical system 100a. Is approximately coaxial.

  With the configuration as described above, the light beam emitted from the light source 1 is converted into parallel light by the lens 2 and then reflected by the plane mirror 8 from the beam splitter 71 and the concave mirror 6 through the polarization beam splitter 4 to be a wavefront compensation device. 72 is incident. The light beam reflected by the wavefront compensation device 72 is reflected by the concave mirrors 11 and 12 via the beam splitter 75 and enters the resonant scanner 15.

  The light beam reflected by the resonant scanner 15 is reflected from the concave mirror 16 by the plane mirror 21, condensed on the lens 22, reflected by the plane mirror 23, and further via the diopter correction unit 10, The light is reflected by the concave mirror 26 and enters the galvano scanner 40. The light beam reflected by the galvano scanner 40 is reflected by the concave mirror 35 from the dichroic mirror 90 and condensed on the fundus.

The first imaging optical system 100b shares an optical path from the dichroic mirror 90 to the beam splitter 71 of the first illumination optical system 100a, and includes a plane mirror 51, a polarizing beam splitter 52, a lens 53, a pin on the reflected optical path of the beam splitter 71. It has a hall plate 54, a lens 55, and a light receiving element 56. The polarization beam splitter 52 passes only the S-polarized light beam and blocks the P-polarized component (other) light beam. The pinhole plate 54 is placed at a conjugate position with the fundus. As the light receiving element 56, an APD (avalanche photodiode), a photomultiplier tube, or the like is used.

  Reflected light from the fundus traverses the first illumination optical system 100 a in reverse, and is reflected by the beam splitter 71 and the plane mirror 51. Then, only the light beam of the S polarization component is transmitted by the polarization beam splitter 52. The light beam of the S deflection component transmitted through the polarization beam splitter 52 is focused on the pinhole of the pinhole plate 54 through the lens 53, and is received by the light receiving element 56 through the lens 55.

  Since most of the reflected light from the cornea is removed by the pinhole plate 54, the light receiving element 56 preferably receives the reflected light from the fundus. The wavefront compensation unit 110 shares an optical member from the beam splitter 71 to the concave mirror 35 placed in the optical path of the first illumination optical system 100a, and also includes a light source 76, a lens 77, a polarization beam splitter 78, beam splitters 75 and 71, and a dichroic. It has a mirror 86, a polarization beam splitter 85, a lens 84, a plane mirror 83, a lens 82, and a wavefront sensor 73. As the light source 76, a laser diode or the like that emits a light beam in an infrared region different from that of the light source 1 is used. The light source 1 and the aberration detection light source 76 may be used in combination.

  The polarization beam splitter (first polarization means) 78 polarizes the light beam emitted from the light source 76 into a P-polarized light beam orthogonal to the light beam from the light source 1 that has been S-polarized by the polarization beam splitter 4. The beam splitter 75 guides the light beam of the wavefront compensation unit 110 to the optical path of the first illumination optical system. The beam splitter 71 transmits the light having the wavelength of the light source 1 (840 nm) and reflects the light having the wavelength of the aberration detecting light source 76 (780 nm). As a result, the wavefront sensor 73 detects light having an S-polarized component from the scattered light from the fundus occupying the irradiated laser light. The dichroic mirror 86 transmits light (840 nm) having the wavelength of the light source 1 and reflects light having the wavelength (780 nm) of the light source 76 for aberration detection. The polarization beam splitter (second polarization unit) 85 blocks the light beam in the polarization direction (P-polarized light) irradiated to the eye E from the light source 76, and the light beam in the polarization direction (S-polarized light) orthogonal to the polarization direction. To Penetrate.

  As the wavefront sensor 73, a sensor that detects low-order aberrations and high-order aberrations contained in the reflected light of the eye to be examined is used. For example, it includes a microlens array and a two-dimensional imaging element 73a (two-dimensional light receiving element) that receives a light beam that has passed through the microlens array. The light source 76 (third light source) for aberration detection is selected to irradiate a light beam in an infrared band different from that of the light source 1. For example, a laser diode that emits laser light having a wavelength of 780 nm is used. The emission end of the light source 76 is substantially conjugate with the fundus. As the wavefront sensor 73, a Hartmann Shack detector, a wavefront curvature sensor for detecting a change in light intensity, or the like is used. The reflective surfaces of the resonant scanner 15 and the wavefront compensation device 72 are conjugate with the pupil of the eye to be examined. The light receiving surface (two-dimensional imaging element 76a) of the wavefront sensor 73 is substantially conjugate with the fundus of the eye E. In the present embodiment, the wavefront compensation device 72 is controlled based on the detection result from the wavefront sensor 73, but the present invention is not limited to this. For example, the control unit 80 may control the wavefront compensation device 72 using a fundus image acquired based on a light reception signal from the light receiving element 56. With such a configuration, the wavefront of reflected light from the fundus can be compensated using the wavefront compensation device 72 without providing a wavefront sensor. When the wavefront sensor is not used, for example, the control unit 80 may determine the focus state of the fundus image by image processing, and may control the wavefront compensation device 72 so that a fundus image with an appropriate focus state is obtained.

