JPS63138310A - Focus detector - Google Patents

Abstract

PURPOSE:To realize high-accuracy automatic focusing without generating any false focus error signal by the pattern of an object surface by setting an index projection surface where an index image is formed at a position which is not conjugate to an object projection surface about an object lens. CONSTITUTION:The cause of the influence of the pattern of the object surface 3 on the index image 5 on a photodetecting means is that the object surface 3 and index light projection surface 6 are set conjugate to each other about the objective 2, so this is utilized to set the optical arrangement between the object surface 3 and index projection surface 6 not to satisfy the conjugation relation about the objective 2. Consequently, even if the index light incident on the index projection surface 6 is under the influence of the object surface pattern, the pattern lighted by the index light is not image-formed sharply on the index projection surface 6, but formed in an out-of-focus state, so the light quantity distribution of the index image on the index projection surface 6 is averaged and the superposition of an error signal on the output of the photodetecting means is suppressed.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a focus detection device used in a microscope or the like.

(Prior Art) Conventionally, this type of device is disclosed in US Pat. No. 3,721,827.
No. 60-118814 is known.

These devices are equipped with an illumination device that irradiates illumination light onto an object surface using epi-illumination, and an observation device that images the object light from the illuminated object surface as an object image on the object projection surface using an objective lens. Roughly, the configuration for focusing the object image on the object projection plane includes an index irradiation means that irradiates index light such as a slit light having a wavelength different from that of the illumination light toward the object surface, and an index light that is reflected from the object surface and an index projection means for forming an index light that has passed through the lens on an index projection plane as an index image;
a defining means for determining that the index light that has passed through one side of the pupil of the objective lens is focused on the target projection plane in order to move the target image on the target projection plane according to the distance between the object plane and the objective lens; The apparatus includes a light receiving means disposed on the projection plane and generating an output corresponding to the imaging position of the index image, and an interval control means controlling the distance between the object plane and the objective lens based on the output of the light receiving means.

(Problems to be Solved by the Invention) However, in such conventional focus detection devices, when the reflectance of the object surface is not uniform, for example when a pattern is formed, The irradiated index light becomes the illumination light for the pattern, and an image of the pattern is formed on the index projection surface, so the index image on the index projection surface is a distorted one with a positionally different light intensity distribution. became. Therefore, the output obtained from the light receiving means includes an error signal due to the influence of the pattern, and there is a problem that accurate focus adjustment is impossible.

(Means for Solving the Problems) The present invention has been made in view of these conventional problems, and provides a focus detection device that can reliably avoid the generation of error signals due to patterns on the object surface. It is about providing.

In order to achieve this object, the present invention solves the following problems:
Focusing on the fact that the object plane and the index light projection plane were set in a conjugate relationship with respect to the objective lens, we changed the optical arrangement of the object plane and the index projection plane so that they did not satisfy the conjugate relationship with respect to the objective lens. This is the setting.

(Production) According to the configuration of the present invention, even if the index light incident on the index projection surface receives the [1] of the object plane pattern, the pattern illuminated by the index light is clearly visible on the index projection surface. The light intensity distribution of the target image on the target projection plane is averaged, and the error signal due to the pattern is suppressed from being superimposed on the output of the light receiving means. .

(Example) FIG. 1 is an explanatory diagram showing the main configuration of the present invention.

In the figure, objective lenses 2a and 2b are the same lens 2 developed with the object surface 3 as a mirror surface.

Object light a from the object plane 3 illuminated by an illumination means (not shown) using a method such as epi-illumination or transmitted illumination is focused on the object projection plane 4 as an object image.

On the other hand, the focus detection slit 5 as an index
is placed on the non-conjugate surface 5 via the objective lens 2a. Therefore, the index light beam from the slit 5 forms an aerial image as an index image on a plane 6 different from the object plane 3 via the objective lens 2a. Therefore, the finger mass projection surface 7 on which the index image is formed is at a different position from the object projection surface 4, and the object surface 3 illuminated by the index beam becomes a blurred image on the index projection surface 7. Therefore, even if a pattern with different reflectance (transmittance) depending on the position is formed on the object plane 3, an index image with an averaged light distribution is formed on the index projection plane 7. Superimposition of error signals due to the pattern on the output of the light receiving means installed on the index projection surface 7 is suppressed.

