JP3405891B2 - Focusing position calculating apparatus and method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an in-focus position calculating apparatus and method for a scanning electron microscope.

[0002]

2. Description of the Related Art Currently, a scanning electron microscope (Scanning Ele
ctron Microscope: hereinafter referred to as SEM) has a resolution of n
It reaches the m (nanometer) range and is widely used in semiconductor inspection and the like.

In this SEM, it is extremely important to speed up the autofocus performed every inspection, and several speeding up methods have been studied.

In general, autofocus consists of the following two steps.

(1) The degree of focus of an image obtained at each focus (hereinafter referred to as sharpness) is evaluated.

(2) A peak is detected from the change in sharpness with respect to focus, and the peak position is set as the in-focus position.

A clear image contains many high frequency components. Therefore, the strength of the high frequency component of the image may be used as the definition function of the sharpness.

[0008]

By the way, since it takes a very long time to input a full-size image for all stages of focusing, as a method for speeding up autofocus, a method of inputting only a part of the image or Kaihei 7-1
No. 53407), there is a method of predicting the in-focus position and densely inputting steps only near the in-focus position (Japanese Patent Laid-Open No. 56-4).
1663).

In the method of inputting only a part of the image, that is, in the former method, if the density gradient of the input partial image is small, the sharpness change with respect to the focus does not have a clear peak. There is a problem that the accuracy of is very poor.

This is because when the density gradient is small, the image hardly changes even if the focus changes.

For example, FIG. 1A shows the change in sharpness with respect to focus when the density gradient in the partial image is relatively large, and FIG. 1B shows the change in sharpness with respect to focus when the density gradient is small. For example:

In the former method, in order to avoid this problem, an image for one screen is first input, a region with a large density gradient is extracted from the image, and then the focus is changed only for that region for input. . However, since the density gradient in the noise region also becomes large, there is a risk that the noise region is erroneously input and the incorrect focusing position is output, and the processing time is required for the feature extraction time.

Further, in the latter method of predicting the in-focus position and densely inputting a stage close to the in-focus position, an image is input with several focuses and the sharpness of the image at each focus is obtained.

For example, suppose that the sharpness at five focus positions is obtained as a black circle as shown in FIG. Here, assuming that the in-focus position exists between f1 and f2, f
Only the image for the focus between 1 and f2 is input, and the sharpness of the image for the focus between f1 and f2 is evaluated as indicated by the black circles in FIG. The above processing is repeated to obtain the peak position while predicting it. However, the peak positions do not always exist in f1 to f2. For example, when a true peak position exists outside f1 to f2, there is a problem that an incorrect focus position is output.

The present invention has been made in view of the above circumstances, and relates to a focus position calculating method and apparatus for performing focus adjustment at higher speed and with higher accuracy than ever before.

[0016]

According to a first aspect of the present invention, a sample is scanned with an electron beam, secondary electrons generated by the scanning are detected, and the detected amount of the detected secondary electrons is used as contrast information. in the scanning electron microscope conversion and displays an enlarged image of the specimen, the electron beam control means for are scanned interlaced at predetermined intervals in the vertical and horizontal directions of the electron beam to said sample, from said electron beam control means Focus control means for changing the focus of the electron beam on the sample, detection means for detecting secondary electrons generated from the sample by the electron beam, and the detection while changing the focus on the sample by the focus control means. Image of each focus from the detected amount of secondary electrons detected by the means.
Scan the system vertically and horizontally at predetermined intervals.
Image information calculation means for obtaining the image information of the reduced image which is the image obtained by the above, and the sharpness of the image information of the reduced image for each focus obtained by the image information calculation means. It is an in-focus position calculation device comprising image processing means for obtaining a focus corresponding to an in-focus position for an image.

Since the electron beam control means controls the electron beam to scan the surface of the sample at a preset interval, an image having a size smaller than usual is input as a reduced image, and the image input time is shortened. To be done. Further, the image processing performed in the process of obtaining the in-focus position is also performed on the reduced image, so that the processing time is shortened.

A second aspect of the present invention scans a sample with an electron beam, detects secondary electrons generated thereby, converts the detected amount of the detected secondary electrons into contrast information, and converts the detected amount of the secondary electrons. a scanning electron microscope to view the enlarged image, and the electron beam control step of scanning and interlaced at predetermined intervals the electron beam vertically and horizontally <br/> to said sample, results from the sample by the electron beam An image for each focus from the detection step of detecting secondary electrons and the detected amount of secondary electrons detected by the detection step while changing the focus on the sample.
Therefore, the electron beam is vertically and horizontally scattered at predetermined intervals.
An image information calculation step for obtaining image information of a reduced image that is an image obtained by overscanning, and a sharpness is obtained for each image information for each focus obtained in the image information calculation step, and the reduction is performed from these sharpnesses. A focus position calculating method comprising an image processing step of obtaining a focus corresponding to a focus position for an image.