  The laser light emitted from the light source 76 is converted into parallel light by the lens 77, and then the polarization direction is orthogonal to the light beam from the light source 1 by the polarizing beam splitter 78 (P-polarized light). Guided to the optical path of system 100a. The laser light reflected by the beam splitter 75 is condensed on the fundus through the optical path of the first illumination optical system 100a. The reflected light from the fundus is reflected by the wavefront compensation device 72 through each optical member of the first illumination optical system 100 a, removed from the optical path of the first illumination optical system 100 a by the beam splitter 71, and reflected by the dichroic mirror 86. Then, it is guided to the wavefront sensor 73 through the polarization beam splitter 85, the lens 84, the plane mirror 83, and the lens 82. The wavefront sensor 73 detects light having an S-polarized component from the scattered light from the fundus and suppresses detection of the light beam reflected by the cornea or the optical element by the wavefront sensor 73.

  The wavefront compensation device (compensation optical system) 110 controls the wavefront compensation device 72 based on the wavefront aberration of the fundus reflected light of the light source 76 detected by the wavefront sensor 73, and together with the S-polarized component of the reflected light of the light source 76, The wavefront aberration of the luminous flux emitted from the light source 1 and the reflected light thereof is removed, and the high-resolution first fundus image 70a from which the wavefront aberration of the eye E is removed (wavefront compensation is performed) is obtained.

(Second shooting unit)
The second imaging unit obtains a fundus image (second fundus image 70b) having a wider angle of view than the angle of view of the first imaging unit. The second fundus image 70b is used for specifying the imaging position of the first fundus image 70a and confirming the imaging position. The second photographing unit 200 that acquires the second fundus image 70b only needs to be able to acquire the fundus image of the eye E in real time with a wide angle of view (for example, about 20 to 60 degrees) for observation. An imaging optical system, an optical system of a scanning laser ophthalmoscope (SLO), a control system, and the like are used. Here, for convenience of explanation, the configuration of the second photographing unit 200 is shown in a block diagram.

  The second photographing unit 200 has a second illumination optical system and a second photographing optical system. The second illumination optical system includes a second light source 210 that illuminates the fundus with infrared light, a scanning unit 220 that scans the light beam two-dimensionally on the fundus, and the like, and illuminates the fundus two-dimensionally. As the second light source 210, for example, a laser diode that emits laser light having a wavelength of 910 nm is used. The scanning unit 220 includes a mirror that deflects (reflects) laser light in a two-dimensional direction of X and Y directions. The deflection angle of the scanning unit 220 is determined so that the shooting angle of view of the second shooting unit 200 is larger than the shooting angle of view of the first shooting unit 100. For example, when it is possible to simultaneously photograph fundus features such as the macula and nipple, the angle of view is about 20 to 60 degrees. The second photographing unit 200 includes a light receiving element 251 that receives the reflected light from the fundus in order to receive the reflected light from the fundus and acquire the second fundus image 70b.

  The optical path of the second imaging unit 200 is made substantially coaxial with the first imaging unit 100 by the dichroic mirror 90, and the light flux from the second light source 210 is collected on the fundus via the dichroic mirror 90, the concave mirrors 31 and 35, and the plane mirror 33. Shine. The light beam projected on the fundus scans a wide range of the fundus two-dimensionally by driving the scanning unit 220. The reflected light from the fundus passes through the optical path from the plane mirror 33 to the dichroic mirror 90 in the reverse direction and is received by the light receiving element 251 to obtain the second fundus image 70b used for specifying the shooting position of the first shooting unit 100 and the like. .

