JP2023099250A - auto focus device - Google Patents

Abstract

To provide an autofocus device for optical device which can maintain the high accuracy of focusing in an autofocus operation correspondingly to a deviation and a distortion of an optical system for a long period of time, and has a simple structure and a small size, and can reduce the influence of AF light to a sample used in a life science research.SOLUTION: In an autofocus device according to the present invention, an optical element array is arranged on an optical path from a light source to a sample, light emitted from the light source is separated by the optical element array, a plurality of light beams with different incidence angles are made to be incident on the sample, an imaging element for autofocus acquires a plurality of light images formed with reflection light from the sample, and a control unit calculates an interval between the plurality of light images, acquires a difference to a grating interval reference value of the light image that becomes a reference value in preset focusing and outputs a command signal of the movement operation to a distance adjustment mechanism.SELECTED DRAWING: Figure 1

Description

The present invention is an autofocus device installed in an optical device, which prevents defocus due to deviation or distortion of an optical element arranged in an optical system (hereinafter referred to as an optical system), and prevents a sample from being defocused. The present invention relates to an autofocus technique for achieving high-speed focusing while suppressing the amount of light given to the subject.

In recent years, similar to optical devices such as cameras, microscopes equipped with an autofocus device that automates manual focusing operations have been developed, and have come to be used for quality control in life science research and industrial fields. It's here.

In modern life science research, it has become possible to process large amounts of data for screening and statistical analysis due to the reduction in computational costs using computers. For this reason, there is a growing need for automation of microscopes for observing a large number of samples and acquiring a large amount of data.

Autofocus is one of the important elemental technologies for automating microscopes. In optical microscopes, the following two methods are mainly used as autofocus methods.

The first method is the contrast method. In this method, the objective lens or stage is moved in the direction of the optical axis, the actually formed image is photographed with an image sensor, and the focus is scanned while evaluating the sharpness of the image by image processing. The position of maximum sharpness is the in-focus position. In this case, since scanning in the focal direction is required, it takes a long time to focus. In addition, since the sample is exposed to a large amount of light, fading is a problem during fluorescence observation. becomes.

The second method is a phase difference method. In this method, the stage is irradiated with near-infrared light from one direction, and the reflected light is used to form an image of the diaphragm on the image pickup device. , determine whether the current focus state is rear focus (when the focus is on the far side of the sample) or front focus (when the focus is on the front side of the sample), and This is a method of driving a distance adjustment mechanism that moves the objective lens or stage in the optical axis direction. In this case, the scanning described above is not required, and since near-infrared light is used, the problem of fading does not occur, so microscopes of this type are often used in life science research.

The second method is mainly used in life science research, but there is also a problem that the focus shifts after a long period of time. On the other hand, there is a shortage of experimenters for acquiring data in research and development in the field of life science, so there is a high need for automation of microscopic photography in the field. In order to automate the microscope, it is essential to develop an autofocus technology that does not shift focus even after a long period of time. Also, irradiation of a sample with near-infrared light does not cause the problem of fading due to fluorescent molecules, but it does affect the measurement results to some extent. A small amount of near-infrared light passes through the dichroic mirror, and there is also the problem of light leaking into high-sensitivity observation cameras such as those used in life science research. Therefore, the amount of near-infrared light that irradiates the sample is required to be suppressed as much as possible.

As mentioned earlier, the phase-contrast method is often used in the automation of microscopes. There is a problem that the focus is on a position shifted from the glass surface, and the focus is shifted from the target object to be observed. When the focal position is within the range of the optical axis direction, such as with a confocal microscope or fluorescence microscope, that is, when the observation target is the entire thick cell, defocusing is allowed as long as the focal position is within the thickness of the cell. sometimes you can. However, when observing a glass surface, such as with a total internal reflection microscope, which is a type of super-resolution microscope, the focal point shifts from the glass surface in the direction of the optical axis, causing the image to blur and the captured image to be used for observation. An impossible situation arises. From the above, in the super-resolution microscope, the current situation is that the automation of microscopic photography has not particularly progressed, and the photography is still performed by manual focusing operation. In addition, it is difficult to achieve long-term time-lapse with a super-resolution microscope because high accuracy in autofocus operation cannot be maintained for a long period of time.

Patent document 1 or patent document 2 is disclosed as a prior art of the phase difference method which tackles the above-mentioned problem.

WO2019/159627A1 Japanese Patent No. 5626367

In the conventional autofocus technology, light used for focusing is made incident from one direction, and the position of the aperture image detected by the imaging device is controlled to be in the center. However, according to this method, since the position of the diaphragm changes due to the shift or distortion of the optical system, it is difficult to maintain the in-focus state for a long time.

