JP2009128063A - Three-dimensional shape measuring device and its operation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a three-dimensional shape measuring device capable of reducing an accidental error and a quantization error caused by the acquisition of the discrete light intensity information of Z-scanning. <P>SOLUTION: This three-dimensional shape measuring device is equipped with a moving means for moving a sample position relatively in the optical axis direction to an objective lens; a rotation number acquisition means for acquiring the number of times for generating three-dimensional shape information; a movement control means for moving the sample position relatively and discretely to the objective lens by controlling the moving means; and a three-dimensional shape information generation means for performing, in the number of acquisition times, processing for acquiring the discrete light intensity information of observation light from a sample corresponding to a change of the sample position, acquiring the sample position where the maximum light intensity is acquired based on the light intensity information, and generating the three-dimensional shape information based on the sample position. When the three-dimensional shape information is generated in the number of acquisition times, the movement control means controls the moving means so that each sample position where the light intensity information is acquired discretely is made to differ from each other at each generation of the three-dimensional shape information. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a three-dimensional shape measuring apparatus using a confocal microscope.

  In a confocal microscope, a sample is point-illuminated with a point light source, and the transmitted light or reflected light from the sample is collected on the pinhole, and then the intensity of the light transmitted through the pinhole is detected by a photodetector. To obtain the surface information of the sample.

  FIG. 9 shows a schematic configuration of a general confocal microscope. In the confocal microscope shown in the figure, the point light emitted from the point light source 1 is transmitted through the half mirror 2 and then condensed onto the sample surface 4 by the objective lens 3 whose aberration is corrected. The reflected light reflected from the illuminated sample surface 4 is again introduced from the objective lens 3 to the half mirror 2, and then reflected by the half mirror 2 and condensed at a predetermined position. The reflected light from other than the focal point on the sample 4 is cut by the pinhole 5 provided at the condensing position, and only the light transmitted through the pinhole 5 is detected by the photodetector 6.

  Such light detection is performed over the entire sample surface 4, and the reflected light from the sample surface 4 is two-dimensionally scanned in the same manner as the raster scan of a television. Thereby, a two-dimensional image of the sample surface 4 can be obtained.

  By the way, the surface of the sample surface 4 is not a flat surface, and for example, there is a surface deviated from the light condensing position of the objective lens 3 as indicated by reference symbol A. In the confocal microscope, the light reflected from the surface A (shown by a broken line) is not collected on the pinhole 5. Therefore, the reflected light from other than the condensing position is cut by the pinhole 5 and is not detected by the photodetector 6. For this reason, the confocal microscope can measure only the optical image of the sample surface 4 existing at the condensing position of the objective lens 3, that is, the in-focus position, with high accuracy. In addition, the relative distance between the sample and the objective lens is discretely moved by a predetermined moving step amount, the focal point is moved in the optical axis direction, and the light intensity is detected at each position to obtain a slice image in a three-dimensional space. Will be.

  For a sample such as a semiconductor substrate in which light from a light source does not pass through the surface, the position in the optical axis direction where the light intensity is maximum is considered to be in focus on the sample surface. Therefore, the relative distance between the sample and the objective lens is discretely moved by a predetermined moving step amount to obtain a plurality of slice images, and the maximum light intensity in the optical axis direction is obtained from the obtained plurality of slice images. By obtaining the height information for each given pixel, the three-dimensional shape information of the sample can be obtained.

  With this three-dimensional shape information, if measurement is performed with a small amount of movement step in the optical axis direction (Z direction), a high-resolution image can be obtained, but the measurement time is increased accordingly. Therefore, there is a method of improving the Z scanning throughput and increasing the resolution by increasing the moving step amount in the Z direction (for example, Patent Document 1).

  Patent Document 1 discloses the following method. First, the relative position between the focusing position of the objective lens and the sample is discretely changed in the optical axis direction (Z direction). Then, at each relative position, light emitted from the light source is incident on the sample through the objective lens of the optical microscope, and the light intensity of the reflected light from the sample at each position is detected. Thereafter, based on a plurality of light intensity detection values including the maximum light intensity detection value, a change curve indicated by the light intensity is obtained. And the relative position which gives the maximum value on this change curve is estimated, and the relative position is used as height information.