  The concave mirrors 31 and 35 arranged in the first photographing unit 100 and the second photographing unit 200 have a wide reflecting surface (area) so that illumination light can be incident on the periphery of the fundus (with a wide angle of view). It is preferable that what is possessed is used. By expanding the range (view angle) in which the fundus can be illuminated via the concave mirrors 31 and 35, the second fundus image 70b having a wider view angle can be acquired. It is also possible to capture the first fundus image 70a at the periphery of the fundus away from the optical axis L1.

On the other hand, the fundus image of the local region acquired by imaging by the first imaging unit 100 includes distortion due to reflection or refraction of at least one optical member (for example, a concave mirror or lens) arranged in the optical system. This distortion varies depending on the local imaging position on the fundus. That is, since the incident angle of light to the optical member differs depending on the imaging position on the fundus, the distortion characteristics of the fundus image at each imaging position are different. It may be difficult to obtain a good fundus image suitable for diagnosis or analysis without considering such distortion characteristics of the fundus image.
Depending on the constructed optical system, in particular, when obtaining a fundus image in a local region using a peripheral portion away from the optical axis in at least one optical member (for example, a concave mirror or a lens) arranged in the optical system, In some cases, the entire fundus image may be distorted.

  FIG. 10 is an explanatory diagram of the state of occurrence of distortion at each imaging position. FIG. 10A shows the imaging positions B1 to B9 of the first fundus image 70a with respect to the second fundus image 70b (fundus), and FIG. ) Are first fundus images 70a1 to 70a9 acquired at the respective imaging positions B1 to B9. Here, a difference in aberration that occurs in the first fundus image 70a when the first imaging unit 100 captures dots arranged at equal intervals is shown. That is, at the imaging position B1 including the optical axis L1, the illumination light is incident on the central portions of the concave mirrors 31 and 35, so that the distortion generated is small and the first fundus image 70a1 is displayed with relatively high accuracy. On the other hand, at the imaging positions B2 to B9 decentered from the optical axis 1, the illumination light is incident on the peripheral portions of the concave mirrors 31 and 35, so that distortion is more likely to occur than at the imaging position B1. That is, it can be seen that the aberrations of the first fundus images 70a2 to 70a9 appear more conspicuously than the first fundus image 70a1. As described above, the image distortion changes depending on the difference in the incident position of the illumination light on the concave mirrors 31 and 35. Therefore, a common distortion correction pattern (lookup table) is used for the aberration of the first captured image captured at each imaging position. It is difficult to correct using.

Note that if various calculations such as calculation of photoreceptor cell density or area are performed using an image (pixel) including distortion, the reliability of the calculation result may be reduced. Also, when confirming the arrangement of photoreceptor cells, the traveling of the optic nerve transition bundle, etc., correct recognition is hindered by the influence of aberration. Furthermore, when a plurality of first fundus images 70a photographed at different positions of the fundus are connected to form the first fundus image 70a having a wide angle of view, due to the aberrations included in each image individually, Adjacent images may not be connected with high accuracy.
Therefore, in the present embodiment, distortion caused by reflection of the concave mirrors 31 and 35 is corrected for each imaging position of the first fundus image 70a with respect to the fundus. A detailed description of the distortion correction will be described later.

  The control unit 80 controls the operation of the entire fundus photographing apparatus 500. For example, operation control such as alignment operation, imaging operation, wavefront compensation operation, and tracking of the apparatus is performed. The control unit 80 serves as an image acquisition unit that quantizes the signal output from the light receiving element 56 and corrects distortion included in the signal and stores it in the storage unit 81 as information on the captured image. The control unit 80 is also connected to a storage unit 81, a control unit 92, a monitor 70, and the like.

  The storage unit 81 stores information on the shooting angle of view of the first fundus image 70a. In addition, in order to correct distortion caused by reflection of the concave mirrors 31 and 35 for each pixel, at least one of a table for each scanning position and an arithmetic expression (distortion correction program) corresponding to the scanning position is stored. Information on the imaging position of the first fundus image 70a designated on the second captured image is also stored. For example, the imaging position B of the first fundus image 70a is stored as a displacement amount (offset amount) in the XY direction of the first fundus image 70a with respect to the reference position A of the second fundus image 70b (see FIG. 4). In addition to various programs, the storage unit 81 stores various information such as input information from the control unit 92 and acquired photographing information (first fundus image 70a and second fundus image 70b).

  The control unit 92 is used for various input operations. For example, it is used as input means for selecting and instructing the imaging position of the first fundus image 70a (imaging position of the local region) on the second fundus image 70b by the examiner's operation. For the control unit 92, a known input member such as a touch panel, a mouse, or the like installed on the monitor 70 is used.