[0019]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to FIGS.

FIG. 3 shows a schematic structure of the SEM in this embodiment.

Here, the cathode 1a and the anode 1b
A field emission type electron gun 1 for generating a high-intensity electron beam, a focusing lens 2 for focusing the electron beam on the sample surface, a stigmator (astigmatism correction device) 3 for correcting astigmatism, and an electron. Deflector 4 for scanning the beam on the sample surface, and objective lens 5 for focusing the electron beam on the sample surface.
A detector 6 for detecting secondary electrons emitted from the surface of the sample; a monitor 7 for displaying the signal detected by the detector 6 as an image; and an electron beam controller 8 for controlling the irradiation position of the electron beam. A focus control unit 9 that controls the focus of the objective lens 5 and an image processing unit 10 that processes an image generated from a signal of the detector 6 and calculates a focus position. Further, a portion surrounded by a dotted line in FIG. 3, that is, the electron gun 1, the focusing lens 2, the stigmator 3, the deflector 4, the objective lens 5, and the detector 6 are collectively included in the image input unit 1.
Call 1.

FIG. 4 shows a flowchart of autofocus. Note that, in FIG. 4 and the following, the image at the focus f is expressed as _If (f = 1, 2, ..., F).

The initial value of f is set to f = 1 (step 1).

Next, a reduced image I _f in the focus value f which is transmitted from the focus control unit 9 inputs the image input unit 11 (Step 2). This reduced image is at the position of the sample to be measured, and the enlargement ratio is adjusted to the size of the image to be measured.

The "reduced image" is an image obtained by interlaced scanning with an electron beam at preset intervals. FIG. 5 shows an explanatory diagram of the reduced image. In the figure,
Each cell corresponds to a pixel. A normal image is obtained by irradiating all the grid positions with an electron beam. In the example of FIG. 5, the 12 × 12 pixel image is a normal image. For the reduced image, the electron beam control unit 8 controls the irradiation position of the electron beam, and the electron beam is scanned at an interval larger than that during normal image input. For example, in the example of FIG. 5, an image of 3 × 3 pixels obtained by irradiating the electron beam only on the positions marked with “x” by skipping 3 pixels in both vertical and horizontal directions is a reduced image.

In "interlaced scanning", the same scanning as in the case of inputting a magnified image from the size of the image currently desired to be input to the sample is performed. That is, if the size of one side of the image of FIG. 5 is 12 nm (1 square is 1 nm), the interlaced scanning is performed every 4 nm. This scan has a side dimension of 1
This can be realized by performing the same operation as in the case of scanning an image of 48 nm which is four times as large as 2 nm (the enlargement ratio is 1/4) by the image input unit 11. However, even if the amount of information to be input is the same as when scanning an image with a side of 48 nm, the amount of information is very small because the information of an image with a side of 12 nm is input. In particular,
(1/4) ² = 1/16, which speeds up the processing.

The image processing unit 10 evaluates the sharpness E (f) of the reduced image _If . Generally, a clear image has a higher intensity of high frequency components than a blurred image. Therefore, the sum of all pixels of the output image of the high-frequency emphasis filter is defined as the sharpness. Here, a Sobel differential filter that calculates a local density gradient of an image is used as the high-frequency emphasis filter.
The weighting of the Sobel differential filter is shown in FIG.
I _x and I _y represent concentration gradients in the x and y directions, respectively. Gradient of a point on the image _{I f} (x, y) is in the focus f,

[Equation 1] The sharpness E (f) of the image _If is calculated as

[Equation 2] It can be obtained more (step 3).

However, w and h are the numbers of vertical and horizontal pixels of the image, respectively.

If f <F after calculating the sharpness E (f), an instruction to increase f by one step is transmitted to the focus control section 9 (step 4).

The focus control unit 9 increases the focus by 1 (step 5), inputs an image, and similarly calculates the sharpness.

These series of processes are repeated as long as f <F is satisfied, and the change in sharpness with respect to the focus as shown in FIG. 7 is obtained. This curve is called the sharpness curve.

When f ≧ F, the iterative calculation process is terminated, the peak position of the obtained sharpness curve is obtained, and the peak position f _peak is transmitted to the focus control section 9 (step 6). As a method for obtaining the peak position, here, the simplest method is to use f having the maximum sharpness as the peak position.

In the focus control unit 9, the image processing unit 10
The focus is adjusted to the position of f _peak transmitted from the image input unit 11, and the image input unit 11 inputs the focused image at the focus f _peak (step 7).

Here, the electron beam controller 8 controls the irradiation position of the electron beam so that the image is input with the closest sampling interval instead of the reduced image.

As described above, the autofocus function by image processing is realized.

The interval of the interlaced scanning of the electron beam control unit 8 is a constant interval set in advance in one image, but instead of this, the interval set in advance for each scan, or You may go at different intervals.