A phase-difference autofocus technique developed to solve the problems of the prior art is disclosed. In Patent Document 1, first, autofocus light (hereinafter referred to as AF light) incident on the diaphragm is incident from two directions, and each AF light irradiated to the sample passes through the diaphragm. An image of the formed image (hereinafter referred to as an aperture image) is obtained, and the position of the aperture image is measured from the image. Then, the difference between the image positions of the diaphragm images is calculated, and the stage is moved in the Z direction (optical axis direction) and controlled so that the difference becomes a constant value. When the optical system is distorted, the position of the diaphragm image accompanies the same change in each image. It does not change. Therefore, it can be said that this technique is capable of focusing on the glass surface even after a long period of time. However, in Patent Literature 1, a galvanomirror is used to change the angle of the AF light multiple times, and focusing is performed by capturing an aperture image, so a mechanism for moving the galvanomirror is required. In addition, it is necessary to photograph the aperture image multiple times. As a result, there is a problem that autofocusing takes a long time, and the number of constituent parts increases, which poses a problem of miniaturization and cost reduction.

Japanese Patent Application Laid-Open No. 2002-200001 discloses a technique of separating AF light reflected at one point on a sample glass surface by a microlens array and photographing with an autofocus camera for autofocusing. In the autofocus camera, light images condensed on the sample are separated by a microlens array and displayed in a grid pattern. Moving the stage along the optical axis changes the spacing of the grid. Focusing is performed by controlling a distance adjustment mechanism that moves the objective lens or the stage in the optical axis direction so that the distance becomes a predetermined value. Positional deviation of the AF light image occurs due to deviation or distortion of the optical system, and the position of the grating moves, but the spacing of the grating does not change, so it is possible to maintain the focused position for a long time. Autofocus is disclosed. However, if the light concentrated at one point on the sample is split into two directions by a half mirror in the observation system, and if the light is split by the number of gratings by the microlens array, the intensity of the light becomes weaker and the exposure for photography during autofocusing. There is a problem that the autofocus speed decreases because the time is long. Therefore, in the prior art of Patent Document 2, it is necessary to make the light intensity stronger than at least the number of gratings to be separated compared to Patent Document 1. When the light intensity is increased, even near-infrared light causes the following problems. First, as the light intensity increases, the phototoxicity to biological samples such as cells increases. When cells are affected by phototoxicity, the problem arises that the current state of life cannot be accurately observed. Secondly, if the maximum light intensity of the AF light is high at the focal point of the sample, there is a problem that the light is likely to enter the observation camera as stray light.

In view of the above problems, the inventors of the present application have developed an autofocus device that is resistant to optical system deviation and distortion and that suppresses the amount of AF light that irradiates the sample. If high accuracy of autofocus operation can be maintained for a long time with low phototoxicity, the probability of defocusing during automatic imaging with a super-resolution microscope can be suppressed, and the toxicity to cells is reduced, which is life-threatening. An accurate observation of the state can be achieved. Also, in the case of fluorescence microscopes, etc., if the autofocus accuracy is high, images can be taken for a long time at a fixed height from the glass surface, and the focal point does not shift in the direction of the optical axis. There is an advantage that it can be improved and the reliability of the data increases.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems. An object of the present invention is to provide an autofocus device for an optical device that is structured and compact. A further object of the present invention is to provide an autofocus device for an optical device that can reduce the influence of AF light on samples used in life science research.

In order to solve the above problems, the autofocus device of the present invention is an autofocus device that automatically adjusts the distance between a sample to be observed and an objective lens for focusing, wherein an optical element array extends from a light source to the sample. The light emitted from the light source is separated by the optical element array, and a plurality of lights with different incident angles are made incident on the sample, and the autofocus imaging device is arranged on the optical path of the A plurality of light images formed by the reflected light from the sample are acquired, and the control unit calculates the intervals between the plurality of light images, and calculates the distance between the plurality of light images, which is set in advance as a reference value for focusing. It is characterized by obtaining a difference from the grid spacing reference value and outputting a movement operation command signal to the distance adjusting mechanism.

Further, the autofocus device of the present invention can be configured by moving an optical element arranged between the optical element array and the autofocus imaging device in the optical axis direction of the light emitted from the light source, or by moving the grating interval. The focal position in the optical axis direction is changed or corrected by changing the reference value.

Further, the autofocus device of the present invention is characterized by arranging a polarizer and a wavelength plate between an optical element for extracting or separating autofocus light from an observation optical system and an autofocus imaging device. .

Further, a microscope of the present invention is characterized by having the aforementioned autofocus device.