  On the monitor 70, fundus images having different angles of view (that is, the first fundus image 70a and the second fundus image 70b) are formed based on the light reception signals of the light receiving elements 56 and 251 by the control unit 80. For example, the monitor 70 is updated at a predetermined frame rate (for example, about 10 to 100 Hz). The fundus image (the first fundus image 70a and the second fundus image 70b) is displayed as a moving image, and the fundus image stored in the storage unit 81 or directly acquired from the light receiving elements 56 and 251 is displayed as a still image. The Note that the fundus photographing apparatus 500 and the monitor 70 may be incorporated in the fundus photographing apparatus 500 in addition to separate housings.

<Description of operation>
Next, the operation of the fundus imaging apparatus 500 having the above configuration will be described.
A fixation lamp (not shown) is turned on to fix the eye E, and the diopter correction unit 10 is driven by the operation of the control unit 92 to correct the diopter of the eye E. As shown in FIG. 4A, when the second fundus image 70b photographed by the second photographing unit 200 is displayed on the monitor 70, the eye E and the photographing unit 503 are aligned.

  When the alignment is completed, the examiner uses the control unit 92 to specify the imaging position B of the first fundus image 70a on the second fundus image 70b. Based on the input signal, the control unit 80 forms a mark M by image processing at a position corresponding to the imaging position B of the first fundus image 70a on the second fundus image 70b, as shown in FIG. 4B. In addition, the control unit 80 obtains a deviation amount (offset) ΔX in the X direction of the imaging position B with respect to the reference position A of the second fundus image 70 b and a deviation amount (offset) ΔY in the Y direction, and stores them in the storage unit 81. Remember. Here, the reference position A is assumed to be the center coordinates of the captured image 70a.

  The control unit 80 adjusts the inclination angle of the galvano mirror 41a based on the deviation amount ΔX under the control of the driving unit 42. Further, the tilt angle of the galvanometer mirror 41b is adjusted based on the deviation amount ΔY. When the inclination angle of the galvanometer mirrors 41a and 41b changes, the incident angle of the illumination light with respect to the concave mirrors 31 and 35 changes. On the other hand, as described above, when the incident angle of the light flux on the concave mirrors 31 and 35 changes, the state of distortion included in the illumination light reflected by the concave mirrors 31 and 35 changes.

  Further, the control unit 80 starts detection and correction of the wavefront aberration of the eye E by the wavefront compensation unit 110. Based on the optical distribution (light reception signal) detected by the wavefront sensor 73, the control unit 80 controls the wavefront compensation optical system 110 so that the spread of the diffraction image of the reflected light from the fundus is minimized.

  The reflected light of the illumination light collected on the fundus travels back in the optical path via the resonant scanner 15, undergoes modulation by the wavefront compensation device 72, is reflected (deflected) by the beam splitter 71, and is reflected by the first imaging optical system 100b. Led to. The reflected light is focused on the pinhole of the pinhole plate 54 by the lens 53 via the polarization beam splitter 52 and is incident on the light receiving element 56 via the lens 55.

  In a state where the illumination light irradiation position is adjusted to the imaging position B by the deflection of the mirrors 41 a and 41 b of the galvano scanner 40, the control unit 80 scans the resonant scanner 15 and the galvano scanner 40 around the imaging position B. A local area of the fundus is scanned at a predetermined angle of view to obtain a fundus image in the designated local area. That is, the control unit 80 acquires a light reception signal for each pixel constituting the first fundus image 70a.

<Correcting the position of the shot image>
Here, the distortion correction by the distortion correction program will be described. FIG. 5 is a flowchart for explaining the operation of the distortion correction program. The control unit 80 acquires each pixel constituting the first fundus image 70a as described above, and obtains a distortion correction amount per pixel using the following distortion correction program.

  Note that the program may be configured to acquire a distortion correction amount by being executed during acquisition of a captured image, for example, or may be executed in advance before acquisition of a captured image. The structure which acquires may be sufficient. Of course, the configuration may be such that the distortion correction amount is acquired after acquiring the fundus image by being executed after acquiring the captured image.

  First, in step S1, when the imaging position B of the first fundus image 70a is designated on the second fundus image 70b, the inclination angle of the galvano scanner 40 is set based on the offset amount of the imaging position B with respect to the reference position A. .