Also, step 1 of the flow chart of FIG.
The contents of steps 7 to 7 are stored in a recording medium such as a CD-ROM. And in the SEM that enables interlaced scanning,
The contents of this recording medium are transferred, and step 1 to step 7
May be performed.

In the present embodiment, the maximum sharpness f is the peak of the sharpness curve.
A method of fitting a curve such as a Gaussian curve and obtaining the peak of the curve may be used.

Although the Sobel differential filter is used as the sharpness, another differential filter may be used, or another high frequency component extracting filter, for example,
Local dispersion may be used, or frequency analysis may be performed using Fourier transform.

Other modifications can be made without departing from the scope of the present invention.

[0041]

According to the present invention, since a reduced image is targeted through image input and processing, a much faster autofocus function than usual can be realized. Further, the reduced image is obtained by scanning the electron beam over the entire area of the image with a narrow interval, so that stable focus adjustment is always realized. Further, since the image processing is mainly a local calculation, it is easy to increase the speed by using the image processing hardware, and its practical effect is great.

[Brief description of drawings]

FIG. 1 is a diagram for explaining a conventional method.

FIG. 2 is a diagram for explaining a conventional method.

FIG. 3 is an overall configuration diagram of the present embodiment.

FIG. 4 is a diagram showing a flowchart of this embodiment.

FIG. 5 is an explanatory diagram of a reduced image.

FIG. 6 is a diagram showing a Sobel differential mask.

FIG. 7 is a diagram showing a sharpness curve.

[Explanation of symbols]

1 electron gun 2 Focusing lens 3 Stigmata 4 deflector 5 Objective lens 6 detector 7 monitors 8 Electron beam controller 9 Focus control section 10 Image processing section 11 Image input section

─────────────────────────────────────────────────── --Continued from the front page (56) References JP-A-4-104446 (JP, A) JP-A 8-138605 (JP, A) JP-A 7-153407 (JP, A) JP-A 58- 11842 (JP, A) JP 63-202835 (JP, A) JP 59-54159 (JP, A) JP 48-52467 (JP, A) JP 56-41663 (JP, A) JP-A-3-219543 (JP, A) Actual development Sho 51-6861 (JP, U) (58) Fields investigated (Int. Cl. ⁷ , DB name) H01J 37/21

Claims (6)

(57) [Claims] 1. A sample is scanned with an electron beam, secondary electrons generated thereby are detected, the detected amount of the detected secondary electrons is converted into contrast information, and an enlarged image of the sample is displayed. change in the scanning electron microscope to view, and the electron beam control means for the relative sample is scanned interlaced at predetermined intervals in vertical and horizontal directions of the electron beam, the focus for the specimen of the electron beam from the electron beam control means Focus control means for detecting the secondary electrons generated from the sample by the electron beam; detection amount of the secondary electrons detected by the detection means while changing the focus on the sample by the focus control means. From the image for each focus
To jump the electron beam vertically and horizontally at specified intervals.
Image information calculating means for obtaining image information of a reduced image which is an image obtained by scanning, and sharpness for each image information of the reduced image for each focus obtained by the image information calculating means. An in-focus position calculating device comprising image processing means for obtaining a focus corresponding to an in-focus position for the reduced image from the degree. 2. The in-focus position calculation device according to claim 1, wherein the electron beam control means performs interlaced scanning at a constant interval or at different intervals. 3. The image processing means applies a filter for extracting a high frequency component to the image information of the reduced image for each focus obtained by the image information calculating means, and outputs each focus from the output information. The focus position calculation device according to claim 1, wherein the sharpness of the image information of the reduced image is calculated for each, and the focus having the maximum calculated sharpness is set as the focus position. 4. A sample is scanned with an electron beam, secondary electrons generated thereby are detected, the detected amount of the detected secondary electrons is converted into contrast information, and an enlarged image of the sample is displayed. a scanning electron microscope to view, detecting a secondary electron generated from the sample the relative sample and the electron beam control step of scanning and interlaced at predetermined intervals in vertical and horizontal directions of the electron beam by the electron beam While changing the focus on the sample , it is an image for each focus from the detected amount of secondary electrons detected by the detecting step, and the electron beam is changed vertically and horizontally.
Obtained by interlaced scanning at a predetermined interval
Image information calculation step for obtaining image information of a reduced image that is an image, and a sharpness is obtained for each image information for each focus obtained in the image information calculation step. An in-focus position calculation method comprising an image processing step of obtaining a corresponding focus. 5. The focusing position calculation method according to claim 4, wherein the electron beam control step performs interlaced scanning at a constant interval or at different intervals. 6. The image processing step applies a filter for extracting a high frequency component to the image information of the reduced image for each focus obtained in the image information calculating step, and outputs each focus from the output information. The focus position calculation method according to claim 4, wherein the sharpness of the image information of the reduced image is calculated for each of the images, and the focus having the maximum calculated sharpness is set as the focus position.