In Japanese Patent Application Laid-Open No. 2002-100000, the autofocus is performed by operating the galvanomirror to change the angle of light incident on the diaphragm, capturing images a plurality of times, and detecting the state of defocus. On the other hand, according to the autofocus device of the present invention, a plurality of light beams incident on different positions on a plane perpendicular to the optical axis have different angles. It is possible to measure the distance, acquire the state of defocus, and adjust the focus using a distance adjustment mechanism (hereinafter referred to as a focus movement mechanism) that moves the objective lens or stage in the optical axis direction. have some effect.

According to the autofocus device of the present invention, since a galvanomirror is not required, the mechanism of the optical system can be simplified. In addition, since it is not necessary to photograph the optical image multiple times by changing the angle of the galvanomirror, the time required to obtain the focus state can be shortened, and the speed of autofocus can be increased.

Furthermore, since a mechanism for moving a galvanomirror or the like is not required, the number of parts can be reduced, and the manufacturing cost of the autofocus device can be reduced.

In Patent Document 2, a single line of light is incident locally on the sample to be observed and then divided by the lens array, so the sample may be affected by phototoxicity depending on the intensity of the AF light. On the other hand, according to the autofocus device of the present invention, since the AF light is separated by the lens array and then irradiated onto the sample to be observed, there is no need to separate the reflected light after that, and the reflected light is strong enough to affect the sample. No light is incident locally on the sample. For example, when comparing the case where the light is dispersed by a lens array with a 3×3 grid and is irradiated onto the sample, and the case where the sample is irradiated with a single line of light without being dispersed, the result is that the light is concentrated at one point on the sample. The amount of light to be applied is 1/9. Therefore, by separating the AF light before irradiating the sample, the intensity of the light can be reduced to about 1/10, and the effect on the sample can be suppressed. Furthermore, by placing a shielding plate before irradiating the sample to reduce the number of light rays passing through, it is possible to further suppress the amount of light applied to the sample.

Furthermore, according to the autofocus device of the present invention, the lenses placed between the sample and the autofocus camera are, in principle, only the objective lens and the imaging lens for AF. Loss of light quantity can be suppressed.

According to the autofocus device of the present invention, the position of the diaphragm is measured from the images obtained by the light incident at a plurality of angles at the same time, and the relative difference is obtained. The optical system is resistant to distortion and is less likely to be out of focus even after a long period of time.

INDUSTRIAL APPLICABILITY According to the microscope of the present invention, in life science research that requires acquisition of a large amount of data, it is possible to inexpensively increase the speed of microscopic observation and save labor. In addition, by combining a microscope equipped with an autofocus device with an automatic dispensing machine, it becomes possible to apply it to automation of drug screening. Specifically, an image of the state of the living body before adding the drug is taken with an automated microscope, and an image of the state of the living body is taken again after the drug is added, and the effects of the drug are evaluated by comparing both images. can do.

1 is a conceptual diagram showing the basic configuration of an optical system of an autofocus device 1 of the present invention; FIG. 2 is a diagram showing an image of AF light captured by an autofocus camera 50 in the optical system of FIG. 1; FIG. FIG. 5 is a diagram showing a flowchart for calculating a deviation distance amount from an in-focus position in the autofocus device 1 of the present invention; It is a figure explaining the principle of the autofocus apparatus 1 of this invention. 5 is a diagram showing an image of the diaphragm 30 photographed by the autofocus camera 50 in the optical system of FIG. 4. FIG. FIG. 10 is a diagram showing an example of an image of an aperture image AI(n) in an N×N grid obtained by an autofocus camera 50 having a wide imaging range; 1 is a diagram showing one embodiment of an autofocus device 2 of the present invention; FIG. FIG. 4 is a diagram showing another embodiment of the autofocus device 3 of the present invention;

FIG. 1 is a conceptual diagram showing the basic configuration of the optical system of an autofocus device 1 of the present invention. In FIG. 1, the autofocus device 1 is a camera (hereinafter referred to as an AF camera 50) including a light source 20, an optical element array, a half mirror 40, an objective lens 110, an AF imaging lens 66, and an autofocus imaging device. ). A light source 20 in FIG. 1 indicates a point light source 20 . A light ray R emitted from a point light source is indicated by a solid line (same below). A planar lens array 10 in which a plurality of lenses are arranged in the XY axis direction is suitable for the optical element array. In FIG. 1, it is assumed that three lenses are arranged in the X-axis direction and three lenses are arranged in the Y-axis direction centering on the Z-axis. The arrangement of the lenses is arbitrary, but for ease of explanation, a grid pattern is used in this specification. In the optical system of FIG. 1, the point light source 20, the surface of the sample 100, and the AF camera 50 are conjugate.