  Specifically, the change curve indicated by the light intensity is approximated by a quadratic function using a total of three light intensity detection values of the Z-direction position where the maximum light intensity is detected and the Z-direction positions before and after that position. Thus, the height position where the light intensity becomes the maximum value is estimated. By the method disclosed in Patent Document 1, it is possible to obtain three-dimensional shape information by increasing the amount of movement step that moves at one time in Z scanning. This reduces the number of movements and improves the Z scanning throughput.

  Patent Document 2 discloses the following Z scanning method. First, a plurality of confocal images are acquired by changing the relative distance between the sample and the objective lens. Then, an estimated value of the height of the sample is obtained from the plurality of confocal images using a plurality of estimation methods. Then, a plurality of estimated values obtained by each of the plurality of estimation methods are added and averaged to obtain height information. Thereby, the influence of the error included in the height estimation value by one estimation method can be reduced by the other estimation values.

Patent Document 3 discloses the following method. A plurality of confocal images are sequentially acquired and stored at the same position (Z position) in the optical axis direction, and these images are combined in units of pixels. The mixed image generated in this way is set as three-dimensional shape information. Thereby, the dispersion | variation in the light intensity information in each Z position is reduced, and the resolution in the XY direction and the Z direction is improved.
JP-A-9-68413 JP 2004-61334 A Special Table 2002-517774

However, since the methods of Patent Document 1 and Patent Document 2 are based on only one Z-scan, the acquired light intensity information includes a chance error.
Further, in order to reduce the accidental error, it is necessary to apply the method of Patent Document 3 or perform a plurality of Z scans. However, even if the Z scan is performed a plurality of times, if the relative distance between the sample and the objective lens is simply changed and the acquisition of the light intensity information is repeated, the quantization error caused by performing the Z scan discretely will be reduced. The effect is inevitable.

  In view of the above-described problems, the present invention provides a three-dimensional shape measuring apparatus that reduces accidental errors and quantization errors caused by Z-scan discrete light intensity information acquisition.

  A three-dimensional shape for measuring a three-dimensional shape of a specimen from a plurality of two-dimensional images picked up by a microscope that can move the position of the specimen relative to the objective lens at least in the optical axis direction according to the present invention. The measuring device controls the moving means for moving the sample position relative to the objective lens in the optical axis direction, and controls the moving means so that the sample position is relatively discrete with respect to the objective lens. Movement control means for moving the sample, frequency acquisition means for acquiring the number of times of generating the three-dimensional shape information of the specimen, and discrete light intensity information of the observation light from the specimen according to the change in the specimen position Acquiring the specimen position for obtaining the maximum light intensity based on the light intensity information for obtaining the maximum intensity, and obtaining the processing for generating the three-dimensional shape information based on the specimen position by the number-of-times obtaining means. Three-dimensional shape information generating means for performing the number of times, wherein the movement control means discretely acquires the light intensity information when the three-dimensional shape information is generated for the number of times acquired by the number of times acquisition means. The moving means is controlled so that the sample positions differ from each other every time the three-dimensional shape information is generated.

  In the three-dimensional shape measuring apparatus, the three-dimensional shape information generating means is generated by making the sample positions from which the light intensity information is discretely obtained different for each generation of the three-dimensional shape information. The three-dimensional shape information corresponding to the number of times is averaged to obtain the three-dimensional shape information of the sample.

  The three-dimensional shape measurement apparatus further includes a measurement target acquisition unit configured to acquire setting information of a predetermined measurement target set based on the three-dimensional shape information generated in advance, and based on the setting information of the measurement target. Measuring means for measuring a three-dimensional shape corresponding to the measurement object from the three-dimensional shape information obtained by the three-dimensional shape information generating means.

  In the three-dimensional shape measuring apparatus, the movement control means obtains the number of deviations when the sample positions from which the light intensity information is discretely obtained differ from one another for each generation of the three-dimensional shape information. It changes according to the frequency | count acquired by the means.

