JP2009288162A - 3D measuring device - Google Patents

Abstract

Provided is a dimension measuring apparatus capable of correcting a pixel shift based on a mechanical error of Z-axis straightness when generating a two-dimensional image and accurately measuring a width between two points of the two-dimensional image.
A confocal microscope 10 is provided so as to be movable in the Z-axis direction by a Z stage 21, and a focus image at a different Z position with respect to a sample 16 is captured by a measurement camera 18 and output to a control unit 20. . The control unit 20 includes a correction table 201 that corrects a pixel shift based on the Z-axis straightness shift of the Z stage 21. The control unit 20 reads the height position of the Z-axis from the linear scale 23 when capturing the camera image, corrects the position of the camera pixel based on the correction data stored in the correction table 201, and is held in the maximum value memory. Compared with the maximum luminance value, a two-dimensional image is generated, and pixel position shift due to Z-axis straightness shift is prevented.
[Selection] Figure 1

Description

  The present invention relates to a three-dimensional measuring apparatus that generates a two-dimensional image related to height and measures a width between two points from the two-dimensional image.

  In recent years, there has been provided a three-dimensional measuring apparatus that acquires a confocal image of a sample such as an LCD substrate or a semiconductor wafer with a confocal microscope and measures the surface shape (height) of the sample using the image. This type of three-dimensional measurement apparatus uses, for example, a confocal optical microscope with a shallow depth of field, and is enlarged by the microscope while moving the in-focus position of the confocal optical microscope in the optical axis direction (Z-axis). An optical image of the sample is picked up by the measurement camera, and the picked-up image data is stored in the memory. Then, by detecting a pixel having the maximum luminance with respect to each pixel of the captured image data, and collecting the pixels having the maximum luminance and composing one image, an omnifocal image (extended focus image) can be acquired. . Alternatively, a height image or a cross-sectional profile can be acquired by creating one image assuming that the focus position (height) when the maximum luminance is obtained is the position of each pixel.

  Conventionally, a width between two points is measured from a two-dimensional image generated by the three-dimensional measuring apparatus.

  The three-dimensional measuring apparatus generates a two-dimensional image as shown in FIG. 4 and measures the width between two points from the two-dimensional image. That is, when each pixel (pixel) 2 of the camera image 1 shown in FIG. 4A captured by the measurement camera is moved (scanned) in one vertical direction as shown in FIG. The brightness value of each pixel ZP1, ZP2,... Obtained from the measurement camera is compared with the maximum brightness value of the corresponding pixel held in the maximum value memory, and the planar image indicated by the peak of the brightness value and the Z at the peak time are compared. The axis height value is stored in the memory. In the three-dimensional measuring apparatus, a linear scale for obtaining height information is mounted on the Z stage driving unit that moves the confocal optical microscope in the Z-axis direction. It is stored in the memory as the Z-axis height value.

  In the generation of the two-dimensional image, when there is an error in the Z-axis straightness as shown in FIG. 5, the vertical movement of the Z-axis is not performed perpendicular to the sample. For example, when the position of the pixel ZP2 is shifted, the maximum value comparison of each pixel is erroneously compared with the neighboring pixel ZPx. Therefore, the generated two-dimensional image includes an error of the Z-axis straightness, and if the width between two points is measured using this image, a correct result may not be obtained.

As a publicly known technique for correcting image displacement, a scanning type that corrects displacement and stores the corrected image data while detecting the displacement amount on the XY plane caused by vibration caused by the floor or an electric part inside the apparatus. A laser microscope (for example, refer to Patent Document 1), a laser scanning microscope (for example, refer to Patent Document 2) that detects and corrects an amount by which a specimen shifts in the XY-axis direction over time are known.
JP 2002-98901 A JP 2004-191846 A

  As described above, when a two-dimensional image is generated by the three-dimensional measuring apparatus and the width between two points is measured from the two-dimensional image, if there is a mechanical error in the Z-axis straightness when generating the two-dimensional image, Z Since the vertical movement of the axis is not performed perpendicularly to the sample, the position of the pixel fluctuates according to the error, and when comparing the maximum value of each pixel of confocal images with different heights, Therefore, there is a problem that a measurement error occurs in the width measurement between two points.

  The present invention has been made to solve the above-described problems, and corrects a pixel shift based on a mechanical error of the Z-axis straightness when generating a two-dimensional image, and measures a width between two points in the two-dimensional image. An object of the present invention is to provide a three-dimensional measuring apparatus that can be performed correctly.