Light emitted from the point light source 20 is AF light. The AF light passes through the lens array 10, the half mirror 40, and the objective lens 110, and the AF light image RI is formed on the sample 100. At that time, the lens array 10 splits the light so that the light spot is at a corner. A plurality of these square images are formed and are basically formed in the same lattice pattern as the arrangement of the lens array 10 . In FIG. 1, only rays in the X-axis direction are shown, and rays in the Y-axis direction are not shown. Therefore, if the light emitted from the point light source 20 is represented by one ray that travels straight on the optical axis (on the Z axis) and two rays that pass from the optical axis in the positive or negative direction of the X axis at an angle, Three light spots are imaged on the sample 100 .

At this time, the angle of the incident light beam differs depending on the position of each lens of the lens array 10 in which the lenses are arranged in a grid pattern. Among the three rays, the central image is formed by rays traveling straight along the optical axis (Z-axis) in the positive direction. When a ray emitted from the point light source 20 passes through a lens distant from the Z-axis, the angle of the ray with respect to the Z-axis increases as the distance from the Z-axis increases. Actually, since a plurality of lenses are also arranged in the Y-axis direction, the optical image RI becomes a grid-like two-dimensional image. The AF light emitted from the point light source 20 to the sample 100 is reflected by the sample 100 (indicated by a dashed line in the drawing), and further, a part of the light amount is reflected by the half mirror 40, and is reflected by the AF camera 50. be filmed. Since the surface of the sample 100 and the surface of the AF camera 50 are conjugate, the AF camera 50 also captures a grid-like image of the AF light emitted from the point light source 20 .

FIG. 2 is a diagram showing an image of AF light captured by the AF camera 50 in the optical system of FIG. The AF optical image RI split by the lens array 10 is photographed in a grid pattern on the autofocus imaging device. A square frame is a photographing area FV of the AF camera. When the position of the objective lens 110 or the stage (not shown) is changed in the Z-axis direction, the lattice spacing of the AF optical image RI widens or narrows in the captured image.

In the AF optical image RI, the grid spacing in focus is determined by the spacing of each lens in the lens array 10, the focal point of the lens, and the magnification of the objective lens 110. FIG. Therefore, the direction and distance for focusing by the autofocus operation are derived by comparing the size of the grid spacing and calculating the dimensional difference between the grid in focus and the grid image currently being photographed. be able to. Autofocus is realized by operating a focus movement mechanism that moves the objective lens 110 or the stage in the optical axis direction so that the grating interval always maintains a predetermined interval.

FIG. 3 shows a flowchart for calculating the amount of deviation from the in-focus position and operating the focus movement mechanism in the autofocus device 1 of the present invention. The flowchart is constructed based on FIG. 1 for ease of explanation. Further, the flowchart is programmed and executed by the control unit.

When the autofocus is started (S10), in step S20, the lens array 10 splits the AF light emitted from the point light source 20 into a grid pattern, and the AF light image RI formed on the sample 100 is captured by the AF camera 50. to shoot. At this time, it is determined whether or not the photographing of the spectroscopic AF optical image RI has succeeded (S30). process and exit. If the AF optical image RI has been successfully photographed, in step 40, the lattice spacing Δ (delta) of the image obtained by photographing the AF optical image RI is measured. Subsequently, the difference dx=Δ−D _f between the grating interval value D _f treated as a reference value for focusing set in advance according to the characteristics of the optical system and Δ is calculated (S50). More specifically, the grating interval value _Df is, in principle, the distance between the AF optical images RI formed when the lens array 10, the surface of the sample 100 and the AF camera 50 are conjugate in the AF optical system. After that, the displacement distance amount in the Z-axis direction dz=a(Δ−D _f )=a×dx is calculated by multiplying by the proportionality constant a, and the focus moving mechanism is driven in the Z-axis direction by dz (S60). In step S70, it is determined whether or not to end the autofocus operation. When the autofocus operation is terminated, the termination process of step S80 is performed. When the autofocus operation is continued, the process returns to the photographing of the AF optical image RI in step S20, and the above steps are repeated. A specific method of calculating dx and dz will be described later.

In Patent Document 1, AF light is applied to a sample at different angles, and a plurality of images taken by an AF camera are compared to calculate the difference in the position of the AF light image. According to the focusing device 1, by using the lens array 10, AF optical images RI at a plurality of angles are acquired in a single photographing operation, and the difference between the AF optical images RI is calculated. You can get results with the same precision at high speed. Further, even if the optical system is misaligned or distorted, according to the autofocus device 1 of the present invention, only the position of the grating formed by the AF optical image RI is changed, and the grating interval is the same as that of the optical system. Since there is no change from before deviation or distortion occurs, there is no effect on autofocus accuracy.
In the autofocus device of Patent Document 1, since a plurality of images are taken with a time difference (several hundred milliseconds), if the optical system shifts or is distorted between the first and second shots, the focus movement mechanism However, according to the autofocus device 1 of the present invention, an image for calculating the difference can be obtained in one shot. Therefore, even if the optical system is misaligned or distorted for about several hundred milliseconds, it does not affect the calculation accuracy of the difference up to the in-focus position. Therefore, the autofocus device 1 of the present invention has a very strong effect on optical system shifts and distortions on all time scales.