FIG. 2 is an explanatory diagram showing an embodiment of the present invention in an epi-illumination microscope based on the basic concept of the present invention shown in FIG.

First, to explain the configuration, numeral 15 is a light source that generates illumination light in a wavelength range including visible light and invisible light (for example, infrared light). The light passes through an aperture stop 17, a relay lens 18, and a field stop 14 to reach a half mirror (semi-transmitting mirror) 12, where it is reflected in the right angle direction and illuminates the object plane 3 through the objective lens 2. Object light a from the object plane 3 illuminated by the light source 15 is reflected by the objective lens 2 as shown by the solid line.
and is imaged onto the object projection plane 4 through the half mirror 12.

Here, a portion consisting of a light source 15, a condensing lens 16, an aperture diaphragm 17, a relay lens 18, a field diaphragm 14, and a half mirror 12 constitutes an illumination optical system Δ. The objective lens 2 for observing the object and the object projection surface 4 constitute an observation optical system.

On the other hand, as an optical system for focus detection using invisible light (infrared light) from the light source 15, a focus detection slit 5 as an index is first provided. This focus detection slit 5
has a slit opening with a long side formed in a direction perpendicular to the optical axis and the plane of the paper, the opening transmits invisible light as an index light from the light source 15, and the part other than the slit opening transmits invisible light. Block out light. Further, the focus detection slit 5 completely transmits visible light.

Therefore, the half mirror 12 is irradiated with a slit beam of index light consisting of invisible light rays shown by broken lines from the focus detection lenses 1 to 5. The index light indicated by the broken line illuminated on the half mirror 12 is reflected in the orthogonal direction, and is reflected by the objective lens 2.
is formed as an aerial image on a plane 6 different from the object plane 3. Thereafter, the light is reflected by the object surface 3 and is extracted by a dichroic mirror 11 provided in the middle of the observation optical system, which reflects only the invisible light and transmits the visible light, and is branched to the focus detection system B.

The focus detection system B is provided with a filter 17 as a defining means at the re-imaging position of the pupil of the objective lens 2, and a cylindrical lens 8 behind it. Filter 17
Of the index light extracted by the dichroic mirror 11, only the light that has passed through the left pupil (hatched area) of the objective lens 2 is allowed to pass through. Cylindrical lens 8
compresses the index light obtained through the filter 16 in the longitudinal direction of the slit (direction perpendicular to the plane of the paper).

The focus detection system A further includes a light receiving member 9 placed on the target projection surface 7 on which the target beam obtained from the cylindrical lens 8 is formed as a slit-shaped target image compressed in the longitudinal direction. ing. The imaging position of the index image on the light receiving member 9 becomes a position corresponding to the distance between the objective lens 2 and the object plane 3 due to the action of the filter 17. The light receiving member 9 generates a light receiving output according to the imaging position of the index image. The calculation means 19 receives the light reception output of the light receiving member 9, calculates a focus error signal according to the imaging position of the index image, and supplies it to the drive means 20, and the objective lens 2 or The object plane 3 is driven in the optical axis direction.

FIG. 2a shows the optical path state of the index light in FIG. 2 with diagonal lines.

By arranging the optical system for focus detection using invisible light for the epi-illumination optical system and observation optical system in such an epi-illumination microscope, it is possible to detect a focus error signal at the imaging position of the target image using invisible light. the light-receiving member 9 and the object plane 3
are not in a conjugate relationship, so even if the index image formed on the light-receiving member 9 includes a pattern due to a difference in reflectance on the object surface 3, the pattern on the object surface 3 will be a blurred image. The distribution of the light of the index image is averaged, so that a false focus error signal due to the pattern of the object plane 3 does not occur.