In order to achieve the above object, a three-dimensional measurement apparatus according to the present invention captures an optical image of a sample that has been magnified by the confocal microscope that magnifies the optical image of the sample and the confocal microscope, and acquires the image data. An output camera; a drive mechanism that variably moves the focal position of the confocal microscope with respect to the sample in the optical axis direction; and a linear scale that measures the position in the optical axis direction of the confocal microscope driven by the drive mechanism And a control unit that controls the drive mechanism and processes image data captured by the camera,
The control unit obtains a correction table storing correction data for correcting pixel shift based on a shift in straightness in the optical axis direction of the driving mechanism, and image data of the sample imaged by the camera Each of the pixels of the image data by referring to the correction data stored in the correction table based on the measurement value of the linear scale when acquiring the image data of the camera by the image acquisition means and the image acquisition means An image for generating a two-dimensional image indicated by the peak of the luminance value by comparing the luminance value of each pixel corrected by the correcting unit with the maximum luminance value held in the maximum value memory. And generating means.

  According to the present invention, at the time of generating a two-dimensional image, two points in the two-dimensional image are generated by correcting the positional deviation of the camera pixels due to the deviation of the Z-axis straightness based on a correction table created in advance. The width measurement can be accurately performed.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a configuration diagram of a three-dimensional measuring apparatus according to an embodiment of the present invention. In FIG. 1, reference numeral 10 denotes a confocal microscope, which is mainly composed of a confocal unit 11. The confocal unit 11 includes an objective lens 12 on the lower side and a beam splitter (not shown) inside, and light irradiated from an external illuminating device 13 is projected onto the beam splitter unit and the light guide 14. The light is guided through the light tube 15 and condensed on the sample 16 from the beam splitter section through the objective lens 12.

  The illuminating device 13 is a light source using, for example, a metal halide lamp, and can change the light amount (illuminance) and the wavelength band by an electric diaphragm or filter according to a command from the control unit 20.

  The sample 16 is placed on the inspection stage 17. The inspection stage 17 is driven in the X and Y horizontal directions by a drive unit (not shown) controlled by the control unit 20. In addition, a measurement camera 18 using a CCD camera, for example, is mounted on the confocal unit 11.

  The light irradiated on the sample 16 from the illumination device 13 via the objective lens 12 is reflected by the sample 16 and passes through the objective lens 12 and enters the confocal unit 11. The confocal unit 11 has a characteristic of allowing only focused light to pass therethrough, and the measurement camera 18 is configured to display only an image of a currently focused part.

  The confocal microscope 10 is provided so as to be movable in the vertical direction by the Z stage 21. The Z stage 21 is driven by a Z motor 22 to move the focal position of the confocal microscope 10 with respect to the sample 16 in the vertical direction (Z-axis direction), that is, in the optical axis direction of the confocal microscope 10.

  The Z motor 22 is driven by a Z stage drive unit (not shown) that operates according to an instruction from the control unit 20. This Z stage drive unit is equipped with a high-resolution linear scale 23 that measures the focal position in the Z-axis direction of the confocal microscope 10 with high accuracy, and the Z-axis coordinates acquired by the linear scale 23 are instructions from the control unit 20. The Z stage 21 is controlled so as to coincide with.

  By moving the Z stage 21 as described above, focus images at different Z positions can be taken by the measurement camera 18. An image captured by the measurement camera 18 is sent to the control unit 20 for processing.

  The control unit 20 has a function of controlling the entire apparatus, and also includes a correction table 201 for correcting pixel positional deviation based on deviation of Z-axis straightness when the Z stage 21 is moved in the Z-axis direction. Yes. In this correction table 201, for example, a high-precision measuring instrument such as a laser interferometer is used to measure each position when the Z stage 21 is moved in the Z-axis direction, and the Z-axis is based on the measurement result. Correction data for correcting a pixel position shift accompanying a straightness shift is stored.

  Although not shown, the control unit 20 obtains image data of the sample 16 imaged by the measurement camera 18, and the linear scale 23 when obtaining the image data of the measurement camera 18 by the image acquisition means. The correction data stored in advance in the correction table 201 is referred to based on the measured value of the correction data, the correction means for correcting the position of each pixel 2 in the image data, and the luminance value of each pixel corrected by the correction means is maximized. Compared with the maximum luminance value held in the value memory, an image processing function including an image generating means for generating a two-dimensional image indicated by the peak of the luminance value and storing it in the memory is provided.