The advantage of simultaneously irradiating the light divided into multiple angles is that the structure is very simple because there is no need for moving parts such as galvanometer mirrors, and the autofocus operation can be processed in a short time because the shooting can be done only once. It is possible.

An advantage of arranging the lens array 10 before the sample 100 on the optical path is that the maximum light intensity applied to the sample 100 can be dispersed by dispersing the AF light that irradiates the sample 100 . In Patent Document 2, the maximum light intensity at the condensing point on the sample increases, and the AF light is likely to enter a camera for observing the sample as stray light. There is an upper limit. On the other hand, if the maximum light intensity of the AF light is lowered, the exposure time required for autofocusing becomes longer, and the time required for the autofocusing operation becomes longer. Since the AF light is dispersed in the above-described manner, the reflected light from the sample 100 can be captured by the AF camera 50 without passing through the lens array 10, unlike in Patent Document 2. As a result, compared to Patent Document 2, the amount of light applied to the sample 100 can be suppressed by the number of gratings to be separated, and the autofocus exposure time can be shortened. Play.

By changing the configuration of the optical system to place a shielding plate between the half mirror 40 and the lens array 10 to reduce the number of light rays passing through, the amount of light applied to the sample 100 can be further suppressed. However, if the light rays that are important for autofocus operation are reduced, the accuracy of autofocus will be reduced and the range will be narrowed, so careful design is required. The details of the relationship between the autofocus accuracy and focus range and the number of grids will be described later.

FIG. 4 is a diagram explaining the principle of the autofocus device 1 of the present invention. A basic autofocus optical system (hereinafter referred to as an AF optical system) comprises a light source 20, a lens A60, a diaphragm 30, a lens array 10 and a lens B62. Here, the lens A60 is a condensing lens, and the lens B62 is a lens for forming the lattice of the aperture image AI. The diaphragm 30 is not essential in terms of construction, and it is sufficient if the characteristics of the light source 20 are suitable for AF light as shown in FIG. For example, it can be applied by using a laser focal point or by making the shape of the diaphragm 30 polygonal. AF light emitted from the light source 20 toward the sample passes through the aperture 30, and the reflected light of the aperture image AI formed on the sample is separated using a half mirror 40 or the like at an arbitrary position on the reflected light path. The current focus state is detected and adjusted to be in focus by forming an image on the take-out AF camera 50 . In FIG. 1, the AF optical image RI formed by the AF light emitted from the light source was captured by the AF camera 50. In FIG. AI is filmed. The virtual plane P, the sample, and the AF camera 50 are in a conjugate relationship.

The light source 20 is not particularly limited, such as a light emitting diode (LED (Light Emitting Diode)), laser light, or the like. As the wavelength of light, a wavelength not used in microscopic photography is selected, but in principle any wavelength is acceptable. It is also assumed that the lens array 10 is arranged in the XY direction, that is, as a surface of N×N (N is an integer) perpendicular to the Z axis. In this specification, the case where N is an odd number will be explained, but in principle there is no need for the center of one lens of the lens array 10 to be on the optical axis, and there is no problem even if it is an even number. The lens array 10 has a central lens on the Z axis when N is an odd number, and n=0 (where n is an integer). It is preferable to arrange so that the number of n increases by ±1.

Since the AF light is emitted from the light source 20 with a predetermined radiation angle, the lens A60 is used to condense the light, pass through the diaphragm 30, approximate the point light source 20, and then irradiate the lens array 10, the diaphragm A grid-like image is obtained with the image AI as the intersection point. FIG. 4 shows only light rays that have passed through individual lens centers in the lens array 10 .

FIG. 5 is a diagram showing an image of the diaphragm 30 photographed by the AF camera 50 in the optical system of FIG. As described above, the aperture image AI appears in a grid pattern. Regarding the aperture images AI at the intersections of the grid, if N is an odd number, the aperture image AI at the center of the image is the aperture image AI(0) of the lens with n=0. The number of n increases by ±1 each time the aperture image AI moves away from the center of the image. Since the lens with n=0 is on the optical axis, the position of the diaphragm image AI(0) with n=0 captured by the AF camera 50 does not change even if the focus changes. If the lens of n=0 is displaced from the optical axis, the position of the aperture image AI(0) changes in proportion to the change in focus.