  Next, in step S2, a light vector Rg reflected by the galvanometer mirror 41a is obtained. FIG. 6 is an explanatory diagram of the processing in step S2. The normal vector Ng of the galvanometer mirror 41a is expressed by the equation (1).

なお式（１）において、角度θ１はガルバノミラー４１ａの振れ角であり、角度４５度は、ガルバノミラー４１ａの振れ角の中心角度である。ガルバノミラー４１ａで反射する光線ベクトルＲｇは式（２）で示される。

In equation (1), the angle θ1 is the deflection angle of the galvano mirror 41a, and the angle 45 degrees is the center angle of the deflection angle of the galvano mirror 41a. The light vector Rg reflected by the galvanometer mirror 41a is expressed by equation (2).

式（２）において、ガルバノミラー４１ａに入射する光線ベクトルＬｇ、ガルバノミラー４１ａで反射する光線ベクトルＲｇである。なおＹ方向に走査されるガルバノミラー４２ｂについても上記の式（１）、（２）で同様に光線ベクトルＲｇが求められるので、ここでの詳細な説明は省略する。

In Expression (2), a light vector Lg incident on the galvano mirror 41a and a light vector Rg reflected by the galvano mirror 41a. Note that the light vector Rg is similarly obtained from the above equations (1) and (2) for the galvanometer mirror 42b scanned in the Y direction, and detailed description thereof will be omitted.

Next, in step S3, a three-dimensional ray trace is calculated. FIG. 7 is an explanatory diagram of the calculation of the angle difference of the incident angle in the X direction, and shows an enlarged view of the frame line D of the optical system of FIG. FIG. 8 is an explanatory diagram of the calculation of the angle difference between the incident angles in the Y direction, and shows a state in which the frame line D of the optical system in FIG. 7 and 8, illustration of the plane mirror 32 and the plane mirror 33 is omitted for convenience of explanation.
The light vector R reflected by the concave mirror 31 is obtained by Expression (3).

式（３）において、凹面鏡３１の反射光線ベクトルＲ、入射光線ベクトルＬ（ここで、ガルバノミラー４０で反射した光線ベクトルRｇをLと表記する）、法線ベクトルＮである。つまり入射光線ベクトルＬと法線ベクトルＮから反射ベクトルＲが求められる。なお、同様な計算は凹面鏡３５でも行われる。

In Expression (3), the reflected light vector R of the concave mirror 31, the incident light vector L (here, the light vector Rg reflected by the galvano mirror 40 is expressed as L), and the normal vector N. That is, the reflection vector R is obtained from the incident light vector L and the normal vector N. A similar calculation is performed by the concave mirror 35.

  Next, in step S4, the incident angle of the illumination light to the eye E (pupil) when distortion due to the concave mirror 31 is included is obtained. The incident angle is obtained from the light vector Rp incident on the eye E (pupil) (the light vector R reflected by the concave mirror 35 is expressed as Rp). Expression (4) shows the incident angle θ1 ′ in the X direction, and Expression (5) shows the incident angle θ2 ′ in the Y direction.

式（４）、（５）において、瞳に入射する光線ベクトルＲｐ，光線ベクトルＲｐのＸ成分Ｒｐｘ，Ｙ成分Ｒｐｙ，Ｚ成分Ｒｐｚとする。

In equations (4) and (5), the light vector Rp incident on the pupil, the X component Rpx, the Y component Rpy, and the Z component Rpz of the light vector Rp are used.

  Next, in step S5, the incident angle of the illumination light to the eye E (pupil) when distortion by the concave mirror 31 is included, and the illumination light to the eye E (pupil) when distortion is not included The angle difference between the incident angles is obtained. The angle difference between the incident angles in the X direction can be obtained by Expression (6), and the angle difference between the incident angles in the Y direction can be obtained by Expression (7).

なお式（６）において、ｔθ１は、歪が無い場合のＸ方向への光線入射角度、ｔθ２は、歪が無い場合のＹ方向への光線入射角度である。またｔは、ガルバノスキャナー４０から被検眼Ｅ（瞳）への倍率関係を考慮して定まる係数である。

In Equation (6), tθ1 is a light incident angle in the X direction when there is no distortion, and tθ2 is a light incident angle in the Y direction when there is no distortion. Further, t is a coefficient determined in consideration of the magnification relationship from the galvano scanner 40 to the eye E (pupil).