  In the three-dimensional shape measuring apparatus, the shift amount is calculated by dividing a single movement step amount by the movement control unit by the number of times acquired by the number of times acquisition unit.

  In the three-dimensional shape measuring apparatus, the amount of deviation when the position at which the light intensity information is discretely acquired by the movement control means is different for each generation of the three-dimensional shape information can be arbitrarily set. Features.

  In the three-dimensional shape measuring apparatus, the three-dimensional shape information generating means extracts a predetermined number of light intensity information from a plurality of the light intensity information when acquiring the sample position for obtaining the maximum light intensity, Calculating a maximum value on a change curve suitable for the predetermined number of light intensity information, estimating the sample position corresponding to the maximum value, and acquiring the estimated sample position as three-dimensional shape information And

  A three-dimensional shape for measuring a three-dimensional shape of a specimen from a plurality of two-dimensional images picked up by a microscope that can move the position of the specimen relative to the objective lens at least in the optical axis direction according to the present invention. The operation method of the measuring apparatus is such that the sample position relative to the objective lens is moved relatively and discretely in the optical axis direction, the number of times of generating the three-dimensional shape information of the sample is obtained, and the change in the sample position According to the method, obtaining discrete light intensity information of the observation light from the specimen, obtaining the specimen position for obtaining the maximum light intensity based on the light intensity information, and obtaining the three-dimensional shape information based on the specimen position When the three-dimensional shape information is generated for the acquired number of times, the sample position where the light intensity information is acquired discretely is generated each time the three-dimensional shape information is generated. In So as to become, and wherein the moving the specimen position.

  According to the present invention, it is possible to reduce the accidental error and the quantization error caused by the Z-scan discrete light intensity information acquisition.

  The three-dimensional shape measuring apparatus according to the embodiment of the present invention provides a 3D image of a specimen from a plurality of two-dimensional images picked up by a microscope that can move the position of the specimen relative to the objective lens at least in the optical axis direction. Dimensional shape information is acquired. The three-dimensional shape measuring apparatus includes moving means, number of times acquisition means, movement control means, and three-dimensional shape information generating means.

The moving means moves the sample position relative to the objective lens relatively in the optical axis direction. The moving means corresponds to the stage 34 in the present embodiment, for example.
The number acquisition means acquires the number of times to generate the three-dimensional shape information of the specimen. For example, in the present embodiment, the number-of-times acquisition unit corresponds to the computer 28 that displays a predetermined input screen and sets the number of times to execute the three-dimensional shape information generation process.

  The movement control means controls the movement means to move the sample position discretely relative to the objective lens. The movement control means corresponds to, for example, the Z-direction movement control circuit 44 in the present embodiment.

  The three-dimensional shape information generating means acquires discrete light intensity information of the observation light from the sample according to the change in the sample position, and acquires the sample position for obtaining the maximum light intensity based on the light intensity information Then, the process of generating the three-dimensional shape information based on the sample position is performed the acquired number of times. The three-dimensional shape information generation means corresponds to, for example, the three-dimensional shape information calculation means 42 (or the computer 28 may be included) in the present embodiment.

  In such a configuration, when the three-dimensional shape information is generated for the acquired number of times, the movement control unit determines the sample position for discretely acquiring the light intensity information for each generation of the three-dimensional shape information. The moving means is controlled to be different from each other. The three-dimensional shape information generating means generates the three-dimensional shape corresponding to the number of times generated by making the sample positions from which the light intensity information is discretely obtained different for each generation of the three-dimensional shape information. The information is averaged to obtain the three-dimensional shape information of the sample.

  Further, the movement control means changes a deviation amount when the sample positions for discretely acquiring the light intensity information are made different for each generation of the three-dimensional shape information according to the acquired number of times. Let The deviation amount is calculated by dividing one movement step amount by the movement control means by the acquired number of times. In addition, you may set arbitrarily the deviation | shift amount at the time of making the said position from which the said light control information discretely acquired by the said movement control means differ mutually for every production | generation of the said three-dimensional shape information.