When an AF ray emitted from a light source 20 passes through the lens array 10, the angle of the AF ray with respect to the Z-axis increases as the AF ray passes through a lens with n larger away from the Z-axis. The same applies to the Y-axis direction. Therefore, the aperture image AI is a two-dimensional image in the same N×N grid pattern as the lens arrangement. FIG. 6 shows an example of an image of an aperture image AI(n) in the form of an N×N lattice captured by the AF camera 50 having a wide imaging range.

The AF light split by the lens array 10 is condensed by the lens B62 to form an aperture image AI on a virtual plane P perpendicular to the optical axis. In an actual microscope, the virtual plane P, the sample plane, and the plane of the AF camera 50 are conjugate. In order to operate the focus movement mechanism to the in-focus position and perform autofocus, it is necessary to calculate the movement direction and movement amount of the focus movement mechanism from the image captured by the AF camera 50 . For this purpose, first, the grid spacing, angle and diameter of a plurality of aperture images AI incident on the sample are calculated.

When one lens of the lens array 10 is on the optical axis, from the center (n=0) on the plane P on the optical axis (Z-axis) of the AF light to n (= . , −2, −1, 0, 1, 2, . . . , n, . Calculation of the angle θ(n) (see FIG. 4) and the diameter D _P (n) of the aperture image AI(n) (see FIG. 6) yields the following. FIG. 6 shows the state in which the lens at the center of the lens array 10 is arranged on the optical axis. No need to place.
(Formula 1) Δ(n)=(f ₀ /f)nd
(Formula 2) tan θ(n)=2nd/ _f0
(Formula 3) D _P (n)=D×f ₀ /f
where f and _f0 are the focal lengths of lens array 10 and lens B 62 respectively, d is the distance between adjacent lenses in lens array 10, and D is the diameter of aperture 30. From Equation 1, the lattice spacing Δ between the adjacent diaphragm images AI(n) and AI(n+1) is given by Δ=(f ₀ /f)d.

When the focus shifts in the direction of the optical axis by a distance dz, the shift dx in the distance between the aperture image AI(n) and the optical axis (Z-axis) means that the aperture image AI(0) at n=0 moves from the optical axis. Therefore, it is equal to the difference in distance between the aperture image AI(0) of n=0 and the n-th aperture image AI(n). Therefore, dx and dz are as follows.
(Formula 4) dx=Δ(n)−D _f
From FIG. 4, tan θ(n)=dx/dz. From Equation 2, dx/dz=2nd/ _f0 .
(Formula 5) dz={f ₀ /(2nd)}×dx
Therefore, if a=f ₀ /(2nd), then dz=a×dx.

It can be seen from Equation 5 that if dz is constant, the change in dx increases as n increases. Therefore, high accuracy can be achieved by performing calculations using grid intervals with a large n. On the other hand, when n is large, the movement of the light spot of the aperture image AI forming the lattice becomes large, so that the selected n aperture image AI(n) may deviate from the photographing area of the AF camera 50 in some cases. Sometimes I end up doing it. If the selected n-th aperture image AI(n) is out of the imaging area, select the aperture image AI with a smaller n within the imaging area, for example, the n-1th aperture image AI(n-1). Then, the focus moving mechanism is driven to return the selected n aperture images AI(n) to the photographing area of the AF camera 50, thereby realizing a wide autofocus range. That is, by switching n when performing calculations, it is possible to achieve both a wide autofocus range and high accuracy.

As described above, the current defocus distance amount dz (equation 5) can be known from dx calculated by the control unit using the image captured by the AF camera 50 . By driving the focus movement mechanism so that the deviation distance amount dz becomes 0, wide-range, high-speed, and high-precision autofocus can be achieved.

FIG. 7 is a diagram showing one embodiment of the autofocus device 2 of the present invention. The observation optical system is composed of an observation camera 130 , an observation imaging lens 120 , a dichroic mirror 44 and an objective lens 110 . On the other hand, the AF optical system comprises a light source 20, a lens A60, a diaphragm 30, a lens B62, a lens C64, a half mirror 40, a dichroic mirror 44, an objective lens 110, an AF imaging lens 66, and an AF camera in order through which AF light passes. 50. Here, the lens A60 is a condensing lens, the lens B62 is a lens for forming the lattice of the diaphragm image AI, and the lens C64 is a lens for forming the diaphragm image AI. The observation optical system is configured in the microscope device. The observation optical system and the AF optical system share some optical elements. In Example 1, the dichroic mirror 44 and the objective lens 110 are used as common components. In the AF optical system of Example 1, the diaphragm 30, the virtual plane P, the surface of the sample 100 and the AF camera are in a conjugate relationship.