Next, in step S6, the angle difference between the incident angles in the XY directions is obtained for each pixel, and a distortion correction pattern (correction amount) for the image (each pixel) acquired at the imaging position B is obtained. Expression (8) is a calculation expression for obtaining the correction amount Δpixel (x) in the X direction in pixel units. Expression (9) is a calculation expression for obtaining the correction amount Δpixel (y) in the Y direction in pixel units.

なお式（８）、（９）でθpixelは歪が無いときの１画素（ピクセル）当りの角度であ
る。式（８）、（９）によって、１画素当りの歪の補正量が得られる。そして制御部８０は、ガルバノミラー４０及びレゾナントスキャナー１５による各走査位置に対応する画素に関して、ステップＳ２〜Ｓ６の処理によって、１画素当り（１画素毎）の歪の補正量をそれぞれ求める。つまり、制御部８０は、ステップＳ２〜Ｓ６の処理において、角度θ１、θ２に、第１眼底画像７０ａの各画素に対応する値を与えることで、１画素当りの歪の補正量を求める。
そして第１眼底画像７０ａを構成する画素全体に対する歪の補正量を求めることで、撮像位置Ｂに対する歪の補正パターンが得られる。

In equations (8) and (9), θpixel is an angle per pixel (pixel) when there is no distortion. Expressions (8) and (9) provide a distortion correction amount per pixel. And the control part 80 calculates | requires the corrected amount of distortion per pixel (each pixel) by the process of step S2-S6 regarding the pixel corresponding to each scanning position by the galvanometer mirror 40 and the resonant scanner 15, respectively. That is, the control unit 80 obtains a distortion correction amount per pixel by giving values corresponding to the respective pixels of the first fundus image 70a to the angles θ1 and θ2 in the processing of steps S2 to S6.
Then, a distortion correction pattern for the imaging position B is obtained by obtaining a distortion correction amount for all the pixels constituting the first fundus image 70a.

  In step S7, the control unit 80 corrects the distortion of the first fundus image 70a by displacing the coordinates of each pixel based on the distortion correction pattern with respect to the image acquired at the imaging position B. Based on the correction amount acquired for each pixel as described above, the control unit 80 determines the correction amount Δpixel (x) in the X direction and the correction amount Δpixel (y in the Y direction based on the correction amount acquired for each pixel as described above. ) Shift.

In this way, the coordinate position of the first fundus image 70a when it is assumed that no distortion has occurred due to at least one optical member arranged in the first imaging optical system 100b, and the actually acquired first fundus image 70a. Based on the displacement amount with respect to the coordinate position, the displacement amount of the coordinate of each pixel is set for each scanning position of the scanning member (optical scanner) at the imaging position.
According to the above processing, when the distance from the optical axis L1 of the first imaging optical system 100b to the imaging position B is increased, the displacement amount for correcting the coordinates in the first fundus image 70a is set large.

In the above calculation, it is assumed that scanning by the resonant scanner 15 (scanning in the X direction) is performed by the galvanometer mirror 41a. Therefore, the angle θ1 includes the inclination angle in step S1 and the inclination angle by scanning. (The value of the angle θ1 changes due to the change of the tilt angle by scanning, and the correction amount of the entire image can be determined.)
In the above description, the distortion correction amounts of all the pixels constituting the first fundus image 70a are sequentially obtained. Of the pixels constituting the first fundus image 70a, the pixel for obtaining the correction amount is selected in a predetermined step. May be. In this case, the influence of the aberration is reduced by performing the complementing process on the pixels on which the distortion correction calculation is not performed.

  In the above description, the distortion correction is performed after obtaining the distortion correction pattern corresponding to the entire first fundus image 70a for each imaging position B. In addition to this, the correction process of step S7 may be executed every time the correction amount per pixel is obtained in step S6 without obtaining the distortion correction pattern. As described above, the first fundus image 70 a is displayed on the monitor 70. FIG. 9 shows an example of a captured image 70a before and after distortion correction. FIG. 9A shows a first fundus image 70a before distortion correction, and FIG. 9B shows a first captured image 70b after distortion correction. It can be seen that the first fundus image 70a from which the aberration is removed is accurately displayed on the monitor 70 by executing the distortion correction program as described above.

  In step S8, it is determined whether or not the first captured image 70b has been captured. If it is determined that the imaging has not been completed, the process returns to step S1 again, and the imaging position B for the second fundus image 70b is switched. The designation of the imaging position B may be performed manually by operating the control unit 92 or may be performed automatically by the control unit 80. The control unit 80 repeatedly performs the processing of steps S1 to S8, obtains a distortion correction pattern for each imaging position B, and performs distortion correction processing. Then, when it is determined in step S8 that the shooting has been completed, a series of processing ends.