Thereby, it is possible to reduce the accidental error and the quantization error caused by the Z-scan discrete light intensity information acquisition.
The three-dimensional shape measurement apparatus may further include a measurement object acquisition unit and a measurement unit.

  The measurement target acquisition unit acquires setting information of a predetermined measurement target set based on the three-dimensional shape information generated in advance. For example, in the present embodiment, the measurement target acquisition unit is a computer 28 that can set a position to be measured and a position between the positions with respect to the three-dimensional shape information generated in advance and displayed on the monitor 46. Equivalent to.

  The measurement unit measures a three-dimensional shape corresponding to the measurement target from the three-dimensional shape information obtained by the three-dimensional shape information generation unit based on the setting information of the measurement target. For example, in the present embodiment, the measurement means corresponds to the three-dimensional shape information calculation means 42 (or the computer 28 may be included) that executes S14 and S15 in FIG.

  By configuring in this way, it is possible to reduce accidental errors and quantization errors due to discrete Z scanning, and realize a measurement method that is easy to operate for the observer. is doing.

  Further, the three-dimensional shape information generating means extracts a predetermined number of light intensity information from a plurality of the light intensity information when acquiring the sample position for obtaining the maximum light intensity, and the extracted predetermined number of light intensity A maximum value on a change curve that matches the information may be calculated, the sample position corresponding to the maximum value may be estimated, and the estimated sample position may be acquired as three-dimensional shape information.

With this configuration, an effect of further reducing the Z scanning quantization error can be obtained. Therefore, it is possible to obtain more accurate three-dimensional shape information with the same number of movements.
Hereinafter, embodiments of the present invention will be described in detail.

<First Embodiment>
In the confocal scanning microscope according to the present embodiment, when the three-dimensional shape information of the sample is obtained a plurality of times using the confocal optical system, the positions at which the three-dimensional shape information of the sample is obtained discretely are different from each other. The three-dimensional shape information of the sample is obtained by averaging the obtained three-dimensional shape information.

  FIG. 1 shows a configuration of a confocal scanning optical microscope system according to the first embodiment. The first embodiment is an example in which surface information is acquired by two-dimensionally scanning a sample using the confocal optical system 16.

In the confocal optical system 16, the scanning laser light emitted from the laser light source 18 is reflected by the mirror 20 and enters the two-dimensional scanning mechanism 24 through the half mirror 22.
The two-dimensional scanning mechanism 24 is connected to a computer 28 via a scanning control unit 26. The driving of the two-dimensional scanning mechanism 24 is controlled based on a scanning control signal P1 output from the scanning control unit 26 in response to a command from the computer 28. The two-dimensional scanning mechanism 24 condenses the scanning laser light on the sample 36 on the stage 34 to a minute spot via the objective lens 32 set on the revolver 30 based on the scanning control signal P1. In this state, the two-dimensional scanning mechanism 24 scans the sample 36 in the XY directions on the sample 36 in the same manner as the raster scanning of the television.

  The reflected light reflected from the sample 36 in the two-dimensional scanning by the scanning laser light is guided to the half mirror 22 through the objective lens 32 and the two-dimensional scanning mechanism 24, and reflected by the half mirror 22 toward the photodetector 40 side. Is done.

  The reflected light reflected by the half mirror 22 passes through the pinhole 38 disposed at a position conjugate with the condensing position of the objective lens 32 and then enters the photodetector 40. The photodetector 40 converts the incident reflected light into an electrical signal corresponding to the amount of light and outputs it to the three-dimensional shape information calculation means 42.

  The three-dimensional shape information calculation means 42 is connected to the photodetector 40, the computer 28, and the Z direction movement control circuit 44. The three-dimensional shape information calculation unit 42 can store the electrical signal output from the photodetector 40 and the command information to the computer 28 and the Z direction movement control circuit 44.

  The stage 34 is controlled to move in the Z direction by a predetermined amount based on the Z control signal P2 output from the Z direction movement control circuit 44 in response to a command from the computer 28. At this time, the amount of movement step per stage of the stage 34 is controlled by the computer 28.