  Next, an operation when generating a two-dimensional image using the three-dimensional measuring apparatus configured as described above will be described. At the time of generating a two-dimensional image, the illumination device 13 emits predetermined illumination light in accordance with a control command from the control unit 20. The light emitted from the illumination device 13 is guided into the confocal unit 11 through the light guide 14 and the light projecting tube 15, and is condensed on the sample 16 by the objective lens 12.

  The light irradiated on the sample 16 is reflected by the sample 16 and passes through the objective lens 12 and enters the confocal unit 11. Only the focused light passes through the confocal unit 11 and is imaged by the measurement camera 18. . The image of the sample 16 captured by the measurement camera 18 is sent to the control unit 20.

  As shown in FIG. 4A, the control unit 20 moves the Z axis in one vertical direction as shown in FIG. 4B in each pixel 2 of the camera image 1 captured by the measurement camera 18 as shown in FIG. The brightness value of each pixel ZP1, ZP2,... Obtained from the hour measurement camera 18 is compared with the maximum brightness value held in the maximum value memory, and the two-dimensional image indicated by the peak of the brightness value and the Z at the peak The axis height value is stored in the memory.

  That is, the control unit 20 drives the Z stage 21 in a state where the illumination light is irradiated from the illumination device 13 to move the focal position of the confocal microscope 10 stepwise in the optical axis direction. The image of the sample 16 at the focal position is taken by the measurement camera 18. Further, every time an image of the sample 16 at each focal position is obtained by this imaging, the luminance value is compared with the pixel at the same position of the maximum luminance image. At this time, if the luminance value of the pixel 2 in the currently captured camera image 1 is higher, the luminance value of the pixel of the maximum luminance image is updated and stored in the maximum value memory. At the same time, the focal position of the confocal microscope 10 when the pixel having the high luminance value is imaged is stored in the memory in association with the pixel. This is the height image. As the focal position (the height in the optical axis direction) of the confocal microscope 10, a value measured with the linear scale 23 is used. Since the initial values of the maximum brightness image are all 0, the first image obtained at the first position where the focal position is moved becomes the maximum brightness image at that time.

  When creating the two-dimensional image, the control unit 20 reads the Z-axis height position at the time of image capture from the linear scale 23 and calculates the Z-axis straightness based on the correction table 201 created in advance. Then, the pixel position of the camera image 1 is corrected, compared with the maximum luminance value held in the maximum value memory, and a two-dimensional image indicated by the peak of the luminance value is generated and stored in the memory.

  A correction operation in consideration of the Z-axis straightness at the time of generating the two-dimensional image will be described with reference to FIG. FIG. 2 shows a case where, in the Z-axis straightness, an error in the right direction indicated by the arrow a is generated for the pixel ZP2 and an error in the left direction indicated by the arrow b is generated for the pixel ZP3. The lengths of the arrows a and b represent actual straightness, and the pixels ZP2 and ZP3 are corrected in the direction opposite to the direction of straightness according to the length as indicated by arrows a ′ and b ′. Compare with the maximum brightness value stored in the maximum value memory. These processes are performed on all pixels of the image captured by the measurement camera 18.

  As described above, at the time of generating a two-dimensional image, the two-dimensional image is used by generating an image by correcting the positional deviation of the camera pixel accompanying the deviation of the Z-axis straightness based on the correction table 201 created in advance. In measuring the width between two points, it is possible to accurately measure without causing an error.

  In the case of measuring the width between two points using the two-dimensional image, if it is desired to further improve the measurement resolution, the camera image 1 captured by the measurement camera 18 as shown in FIGS. 3 (a) and 3 (b). For each pixel 2, the number of pixels is expanded several times by interpolating between adjacent pixels, and for example, the pixel 2 b is interpolated between the left and right adjacent pixels 2 a and 2 c to increase the number of pixels by 5 times. By performing the correction considering the Z-axis straightness based on the correction table 201, the measurement can be performed with a resolution of 1/5 of the pixel resolution of the measurement camera 18. Note that the pixels 2a to 2c shown in FIG. 3B represent arbitrary three pixels in the camera pixels. An actual two-dimensional image is interpolated by a product-sum operation (filter) of 3 × 3 pixels.