The autofocus device 2 includes a control unit, which is not shown in FIG. The control unit has at least a power supply unit, a calculation unit that calculates the displacement distance (dz), etc., an input unit that acquires an image from the AF camera 50, an output unit such as a command signal to the focus movement mechanism, and a storage unit. . However, it is not limited to this, and may have functions similar to those of a general-purpose computer.

The AF light emitted from the light source 20 is condensed through the lens A60 and enters the diaphragm 30 to approximate the light emitted from the point light source 20 . After passing through the diaphragm 30, the AF light passes through the lens array 10 and the lens B62 and is focused on the position of the virtual plane P as the diaphragm image AI. On the virtual plane P, a grid-like aperture image AI split by the lens array 10 appears. The AF light that forms the grid-like aperture image AI on the virtual plane P passes through the lens C64 and the half mirror 40, and the dichroic mirror 44 reflects light of wavelengths other than the specific region, and enters the objective lens 110. The sample 100 is irradiated with an image formed in a grid pattern.

Reflected light from the sample 100 passes through the objective lens 110 and the dichroic mirror 44 , is reflected by the half mirror 40 , passes through the AF imaging lens 66 , and forms an image on the imaging device of the AF camera 50 . The AF camera 50 captures an aperture image AI on the virtual plane P and in the same grid pattern as the sample 100 surface. The method of separating the reflected light of the AF light by arranging the half mirror 40 between the dichroic mirror 44 and the lens C64 can be used when constructing an integrated optical observation apparatus in which the autofocus device 2 is built into the microscope. This is effective because the size of the device 2 can be reduced and the number of components can be reduced.

When the objective lens 110 or the stage is moved in the optical axis direction to change the position of the sample 100, the grating interval changes. The optical axis direction corresponding to the difference ( _dx ) between the changed grating spacing and the predetermined grating spacing value _Df is adjusted so that the changed grating spacing value becomes equal to the predetermined grating spacing value Df. Auto-focusing is performed by moving the objective lens 110 or the stage in the optical axis direction by the distance difference, ie, the displacement distance amount (dz). The direction of movement at that time can be determined by the positive or negative value of the shift distance amount (dz).

To move the focal position from the glass 102 surface in the optical axis direction, change the lattice spacing value _Df or physically change the position of the lens C64 or the AF imaging lens 66 in the optical axis direction. The former can be performed at high speed because the lens does not actually move. Therefore, it is necessary to correct the blurred diaphragm image AI to a sharp image by image processing and detect the center of the diaphragm image AI before calculating the displacement distance (dz).

An image of the sample 100 itself is captured by the observation camera 130 through the objective lens 110 of the observation optical system, the dichroic mirror 44 and the observation imaging lens 120 . The dichroic mirror 44 reflects only the AF light and transmits the illumination light for observation. In Example 1, an illumination light source for observation is not shown.

FIG. 8 is a diagram showing another embodiment of the autofocus device 3 of the present invention. The observation optical system is composed of an observation camera 130 , an observation imaging lens 120 , a dichroic mirror 44 and an objective lens 110 . On the other hand, the AF optical system is composed of a light source 20, a lens A60, a diaphragm 30, a lens B62, a polarizer, a lens C64, a wave plate, a dichroic mirror 44, an objective lens 110, and an AF camera 50 in order through which AF light passes. . The polarizer is preferably a polarizing beam splitter 42 capable of splitting incident light into P and S polarization components. Also, the wavelength plate is preferably a λ/4 wavelength plate 70 that adds a phase difference of π/2 (=λ/4) to the incident light in the plane of polarization. By combining the polarizing beam splitter 42 and the λ/4 wavelength plate 70, there is an advantage that unnecessary return reflected light can be removed. Here, lens A60, lens B62 and lens C64 play the same role as in FIG. The observation optical system and the AF optical system share some optical elements. In Example 2, the dichroic mirror 44 and the objective lens 110 are used as common components. In addition, although the controller is not shown in FIG. 8, it may have the same configuration as the first embodiment. In the AF optical system of Example 2, the virtual plane P, the sample 100 plane, and the AF camera 50 are in a conjugate relationship.

The AF light emitted from the light source 20 is condensed by the lens A60, passes through the diaphragm 30, and the diaphragm image AI is formed on the virtual plane P in a grid pattern, as in the first embodiment. In Example 2, the AF light that has passed through the virtual plane P passes through the polarizing beam splitter 42, the lens C64, and the λ/4 wavelength plate 70, is reflected by the dichroic mirror 44, passes through the objective lens 110, and forms a grating on the sample 100. An aperture image AI is formed.