  The setting of the measurement range, the setting of the moving step amount of the stage 34 in each measurement range, the image display, the control of the microscope system, and the like are set by the observer via the monitor 46 connected to the computer 28.

  In the confocal scanning optical microscope system configured as described above, after the observer places the sample 36 on the stage 34, the micro spot focused on the sample 36 by the control of the computer 28 is arranged in the XY directions. Scan.

  Then, the stage 34 is moved and controlled in the Z direction to perform a focusing operation on the sample 36. At this time, whether or not the sample 36 is in focus is determined while viewing the image displayed on the monitor 46.

Next, the observer sets parameters related to the measurement operation. First, after setting the measurement range L of the sample 36 and the position Z ₀ of the stage 34 where measurement is started by the computer 28, the movement step amount Δ per stage of the stage 34 in Z scanning is set.

  In this embodiment, the movement step amount Δ is a single movement amount when the position of the sample 36 is moved in the Z direction with respect to the focal position of the objective lens 32. When the measurement range L and the movement step amount Δ per stage 34 are set, the number N of movements of the stage 34 is determined according to the relationship L / Δ ≦ N.

  In addition, the confocal microscope according to the present embodiment has two measurement modes, a normal measurement mode and a repeated measurement mode. The normal measurement mode is a method for generating the three-dimensional shape information of the sample by performing the Z scan once for the set measurement range L, and is the normal measurement method described in the prior art. On the other hand, the repeated measurement mode will be described in detail later, but is a measurement method in which three-dimensional shape information of a sample is generated after repeatedly performing Z scanning for the designated number of times. When the repeated measurement mode is selected, the observer sets the repeated acquisition count R.

  Hereinafter, the operation in the repeated measurement mode will be described in detail. The observer selects the repeated measurement mode, sets the measurement range L, the moving step amount Δ, and the repeated acquisition count R, and then starts measuring the sample 36. Although the number of movements N is automatically determined according to the relationship L / Δ ≦ N, a desired number of movements N can be set.

The measurement operation will be described below with reference to the flowchart shown in FIG.
FIG. 2 shows an operation flow in the repeated measurement mode in the present embodiment. When measurement is started, the following processing is executed by the three-dimensional shape information calculation means 42. First, the repeated acquisition count value r is reset (S1). Next, the measurement start position Z ₀ is set (S2).

Next, the stage 34 moves to the measurement start position Z ₀ , and the counter n of the number of movements of the stage 34 is reset. At this time, the electrical signal I ₀ output from the photodetector 40 is stored in the memory in the three-dimensional shape information calculation means 42 as M (0) (S3).

Next, the stage 34 is moved by the moving step amount Δ, and the counter n of the number of movements of the stage 34 is counted up (S4).
Then, the electric signal I _n and the three-dimensional shape information value M (0) stored in the computing means 42 which is output from the photodetector 40 and compares the (S5). The greater the direction of I _n from M (0) ( "Yes" in S5), and overwrites the value of I _n the M (0). In C (0), the count value n of the number of movements of the stage 34 at this time is stored (S6).

S4~S6 processing of by repeating N times (S7), the maximum value of the electric signal I _n output from the photodetector 40 M (r) and the count of the number of movements of the stage 34 when showing the maximum light intensity The value C (r) is stored in the memory of the three-dimensional shape information calculation means 42.

Next, the repeated acquisition count value r is counted up (S8). Further, the repeated acquisition count value r is compared with the repeated acquisition count R (S9). If r = R, the measurement is completed (“Yes” in S9). If r <R (“No” in S9), the process returns to S2. In this case, the stage initial position Z ₀ is shifted from the previous Z ₀ by Δ / R and overwritten as Z ₀ + r · (Δ / R) (S 2).