Next, the role of the cylindrical lens provided in the focus detection system B of FIG. 2 will be explained.

FIG. 3(a) is a perspective view showing the index image 22 created by the objective lens 2 when there is no cylindrical lens.

FIG. 3(b) shows the index image 22 when the cylindrical lens 8 is provided, and this index image 22 is shown in FIG. 3(a).
) as seen from the arrow X direction. That is, in FIG. 3(b), since the cylindrical lens 8 has no power for the short side of the index image 22,
Edges el and e2 in the long side direction of the index image 22 are formed on the light-receiving surface of the light-receiving member 9, that is, the index image projection surface 9a.

FIG. 3(C) is a diagram of the index image 22 viewed from the X direction in FIG. 3(a>) with the cylindrical lens 8 installed.
When there is no cylindrical lens 8, the luminous flux f(
An index image 22 is created as shown. On the other hand, when the cylindrical lens 8 is provided as shown, an index image whose size in the long side direction is compressed is formed at a position M23 in front of the light receiving surface 9a of the light receiving member 9. .

Therefore, it becomes a blurred index image on the light-receiving surface 9a, and serves to average out the uneven brightness in the X direction in FIG. 3(a) k.

That is, in the focus detection system B of FIG. 2, by making the light-receiving member 9 and the object surface 3 in a non-conjugate relationship, it is possible to average out the unevenness in the amount of light given to the index image by patterns of different reflectances on the object surface. In addition, the cylindrical lens 8 compresses only the long side direction of the target image to form a blurred image on the light receiving means in the long side direction, thereby further reducing uneven brightness of the target image in the long side direction. This is done to prevent false focus error signals from being generated due to patterns of different reflectances on the object surface.

Further, the index image 22 is displaced in accordance with the change in the distance between the objective lens 2 and the object plane 3 in the V direction orthogonal to the X direction,
The index image 22 has a clear edge e without being blurred in the X direction.
Since L e 2 is formed, focus detection accuracy does not deteriorate.

FIG. 4 is an explanatory diagram showing a specific embodiment of the focus detection system B shown in FIG. 2.

In FIG. 4, the target image 24 formed by the objective lens 2 shown in FIG. 2 is re-imaged on the light receiving surface of the light receiving member 9 by the relay lenses 25 and 26. In FIG.

At this time, in the relay lenses 25 and 26, a filter 1 for pupil division is provided on a plane conjugate with the pupil of the objective lens 2 shown in FIG.
7 to cover most of the moving range of the index image.

This filter 17 has one surface 17 of a boundary line 28 having a diameter passing through the optical axis 27 as shown in FIG. 5(a), for example.
The index light consisting of invisible light is transmitted through the surface 17a, and the other surface 17b
It has a structure that blocks light of all wavelengths, including invisible light.

Further, as shown in FIG. 5(b), the filter 16 has one surface 17a of the boundary line 28 located off the optical axis 27.
It is also possible to adopt a structure in which the index light is transmitted through the optical axis 17b and all wavelength light is blocked at the other surface 17b.Furthermore, as shown in FIG. A structure may be used in which the light is transmitted and the remaining portion 17b blocks all wavelength light. In either case, the boundary line 28 needs to have a directional component perpendicular to the direction of displacement of the index image 22.

FIG. 6 shows the imaging state of the index image on the light-receiving surface 9a of the light-receiving member 9, the intensity distribution 30 of the index image, and the position of its center of gravity when using the filter 17 having the structure shown in FIG. 5(a). FIG. 6(b) shows a focused state, and FIG. 6(a) shows a front focus state shifted from the focused state to an imaging position forward in the optical axis direction. 6(C)
indicates a state in which the image is out of focus after being shifted from the in-focus state to an imaging position rearward in the optical axis direction.