  Measurement resolution can be improved by performing pixel interpolation on the image 1 captured by the measurement camera 18 as described above, expanding the number of pixels several times, and correcting the Z-axis straightness in units of subpixels. .

  Hereinafter, another example of a method for creating the correction table 201 will be described. In this preparation method, a sample having a linear structure derived from crystallinity is used for calibration. As the calibration sample, for example, a pyramid-shaped silicon single crystal grown on a silicon (100) substrate can be used. This Si pyramid has exactly four (111) planes, and its sides (facet edges). Is a straight line. Such a calibration sample is measured with the above-mentioned confocal microscope, and an omnifocal image and a height image are obtained.

  FIG. 6 is an omnifocal image of the Si pyramid when an incomplete correction table is used. Since strong reflected light returns only from the substrate surface and the edge, the edge can be easily distinguished from other parts. Since straightness is not fully compensated, the edges that should be straight are slightly curved.

  Next, the obtained omnifocal image and height image are taken into a personal computer or the like, and an area where an edge is to be detected by image processing is designated in a band shape by an operation of the operator. In FIG. 6, this area is indicated by hatching. The belt-like area is set so that only one edge is included therein. Then, using a known edge detection method such as the Canny method, an edge is detected from the omnifocal image in the band-like area, and the X and Y coordinates of the pixel detected as the edge and the height image value (Z coordinate) are detected. ) To obtain a plurality of sets.

  Next, a regression line is calculated for these sets by the least square method, and the deviation of x and y from the regression line is set as a necessary compensation amount. That is, the deviations Vx and Vy are each interpolated by polynomial (quadratic) approximation, Nyquist interpolation, etc. as functions of Z, and the deviations Vx (Z) and Vy (Z) are added to the current correction table values. .

  According to this method, the compensation fee is calculated based on the omnifocal image and the height image. Therefore, the three-dimensional measurement method itself does not need to be changed, and the Z scan start / end position, the scan speed, and the like are actually measured. There is an advantage that a correction table can be created under the same conditions (Z scan speed). However, the height of the calibration sample must be adjusted to the same level as the actual sample (by raising the substrate, etc.) so that a focused image can be obtained in the Z scan range. If it is repeated, the accuracy of the correction table can be further improved.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying constituent elements without departing from the scope of the invention in the implementation stage.

It is a block diagram of the three-dimensional measuring apparatus which concerns on one Embodiment of this invention. It is a figure for demonstrating the correction | amendment operation | movement which considered the Z-axis straightness at the time of the two-dimensional image generation in the same embodiment. FIG. 6 is an operation explanatory diagram in the case where the measurement resolution is improved by extending camera pixels by pixel interpolation when generating a two-dimensional image in the embodiment. It is explanatory drawing in the case of producing | generating a two-dimensional image with a three-dimensional measuring apparatus. It is a figure for demonstrating the influence by Z-axis direction straightness in the case of producing | generating a two-dimensional image with the conventional three-dimensional measuring apparatus. It is an omnifocal image of Si pyramid when a correction table which is not perfect is used.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Confocal microscope, 11 ... Confocal unit, 12 ... Objective lens, 13 ... Illuminating device, 14 ... Light guide, 15 ... Projection tube, 16 ... Sample, 17 ... Inspection stage, 18 ... Measurement camera, 20 ... Control Unit: 21 ... Z stage, 22 ... Z motor, 23 ... linear scale, 201 ... correction table.

Claims (1)

A confocal microscope that magnifies an optical image of the sample, a camera that captures an optical image of the sample magnified by the confocal microscope and outputs the image data, and a focal position of the confocal microscope with respect to the sample A drive mechanism that variably moves in the optical axis direction, a linear scale that measures a position in the optical axis direction of the confocal microscope driven by the drive mechanism, and an image captured by the camera while controlling the drive mechanism A control unit for processing data,
The control unit is
A correction table that stores correction data for correcting a shift of a pixel based on a shift in straightness in the optical axis direction of the drive mechanism;
Image acquisition means for acquiring image data of the sample imaged by the camera;
Correction for correcting the position of each pixel of the image data by referring to the correction data stored in the correction table based on the measurement value of the linear scale when acquiring the image data of the camera by the image acquisition means Means,
Image generation means for comparing the luminance value of each pixel corrected by the correction means with a maximum luminance value held in a maximum value memory, and generating a two-dimensional image indicated by the peak of the luminance value. 3D measuring device.

Images