JPH0787378A - Focusing detection device - Google Patents

Abstract

PURPOSE:To provide a focusing detection device which can obtain the deviation quantity/direction of a measured object with simple constitution by detecting a focusing state based on a position where contrast becomes a maximum in a signal corresponding to a pattern image. CONSTITUTION:A contrast detection part 20 extracts the signal of a horizontal scanning line crossing a direction inclined to an optical axis among picture signals from a camera part 10, obtains the peak position and a bottom position and obtains the position of the pattern image whose contrast is the largest. A control arithmetic part 100 obtains the deviation between the contrast maximum position and a pattern center and displays the size and the direction of the deviation between a prescribed position and the position of the measured object. A measured part shift part 40 is controlled and the measured object is shifted to the prescribed position. A pattern part consists of periodical repetitive patterns, and the detection part 20 assumes an approximate curve passing the plural peaks of a detection signal. It is preferable that the peak position of the approximate curve becomes the contrast maximum value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detection device.

[0002]

2. Description of the Related Art In an autofocus mechanism, it is important to detect the focus accurately and quickly. Conventionally, various focus detection methods have been proposed, and typical methods include a method using a light spot and a method using image contrast.

[0003]

In the former method using a light spot, since light is scattered by unevenness or steps on the surface of the work, focus detection is performed with high accuracy when the surface condition of the work is poor. It was difficult.

The latter method of utilizing image contrast is a method of projecting a stripe pattern on the surface of a work and detecting the contrast of the image with a photodiode array or a CCD camera. In this method, for example, a signal relating to a projected fringe pattern is captured while changing the distance between the microscope unit and the work, and the signal is processed to perform focus detection and focus operation. Therefore, there is a drawback that it takes a considerable amount of time to finish focusing.

In view of such a problem, the present invention has a simple structure and is capable of promptly and accurately obtaining the deviation amount and the direction of the object to be measured from a predetermined (focus) position. An object is to provide a focus detection device.

[0006]

According to the present invention, there is provided a pattern portion having a pattern in a direction inclined with respect to an optical axis, a projection optical system for projecting this pattern on an object, and an object on which the pattern is projected. A detection unit that detects the image of the object, and a focus state detection unit that detects the focus state based on the position of the detection signal of the detection unit that has the maximum contrast in the signal corresponding to the projected pattern image. The gist is a focus detection device characterized by having:

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a conceptual diagram showing an optical system in an embodiment of the present invention.

The light emitted from the light source 1 once passes through the collimator lens 2 to become a parallel light beam, and illuminates the pattern 3.

The pattern 3 is in the form of a flat plate, and as shown in FIG. 2, is formed by repeating bright and dark portions having the same width. The pattern 3 is perpendicular to the optical axis so that the images of the end 3a on the object side and the end 3b on the light source side, which are formed by the objective lens 7, deviate at least from the depth of focus of the objective lens 7. It is necessary to incline at an angle θ. Further, the repeating pitch p of the bright and dark portions of the pattern 3 satisfies the relation of dz ≧ M * p * sin θ, where M is the projection magnification of the objective lens 7 and dz is the required alignment accuracy. It has been decided.

The pattern 3 is arranged so that when the object M to be measured is arranged at a predetermined position, the central portion of the pattern is formed on the surface of the object M to be measured as a sharp image with focus.

The light passing through the pattern 3 is reflected by the half mirror 6, and thereafter passes through the objective lens 7 to form an image 3'of the pattern 3 in the vicinity of the object K to be measured. Both ends of the pattern 3, that is, the end 3a on the DUT side and the end 3b on the light source side are connected as one end 3a 'and the other end 3b' of the image 3 ', as shown in FIG. Image.

The reflected light from the object K to be measured is the objective lens 7
The image is once formed by and then guided to the camera unit 10 via the relay lens 8.

The camera unit 10 can be composed of an image pickup tube, a CCD camera or the like. From a signal processing perspective,
It is preferable that the repeating direction of the pattern 3, that is, the direction inclined with respect to the optical axis, is made to coincide with the horizontal scanning direction of the camera unit. The camera unit 10 is connected to the extreme value detection unit 20.

The extreme value detecting section 20 extracts a signal of a horizontal scanning line which crosses a direction inclined with respect to the optical axis from the image signal from the camera section 10 and obtains its peak position and bottom position. Then, the contrast is detected and the position of the pattern image with the highest contrast is obtained.

Next, how to obtain the magnitude of the focus shift and its direction will be described.

The control calculation unit 100 receives the signal from the contrast detection unit 20, and regards the peak position where the ratio (or difference) between the peak level and the bottom level adjacent to each other is the largest as the contrast maximum position.

Then, a deviation d between the maximum contrast position and the pattern center is obtained, and this deviation d is converted into a difference between the position of the measured object at that time and a predetermined position.
For that purpose, the relationship between the two may be obtained in advance, and conversion may be performed based on this correspondence relationship.