As is clear from the intensity distribution 30 based on the imaging state of the index image on the light-receiving member 9 in FIG.
When using D, the settings are such that the received light outputs obtained from the two output terminals of the PSD are equal in the focused state shown in Fig. 6(b), and the front focus shown in Fig. 6(a) is set to be equal. In the case of the rear pin of FIG. 6(C), the light receiving output of one of the two output terminals in the PSD becomes larger than the other, and in the case of the rear pin of FIG. becomes larger.

In the calculation means 19 shown in FIG. 2, a focus error signal can be obtained from the deviation of the two light receiving outputs of the light receiving member 9 made of PSD. Of course, as the light-receiving member 9, in addition to the PSD, a -dimensional COD line sensor layer two-split light-receiving element, etc. can also be used.

Furthermore, since the focus detection system B shown in FIG. 4 is provided with the cylindrical lens 8 shown in detail in FIG. 3 following the relay lens 26, the focus detection system B shown in FIG. By using the structured filter 17, the index image whose half is cut around the optical axis is transmitted to the light receiving member 9 by the cylindrical lens 8 as shown in FIG. 3(a).
The image is formed in a compressed state in the long side direction at a position different from a, and inevitably becomes a blurred image on the light receiving section 9a.As a result, the index image in the long side direction due to the pattern due to the difference in the reflectance of the object surface The unevenness is averaged out, and the intensity distribution of the index image with respect to the light receiving member 9 becomes a uniform intensity distribution that is not affected by the pattern on the object plane, as shown in FIG.

FIG. 7 is an explanatory diagram showing another embodiment of the illumination system provided in the embodiment of FIG. 2, and this embodiment is characterized by separately providing a light source for forming an index image using invisible light. shall be.

In FIG. 7, the illumination optical system includes a light source 15, a condenser lens 16, an aperture stop 17, a relay lens 18, and a field stop 14, similar to the embodiment shown in FIG. A light source 32 for forming a single finger image, which generates invisible light such as infrared light, is provided at a position perpendicular to the optical axis of this illumination optical system. As the light source 32, an appropriate light source such as a lamp, LED, laser, etc. can be used. Invisible light from the light source 32 passes through a condensing lens 33 and a relay lens 34, further passes through a slit 5 placed at a position corresponding to the plane 5 shown in FIG. 1, and is further provided in the optical path of the illumination optical system. It is reflected by the guichroic mirror or half mirror 35 and introduced into the light beam of the illumination optical system.

According to the embodiment shown in FIG. 7, visible light for epi-illumination is completely transmitted through the focus detection slit 5 as one finger, and invisible light for focus detection is transmitted through the slit opening as an index. The purpose of the focus detection slit 5 can be achieved by simply using a light shielding plate provided with a slit opening, without requiring any special filter.

FIG. 8 is an explanatory diagram showing another embodiment of the focus detection system B in FIG. 2. In this embodiment, the pupil division by the filter 17 shown in FIG. 4 is performed by pipe rhythm. It is characterized by the following.

In FIG. 8, the index light beam from the objective lens 2 is used as a defining means placed at a position conjugate with the pupil of the objective lens 2 between the relay lenses 25 and 26 after creating the aerial image 24 of the finger. is incident on the pipe rhythm 36. The pipe rhythm 36
consists of a pair of wedge prisms whose polarization directions are opposite to each other. The pipe rhythm 36 divides the incident index light beam into first and second index light beams bl and b2. Rain analysis b 1. b 2 is each cylindrical lens 8
After passing through the first and second light receiving members 9A, respectively.

An index image is projected onto 9B.

FIG. 9 shows the index light α (shown by a solid line) at the time of focusing, the index light β (shown by the broken line) at the time of defocusing, and the light receiving members 9A and 9B when a pipe rhythm 36 is provided in the focus detection system. Intensity distribution of index image by rain analysis α and β 31 (solid line diagram)
, 32 (indicated by broken lines) and the center of gravity position 33 of the intensity distribution of the index light α.