  When a plurality of first fundus images 70a are acquired, the control unit 80 connects the adjacent first fundus images 70a by image processing and displays them on the monitor 70 (as a collage). In the present embodiment, since the aberration of each first fundus image 70a is removed by the distortion correction pattern obtained for each imaging position B, the adjacent first fundus images 70a can be connected with high accuracy. In addition, using the image from which distortion has been removed, photoreceptor cell density analysis (calculation of photoreceptor cell density and area) and fundus morphology information (photoreceptor cell arrangement, optic nerve fiber bundle strike) can be confirmed with high accuracy.

  In the above description, an example in which a distortion correction program (calculation formula) for obtaining a distortion correction pattern is stored in the storage unit 81 has been described. In addition to this, the storage unit 81 may store a distortion correction pattern acquired in advance for each imaging position B (or a correction amount acquired in units of one pixel). In this case, when the imaging position B is designated, the corresponding distortion correction pattern (or the correction amount of the corresponding pixel) is called from the storage unit 81, and distortion correction processing is executed.

  In the above description, the distortion correction pattern is calculated for each imaging position B. However, the distortion correction pattern may be determined stepwise in accordance with the deviation amount (offset amount) from the reference position A. For example, a specific distortion correction pattern is selected according to the distance from the reference position A. Alternatively, a specific distortion correction pattern may be selected according to the direction from the reference position A.

  Alternatively, whether or not to perform distortion correction may be switched according to the state of occurrence of distortion for each imaging position B. For example, when the illumination light incident in the vicinity of the optical axis L1 of the concave mirrors 31 and 35 has a small distortion (for example, a predetermined range including the reference position A), the distortion correction is invalidated, and the peripheral light is a predetermined distance from the optical axis L1. It is also possible to set so that distortion correction is executed (when shooting is performed at an imaging position B that is a predetermined distance away from the reference position A). Alternatively, the examiner may be able to select a plurality of these processes.

  Furthermore, distortion caused by scanning by the resonant scanner 15 may be considered separately. A resonant scanner such as the resonant scanner 15 vibrates in a sinusoidal manner in conjunction with the driving of the motor, so that the scanning angle changes according to the deflection angle, causing variations in the resolution of the captured image. Therefore, the distortion of the resonant scanner 15 is acquired in advance by photographing a corrected image such as a dot matrix. Then, the first fundus image 70a may be formed by combining the distortion correction pattern obtained by the distortion correction program and the distortion information of the resonant scanner 15.

  Furthermore, in the above description, an example in which a concave mirror is used to make illumination light incident on the eye to be examined has been shown. In addition to this, the configuration of the present invention can be applied when illumination light is incident on the eye to be examined using various lenses and mirrors.

In the above description, in order to obtain the incident angle of the illumination light to the concave mirrors 31 and 35, the deviation amount (offset) of the imaging position B of the first fundus image 70a with respect to the reference position A of the second fundus image 70b is used. However, even if an angle detection unit (sensor) that detects the inclination angle of the galvano scanner 40 (galvano mirrors 41a and 41b) is provided and the incident angle of the illumination light is obtained based on the detection result of the inclination angle of the galvano mirrors 41a and 41b. good.

  Further, when performing cell density analysis (various calculations such as visual cell density and area calculation), the first fundus image 70a and the second fundus image 70b need not be displayed on the monitor 70. For example, when a signal for executing various calculations is input by operation of the control unit 92, the control unit 80 automatically adjusts the tilt angle of the galvano scanner 40, and the above-described according to the tilt angle of the galvano scanner 40. The distortion correction by the distortion correction program is executed, and the distortion correction including the signal received by the light receiving element 56 is performed.

  The imaging field angle of the fundus imaging optical system for imaging the local region is preferably 5 degrees or less. More preferably, it is 3 degrees or less. Of course, the present invention is not limited to this, and any angle of view that can capture a local region that cannot be imaged by a normal fundus camera or SLO may be used.

  The shooting angle of view set for imaging the local region is relatively smaller than 1/5 of the imaging angle of view of the imaging optical system for imaging the entire fundus (generally 30 ° to 50 °). And is suitable for obtaining an enlarged fundus image with respect to a local region of the fundus. In order to image the local region, a wavefront compensation device that compensates for the wavefront aberration of the eye to be examined included in the light from the fundus is preferably disposed in the optical path of the imaging optical system.