Further, which side of the predetermined position the object to be measured is displaced is determined as follows. That is, when the contrast on the light source side of the pattern center is the maximum (clear image), it is determined that the object is located in front of the predetermined position. On the other hand, when the contrast of the pattern portion closer to the object than the center of the pattern has the maximum contrast (clear image), it is determined that the object is located behind the predetermined position.

When the object is at a predetermined position, for example, the signal 11 shown in FIG. 4 is obtained. Further, when the object is on the front side of the predetermined position, for example, the signal 12 of FIG. 5 is obtained. On the contrary, when the object is behind the predetermined position, for example, the signal 13 shown in FIG. 6 is obtained. The arrows in FIGS. 4 to 6 indicate the pattern center position in each signal.

As described above, the magnitude of the deviation between the predetermined position and the position of the object to be measured and the direction of the deviation (front side or rear side) are obtained and displayed on the display unit 30.

Further, the control calculation section 100 controls the measured section moving section 40 to move the measured object to a predetermined position based on the magnitude and direction of the deviation obtained as described above. The measured portion moving unit 40 is configured as, for example, a stage for moving the measured object K in the optical axis direction.

When the contrast data is processed by the above-described method in the control calculation unit 100, only a resolution of about one pitch of the projected pattern 3 can be obtained, but the interpolation is performed according to the flowchart shown in FIG. By performing the above, the maximum contrast position can be obtained with higher accuracy.

First, in S1, a peak is extracted from the output of the extreme value detecting section, and in S2, an approximate curve passing through each extracted peak is obtained. Here, as a method for obtaining an approximate curve, there are 1) a spline interpolation method, 2) a linear approximation method, and the like. Both methods are well known methods and will be briefly described.

1) Spline interpolation method In the spline interpolation method, a given section is divided into several subsections, and a polynomial (often used cubic polynomial) is applied to each subsection. However, as a whole, it is determined so that the points pass through the given points and the differential coefficients of the first and second floors coincide with each other.

2) Method based on linear approximation This is a method in which a slope that rises to the right or a slope that rises to the left is linearly approximated, and the intersection point of the approximate straight line is taken as the peak.

In this way, after the approximate curve is obtained in S2, the peak position is determined based on this approximate curve in S3, and this is set as the maximum contrast position.

Next, in S4, the deviation amount and the deviation direction between the position of the maximum contrast obtained in S3 and the pattern center position are obtained, and display or movement control is performed in S5. As shown in FIG. 3, it is displayed on the display unit 30 or the measured object moving unit 40 is driven to move the measured object to a predetermined position. Finally, in S5, the continuation / end of detection is determined.

By performing the interpolation as described above,
It is possible to obtain a finer resolution than one pitch (the length in the optical axis direction) of the projected pattern.

The resolution can be improved by bringing the angle formed by the optical axis and the pattern 3 close to 90 degrees, but in that case, rough focusing becomes somewhat difficult.

[0030]

The focus detection device of the present invention has a pattern portion having a pattern in a direction inclined with respect to the optical axis, a projection optical system for projecting this pattern onto an object, and a pattern projected. A detection unit that detects the image of the object, and a focus state detection that detects the focus state based on the position of the detection signal of the detection unit that has the maximum contrast in the signal corresponding to the projected pattern image. It is possible to obtain the amount and direction of the deviation of the measured object from the predetermined position with a simple configuration.

By performing the interpolation as in the second aspect, the amount and direction of the deviation of the object to be measured from the predetermined position can be obtained with accuracy equal to or lower than the pitch of the projected pattern.

[Brief description of drawings]

FIG. 1 is a conceptual diagram showing an optical system in an example of the present invention.

FIG. 2 is a plan view showing the pattern of FIG.

FIG. 3 is a block diagram showing a signal flow in the embodiment.

FIG. 4 is a diagram showing signals obtained when the object is at a predetermined position.

FIG. 5 is a diagram showing a signal obtained when the object is on the front side of a predetermined position.

FIG. 6 is a diagram showing signals obtained when the object is behind a predetermined position.

FIG. 7 is a flowchart for performing interpolation when obtaining the maximum contrast position.

[Explanation of symbols]

 1 light source 2 collimator lens 3 pattern 6 half mirror 7 objective lens 8 relay lens 10 camera section 20 extreme value (contrast) detection section 30 display section 40 measured section moving section 100 control calculation section

Claims (2)

[Claims] 1. A pattern portion having a pattern in a direction inclined with respect to an optical axis, a projection optical system for projecting the pattern on an object, and detection for detecting an image of the object on which the pattern is projected. And a focus state detection unit that detects a focus state based on the position of the detection signal of the detection unit that has the maximum contrast in the signal corresponding to the projected pattern image. Focus detection device. 2. The focus detection device according to claim 1, wherein the pattern portion is formed of a periodic light and dark repeating pattern, and the focus state detection portion is an approximate curve passing through a plurality of peaks of a detection signal. Assuming that the peak position of the approximated curve is the maximum contrast position, the focus detection apparatus is characterized.