First, during defocusing, the index images on the light-receiving members 9A and 9B become blurred by the intensity, as shown by 32, and simultaneously move in the opposite direction, and the sum of this amount of movement becomes the focus error signal. Therefore, even when the pipe rhythm 36 is used, it is possible to realize movement according to the out-of-focus of the target image due to pupil division, similar to the case where the light beam is cut by the filter 17 shown in FIG.

Next, the advantages of performing pupil division using the pipe rhythm 36 will be explained.

FIG. 10 is an explanatory diagram showing the index light α when the object surface to be observed is not tilted, and the index light β when the object surface is tilted.

That is, since the imaging plane 24 of the index image and the light receiving surfaces of the light receiving members 9A and 9B are not conjugate with the object plane, the intensity distribution 32 of the index image when the object plane is tilted will be the same as the intensity distribution 32 of the index image when in focus. It moves in the same direction with respect to the intensity distribution 31 of the index image.

Therefore, the movement of the index image in the same direction due to the tilt of the object plane, that is, the tilt of the sample to be observed, can be clearly distinguished from the movement of the index image in the opposite direction due to focus error. Therefore, the focus detection system using the pipe rhythm 36 allows detection of a focus error signal that is not affected by the tilt of the sample.

Furthermore, when the pipe rhythm 36 is used, it is possible to separate the influence of patterns having different fouling rates on the object plane from the focus error signal.

FIG. 11 is an explanatory diagram showing the intensity distribution of the index image in the light receiving members 9A and 9B in the focus detection system using the pipe rhythm 36 shown in FIG.

First, the intensity distribution 37 shown by the broken line is the case where there is no influence of the object plane pattern in the index image, and its center of gravity is located at the arrow 3.
It will be in position 8. On the other hand, if the index image is affected by the pattern on the object surface, the reflectance differs depending on the position, so for example, an index image with an intensity distribution 39 shown by a solid line is obtained, and the center of gravity is at the position indicated by the arrow 40. Move as shown. The movement of the center of gravity position due to the index image including the object plane pattern is in the same direction for the light receiving members 9A and 9B, and can be distinguished from the movement of the index image in the opposite direction due to a focus error.

Next, based on the light reception outputs obtained from the light receiving members 9A and 9B of the focus detection system using the pipe rhythm 36, the movement of the index image due to the focus error, and the index image due to the influence of the tilt of the object surface and the pattern of the object surface. An embodiment for performing focus detection and focusing drive by separating the movement of the lens, that is, an embodiment of the calculating means 19 and driving means 20 shown in FIG. 2 will be described with reference to FIG. 12.

In FIG. 12, 9A and 9B are light-receiving members on which an index image is formed by two index lights α and β divided by the pipe rhythm 36, and in this embodiment, the light-receiving position on the light-receiving surface is Let us take as an example the case of using a PSD that produces two light receiving outputs that change in a complementary manner.

The two light receiving outputs obtained from the light receiving member 9A using the PSD are amplified by preamplifiers 42 and 43, and the two light receiving outputs are added and subtracted in an adding circuit 46 and a subtracting circuit 48, respectively, and the results are given to a dividing circuit 50. Information on where the center of gravity of the light intensity distribution of the light that has passed through the pipe rhythm 36 is located in the PSD is output from the dividing circuit 50 to the DC voltage V.
Output as 1. The output of this divider circuit 50 has a waveform generally called an S curve.

Similarly, for the light-receiving member 9B using PSD, the two light-receiving outputs are amplified by the preamplifiers 44 and 45, added and subtracted by the adder circuit 47 and the subtracter circuit 49, and then fed to the divider circuit 51. Direct current iV provides information on where the center of gravity of the light intensity distribution of light passing through the pipe rhythm is located.
Output as 2.

Here, the output signal V1. of the divider circuit 50.51. V2 can be expressed as follows, △V represents an in-phase error signal generated by the influence of the inclination of the object plane and the pattern of the object plane, and vl and v2 are approximately proportional to the amount of movement of the index image due to focus error. voltage.