  Further, the present invention can be applied to various types of scanning fundus photographing apparatuses. For example, a fundus imaging apparatus for observing corneal endothelial cells of a patient's eye, a fundus imaging apparatus that scans a local region on the fundus of the eye to be examined in the transverse direction with emitted light emitted from a light source, for example, optical coherence tomography ( (OCT).

  As another example, the present invention can be applied to distortion correction of a wide-angle shot image. For example, the aberration of the wide-angle captured image captured by the second imaging unit 200 may be corrected based on the above-described distortion correction program (distortion correction pattern). It is particularly useful for the peripheral part of the image.

  That is, (1) a light source, an optical scanner for scanning the entire fundus of the eye to be examined in at least one of the X and Y directions with the emitted light emitted from the light source, and light from the fundus for detecting each scanning position And a fundus imaging optical system for imaging the entire fundus of the eye to be examined, to process a light reception signal from the detector at each scanning position to obtain a fundus image An image processing unit is provided, and the image processing unit corrects the distortion of the fundus image according to each scanning position.

(2) The image processing means is an image processing means for correcting the distortion of the fundus image by displacing the coordinates of each pixel in the fundus image when correcting the distortion of the first fundus image,
The displacement of the coordinates of each pixel is
Based on the amount of displacement between the coordinate position of the fundus image and the actually acquired coordinate position of the first fundus image when it is assumed that there is no distortion caused by at least one optical member arranged in the fundus imaging optical system. Thus, each scanning position of the optical scanner is set.

It is an external view of a fundus imaging apparatus. It is explanatory drawing of the optical system of a fundus imaging apparatus. It is a control block diagram of a fundus imaging apparatus. It is an example of the display screen of a monitor. It is a flowchart of operation control of distortion correction. It is explanatory drawing of the light vector of a galvanometer mirror. It is explanatory drawing of the calculation of the angle difference of the incident angle of a X direction. It is explanatory drawing of the calculation of the angle difference of the incident angle of a Y direction. It is an example of the 1st picked-up image before and after distortion correction. It is an example of the generation | occurrence | production state of the distortion which changes for every imaging | photography position.

DESCRIPTION OF SYMBOLS 1 Light source 15 Resonant scanner 40 Galvano scanner 41a, 41b Galvano mirror 31, 35 Concave mirror 70 Monitor 70a 1st fundus image 70b 2nd fundus image 80 Control part 81 Memory | storage part 92 Control part 100 1st imaging unit 100a 1st illumination optical system 100b First imaging optical system 110 Wavefront compensation unit 200 Second imaging unit 500 Fundus imaging device

Claims (3)

An optical scanner for scanning a local area on the fundus of the eye to be inspected with light emitted from a light source, and a detector for detecting light from the fundus for each scanning position in the local area, Fundus imaging optical system for imaging the local region;
A light deflector arranged in an optical path of the fundus imaging optical system, for deflecting light from the light source, and the position of the local region imaged by the fundus imaging optical system is set to be larger than the angle of view of the local region. An imaging position changing means for changing the light deflecting means up, down, left and right by driving and controlling the light deflecting means in a wide range;
Image processing means for acquiring a fundus image having the local region set by the imaging position changing means as an imaging range by processing a light reception signal from the detector;
The image processing unit acquires position information of the local region set by the imaging position changing unit, and generates distortion of the fundus image that differs for each position of the local region by the imaging position changing unit. A fundus photographing apparatus that performs correction based on the set position information of the local region. The image processing means includes
When correcting distortion of the fundus image,
Image processing means for correcting distortion of the fundus image by displacing the coordinates of each pixel in the fundus image,
The displacement amount of the coordinates of each pixel is
Based on the amount of displacement between the coordinate position of the fundus image and the actually acquired coordinate position of the fundus image when it is assumed that no distortion has occurred due to at least one optical member arranged in the fundus imaging optical system The fundus imaging apparatus according to claim 1, wherein the fundus imaging apparatus is set for each scanning position of the optical scanner in the local region set by the imaging position changing unit. The image processing means includes
The fundus imaging apparatus according to claim 1 or 2, wherein distortion of the fundus image is corrected according to a distance from an optical axis of the fundus imaging optical system to the local region set by the imaging position changing unit.

Images