As described above, the stage initial position Z ₀ is shifted by the number of times of repeated acquisition R, and the moving step count at the maximum light intensity M (0) to M (R−1) and the maximum light intensity of each Z _0. Values C (0) to C (R-1) are stored in the memory of the three-dimensional shape information calculation means 42. Furthermore the intensity of reflected light from the sample is the maximum true, to determine the relative position Z _t of the objective lens and the sample, moving step count value at maximum light intensity which is stored in the memory of the three-dimensional shape information calculating unit 42 The following calculation is performed by the computer 28 using C (0) to C (R-1).

Z _t = (Z (1, m) + Z (2, m) +... + Z (R, m)) / R
However, Z (1, m), Z (2, m),..., Z (R, m) are respectively the maximum light intensity M (0), stored in the memory of the three-dimensional shape information calculation means 42, This is the value of Z when M (1),..., M (R-1). In this way, finally, a value obtained by averaging the three sets of three-dimensional shape information of the repeated acquisition times R is generated as the sample height information.

The effect of the measurement operation described above will be described with reference to FIG.
FIG. 3 shows the relationship between the position Z in the Z direction and the light intensity I output from the photodetector 40 when the position is fixed in the XY direction and the position is moved discretely in the Z direction in this embodiment ( Hereinafter referred to as “IZ curve”). Further, ◯, ×, and □ shown in the figure are Z and I values near the maximum light intensity when the number of repeated acquisitions is 3 in the measurement operation of the present embodiment. Z and I in the first measurement are indicated by ◯, the second time is indicated by ×, and the third time is indicated by □.

  Further, for Z and I in the first measurement indicated by ◯, Z and I when I shows the maximum value are Z (1, m) and I (1, m), respectively. Also, Z and I one step before Z (1, m) and I (1, m) are respectively Z (1, m-1) and I (1, m-1). Also, Z and I after one step from Z (1, m) and I (1, m) are respectively Z (1, m + 1) and I (1, m + 1).

  Similarly, for Z and I in the second measurement indicated by x, Z and I when I shows the maximum value are Z (2, m) and I (2, m), respectively. In addition, Z and I in the previous step from Z (2, m) and I (2, m) are set to Z (2, m-1) and I (2, m-1), respectively. Further, Z and I after one step from Z (2, m) and I (2, m) are respectively Z (2, m + 1) and I (2, m + 1). Also, since the third time is also denoted by the same reference, description thereof is omitted.

As shown in the figure, the acquisition position of the light intensity information at the first measurement, the second measurement, and the third measurement is different by Δ / 3 with respect to the movement step amount Δ. Thus the three Z acquired light intensity is maximized, as a Z _t as described above, these average values _{Z t = (Z (1,} m) + Z (2, m) + Z (3 M)) / 3 is calculated and generated as the final three-dimensional shape information of the sample.

From this figure, it can be expected that the quantization error is obviously reduced by discretely performing the Z scan, and the influence of the accidental error can be reduced by the arithmetic operation of addition averaging.
According to this embodiment, the relative position between the objective lens and the sample is sequentially changed, Z scanning for obtaining the maximum light intensity of the sample and the relative position at that time is repeated a predetermined number of times, and the initial stage position is shifted. By shifting the position of taking in each light intensity information from each other, accidental error and quantization error due to discrete Z scanning are reduced for the three-dimensional shape information finally obtained as an average value of them. It becomes possible to do.

  Although the shift amount of the initial stage position is Δ / R, the observer can arbitrarily set it. Thereby, the deviation | shift amount of the position of taking in each light intensity information can be set arbitrarily.

  Note that, as a modification of the first embodiment in the first embodiment described above, one Z position indicating the maximum light intensity and the Z position 2 immediately before and after the one when the three-dimensional shape information is acquired once. A method may be used in which a total of three points of light intensity at points is approximated by a quadratic function, and the position of the vertex of the quadratic curve is set as the peak position. Hereinafter, it demonstrates in detail using FIGS. 4-5.

  In the confocal scanning microscope, the change in the light intensity when the relative position between the objective lens and the sample is continuously changed is disclosed in, for example, “THEORY AND PRACTICE OF SCANNNING OPTICAL MICROSCOPY” (P126). It is theoretically calculated as follows.