The outputs of the division circuits 50 and 51 are input to the subtraction circuit 52, and Vl-V2=(Vl+Δv)-(v2+ΔV)=v
l −v2 (2) By performing the calculation, the influence of the inclination of the object surface and the pattern of the object surface can be removed.

Here, in the subtraction of equation (2) performed by the subtraction circuit 52, a voltage v1. Since v2 are signals with different signs and equal absolute values, if vl=-v2=V, the subtraction circuit 52 will actually output 2v. Therefore, the subtraction circuit 52 can eliminate the influence of the inclination of the object plane and the pattern of the object plane, and at the same time can improve the sensitivity of the focus error signal itself.

The focus error detection signal obtained from the subtraction circuit 52 in this way is zero at the in-focus position, and at other states.
For example, if the front pin side is a positive voltage, the rear pin side is a negative voltage, and the magnitude of this voltage is approximately proportional to the distance from the in-focus position.

Therefore, the focus error detection signal obtained from the subtraction circuit 52 is applied to the addition circuit 55 as a position control signal for driving the stage to the in-focus position via a buffer 153 and a low-pass filter 54 that removes the 9tE sound component.

The interference circuit 55 constitutes the control loop of the stage drive system, and this stage drive loop is a speed control circuit consisting of a motor drive circuit 57, a stage drive motor 60, a tacho generator 59, a preamplifier 58, and a low-pass filter 56. It constitutes a loop, and a low pass filter 5
Drive control of the stage 61 to the focused position is performed by driving the stage drive motor 60 using the focus error signal from 4, that is, the position control signal, as a control target, and the tilt of the object to be inspected and the pattern shape of the object surface are determined in real time. It is possible to achieve highly accurate autofocus that is not affected by

In the above embodiment, as shown in FIG. 1, by passing through the objective lens 2a, that is, by passing through the objective lens 2 for the first time, the index light beam is connected to the objective lens 2a and the object plane 3. forming an index image on an imaging plane 6 located in the space between the
Thereafter, the index image is re-formed on the index projection surface 7 by being reflected by the object surface 3 and then passing through the objective lens 2b, that is, by passing through the objective lens 2 a second time. It is not limited to. For example, as shown in FIG. 13, by passing the index light beam through the objective lenses 2a and 2b, the index beam is projected onto the finger projection surface 7 located in the space between the objective lens 2b and the object projection surface 4. An image may be formed. In this case, the imaging plane 6 will be located between the index projection plane 7 and the object projection plane 4, but the index image will not actually be formed.

Furthermore, in this embodiment, invisible light such as infrared light is used as the index light in order to distinguish it from the illumination light, but the present invention is not limited to this. For example, a known method of modulating the index light may be employed. Furthermore, even if the wavelength spectra of the illumination light and the index light are the same, if the light intensity of the latter is made somewhat higher than the former, the index light will be a blurred and uniform illumination at the object plane, so there is no problem. On the projection plane, a high-luminance index image is formed in the background of uniform illumination light, so it is possible to detect the imaging position of the index image.

Further, in this embodiment, the object plane is illuminated by epi-illumination, but the present invention is not limited to this, and transmitted illumination may be used.

(Effects of the Invention) As described above, according to the present invention, even if the index light made of invisible light is projected onto the object plane and focus detection is performed from the index image formed by the reflected light, the index projection plane Since the light-receiving surface of the light-receiving member and the object plane are not conjugate, patterns with different fouling rates on the object surface to be observed are formed as blurred images on the light-receiving member, resulting in false focus errors due to the pattern on the object surface. Highly accurate autofocus can be performed without generating a signal.