Reflected light from the sample 100 passes through the objective lens 110 and is reflected by the dichroic mirror 44, then passes through the λ/4 wavelength plate 70, then the lens C64, is reflected by the polarizing beam splitter 42, and is reflected by the imaging device of the AF camera. image on. The AF camera 50 captures an aperture image AI on the virtual plane P and in the same grid pattern as the sample 100 surface.

When moving the focus position from the surface of the glass 102 in the optical axis direction, change the lattice spacing value _Df , or physically change the position of the lens C64 or the AF imaging lens 66 in the optical axis direction. . The former can be performed at high speed because the lens does not actually move. Therefore, image processing is performed to correct the blurred aperture image AI to a sharp image and to detect the center of the aperture image AI before calculating the displacement distance (dz).

The advantage of the configuration of the AF optical system of Example 2 is that it is not necessary to dispose the half mirror 40 between the lens C64 and the objective lens 110. FIG. When attached to a microscope as the autofocus device 3, it is difficult to split the light into two directions by inserting the half mirror 40 between the objective lens 110 and the lens C64 due to the structure of the microscope. Therefore, in the configuration of the second embodiment, the reflected light from the sample 100 is separated by disposing the polarizing beam splitter 42 after the reflected light path of the lens C64, not between the lens C64 and the objective lens 110. FIG. When the autofocus device 3 is attached to the microscope, the focal length of the lens C64 must be set long because the AF optical system is combined with the observation optical system. 42 can be sufficiently secured. In this case, the λ/4 wavelength plate 70 must be placed in front of the reflected light path of the polarizing beam splitter 42 . As a result, as compared with the first embodiment, there is an advantage that the structure of the autofocus device 3 has more space and the design becomes easier. In addition, since the half mirror 40 is not used to split the light into two directions, the total amount of light can be incident on the AF camera 50, so the amount of light to the sample 100 can be reduced, thereby suppressing phototoxicity. be able to.

An image of the sample 100 itself is captured by the observation camera 130 through the objective lens 110 of the observation optical system, the dichroic mirror 44 and the observation imaging lens 120 . The dichroic mirror 44 reflects only the AF light and transmits the illumination light for observation. In Example 2, an illumination light source for observation is not shown.

The autofocus device of the present invention is particularly effective in the life science research field. The autofocus device of the present invention is not affected by misalignment or distortion of the optical system even when the autofocus operation is performed for a long time, so it is possible to realize automation of life science observation. is. By scanning the stage on which a large number of samples are set in the XY directions while continuing the autofocus operation, it is possible to automatically photograph a large number of microscope images of the samples.

Furthermore, by combining with an automatic pipetting machine, it can be used for drug screening. Perform drug screening by taking an image of the state of the living body before adding the drug solution with an automated microscope, taking an image of the state of the living body again after adding the drug with an automatic dispenser, and comparing both images. can be done. In addition, there is a possibility that automatic imaging can be applied to pathological examinations.

1 autofocus device 2 autofocus device (one embodiment)
3 Autofocus device (another embodiment)
10 lens array 20 light source (or point light source)
30 diaphragm 40 half mirror 42 deflection beam splitter 44 dichroic mirror 50 autofocus camera (AF camera)
60 Lens A
62 Lens B
64 Lens C
66 AF imaging lens 70 λ/4 wavelength plate 100 Sample 102 Glass 110 Objective lens 120 Observation imaging lens 130 Observation camera

D aperture diameter d lens array interval f lens array focal length f ₀ lens B focal length AI aperture image FV AF camera imaging area P virtual surface R light ray RI light image Δ(n) lattice interval θ(n) light ray incident angle D _P (n) Diaphragm image diameter

Claims (4)

In an autofocus device that automatically adjusts the distance between the sample to be observed and the objective lens to focus,
the optical element array
arranged on the optical path from the light source to the sample,
The light emitted from the light source is
making a plurality of lights separated by the optical element array and having different angles of incidence with respect to the sample,
The image sensor for autofocus is
Acquiring a plurality of optical images formed by reflected light from the sample,
The control unit
The distance between the plurality of light images is calculated, the difference between the light image lattice distance reference value which is a preset reference value for focusing is acquired, and a command signal for movement operation is output to the distance adjustment mechanism. to do
An autofocus device characterized by: By moving the optical element arranged between the optical element array and the autofocus imaging element in the optical axis direction of the light emitted from the light source,
or
By changing the grid spacing reference value,
changing or correcting the focal position in the direction of the optical axis;
The autofocus device according to claim 1, characterized by: Arranging a polarizer and a wave plate between an optical element for extracting or separating autofocus light from an observation optical system and an autofocus imaging device;
3. The autofocus device according to claim 1 or 2, characterized by: Having the autofocus device according to claim 1 or claim 2,
A microscope characterized by

Images