Note that, as an effect unique to the embodiment, a pipe rhythm is used as a pupil division to realize movement of the index image according to the focal shift on the light receiving surface of the light receiving member, and the index light is divided into two by the pipe rhythm. By forming index images on the respective light-receiving members, the tilt of the object surface and the tilt of the object surface are determined based on the differences in the moving direction of the index image and the tilt of the object surface due to focus errors, and the difference in the moving direction of the index image due to the pattern of the object surface. It is possible to obtain a focus error signal from which the influence of the pattern has been removed, and thereby it is possible to perform highly accurate autofocus without being influenced by the inclination of the object plane or the pattern on the object plane.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram showing the basic concept of the present invention, Fig. 2 is an explanatory diagram showing an embodiment of the present invention in an epi-illumination microscope, and Fig. 2a is a focus based on the index sales target provided in Fig. 2. An explanatory diagram showing the optical path for detection, FIG. 3 is an explanatory diagram showing the function of the cylindrical lens provided in the focus detection system of FIG. 2,
Fig. 4 is an explanatory diagram showing a specific embodiment of the focus detection system in Fig. 2, Fig. 5 is an explanatory diagram showing the pupil division filter shown in Fig. An explanatory diagram showing the movement and intensity distribution of the index image due to a partial cut, Fig. 7 is an explanatory diagram showing another embodiment of the illumination optical system according to the present invention, and Fig. 8 is an explanatory diagram showing another embodiment of the illumination optical system according to the present invention. An explanatory diagram showing pupil division, Fig. 9 is an explanatory diagram showing the formation and intensity distribution of the index image when in focus and out of focus by pipe rhythm, and Fig. 10 is an explanatory diagram showing the index when the object plane is tilted. An explanatory diagram showing the image movement and intensity distribution. Figure 11 is an explanatory diagram showing the index image and its center of gravity movement based on the object plane pattern obtained by pupil division using pipe rhythm. FIG. 13 is an explanatory diagram showing another basic configuration of the present invention. 1.2: Objective lens 3: Object plane 4: Image plane 5: Focus detection slit 6: Projection plane of index light 7: Image formation plane of index image 8 Nijilintrical lens 9.9Δ, 9B: Light receiving member 11 : Dichroic mirror (transmits visible light, reflects invisible light) 12: Half mirror 14: Field diaphragm 15: Light source 17: Aperture diaphragm 16. 33: Condensing lens 18. 25, 26, 24 relay lens 19: Calculating means 20: Driving means 22: Index image 23 Index image 24 of the lindrical lens: Index image 27: Optical axis 28: Boundary line 32: Index image forming light source 36: Pipe rhythm 42. 43, 44, 45, 53. 58: Preamplifier 4
6.47.55: Addition circuit 48.49, 52: Subtraction circuit 50.5 and division circuit 54.56: Low-pass filter 57: Motor drive circuit 59: Tacho generator 60: Stage drive motor 61: Inspection stage

Claims (2)

[Claims] (1) Illumination means for irradiating illumination light onto an object surface; Observation means having an objective lens for forming an object image on an object projection plane using object light from the object surface illuminated by the illumination means; an index irradiation means for irradiating an index light toward the surface; an index projection means for forming the index light reflected by the object surface and passed through the objective lens on an index projection plane as an index image; In order to move the index image on the index projection plane according to the distance from the objective lens, the index light that has mainly passed through one side of the pupil of the objective lens is formed as the index image on the index projection plane. a light receiving means disposed on the target projection surface to receive the target image and generate an output corresponding to the imaging position of the target image; A focus detection device comprising: an interval control means for controlling an interval between an object plane and the objective lens to focus an image of the object plane on the object projection plane, wherein the objective lens is non-conjugate with the object projection plane. A focus detection device characterized in that the index projection plane on which the index image is formed is set at a position, and the light receiving means is disposed there. (2) The defining means is a light splitter (36) disposed at a position conjugate with the pupil position of the objective lens, and the light splitter passes the index light through one side of the pupil of the objective lens. the first index light passing through the other side, and the second index light passing through the other side; images are respectively formed on the second index plane as second index images, and the light receiving means is arranged on the first index projection plane and generates a first output corresponding to the imaging position of the first index image.
a second light receiver disposed on the second target projection plane and generating a second output corresponding to the imaging position of the second target image; The focus detection device according to claim 1, further comprising a subtraction circuit (52) that performs subtraction with the second output.