I (Z) = {sin (u / 2) / (u / 2)} ² (1)
u = 8πZ · sin ² (θ / 2) / λ
NA = sinθ
Z: distance from the focal point I (Z): light intensity at position Z NA: numerical aperture of objective lens λ: wavelength of light (1) According to equation (1), Z = 0, that is, the sample is at the focal position of the optical system In some cases, the reflected light intensity I (Z) from the sample becomes maximum. When the equation (1) is approximated by a quadratic function of Z,
I (Z) = αZ ² + βZ + γ (2)
It is expressed.

FIG. 4 is an explanatory diagram of a height estimation method in a modification of the present embodiment. The solid line shown in the figure is the same as the IZ curve shown in FIG. 3, and I is sampled at the position indicated by ◯ in the figure. As shown in the drawing, in performing the acquisition of sequential light intensity information indicates the position where the output value I of the photodetector 40 (Z) becomes the maximum at Z _m, 1 previous relative Z _m The position where the measurement was performed is represented by Z _m−1 , and the position where the measurement was performed one after is represented by Z _{m + 1} . Then, by substituting the values of the output values I (Z) at the respective positions as I _m , I _m−1 , and I _{m + 1} into the equation (2),
I _m = αZ _m ² + βZ _m + γ (3)
I _m-1 = αZ _m-1 ² + βZ _m-1 + γ
I _{m + 1} = αZ _{m + 1} ² + βZ _{m + 1} + γ
It becomes.

Here, the movement start position in the Z direction when the relative position of the objective lens and the sample is discretely moved is Z ₀ , the movement amount in the Z direction per time is ΔZ, and the output of the photodetector is the maximum position If the number of movements in the Z direction is m,
Z _m = Z ₀ + ΔZ · m (4)
Z _m-1 = Z _m −ΔZ
Z _{m + 1} = Z _m + ΔZ
It becomes. Therefore, when α is obtained from the equations (3) and (4),
α = {(I _m−1 −I _m ) + (I _{m + 1} −I _m )} / (2 · ΔZ ² ) (5)
It becomes.

Here, since I _m is the maximum value of the output I of the photodetector 40, the relationship is I _m > I _m−1 and I _m > I _{m + 1} . Also, α <0 from equation (5). Therefore, I (Z) is maximized when Z = −β / 2α. Therefore, the position Z _p where the light intensity is maximum is
Z _p = −β / 2α
= Z ₀ + ΔZ · m + {(I _{m + 1} −I _m−1 ) · ΔZ} / [2 {(I _m−1 −I _m ) + (I _{m + 1} −I _m )}] 6)
It becomes.

Therefore, the objective lens that maximizes the light intensity when the IZ curve is approximated by a quadratic function from I _m , I _m−1 , I _{m + 1} , m, Z ₀ , and ΔZ based on the equation (6). 32 and the relative position Z _p of the sample 36 can be obtained.

In this modification, when R three-dimensional shape information is acquired, R _Zp are calculated for each pixel. Then, an average value of R Z _p is obtained, and that value is finally used as the three-dimensional shape information of the sample. For example, if three-dimensional shape acquisition is performed three times, the height estimation as described above is performed three times for each pixel. This will be described with reference to FIG.

FIG. 5 shows an IZ curve approximated by a plurality of quadratic functions and its peak position in a modification of the present embodiment. As shown in FIG. 5, Z _p1 is calculated as the height estimation position for the first acquisition of three-dimensional shape information, Z _p2 is calculated for the second acquisition, and Z _p3 is calculated for the third acquisition. And these average values are calculated | required as final three-dimensional shape information Zt. That is, Z _t = (Z _p1 + Z _p2 + Z _p3 ) / 3.

  In such a case, an effect of further reducing the Z scanning quantization error can be obtained as compared with the first embodiment. Therefore, it is possible to obtain more accurate three-dimensional shape information with the same number of movements.

<Second Embodiment>
Next, a confocal scanning optical microscope system according to the second embodiment will be described. In the present embodiment, predetermined three-dimensional shape numerical information is generated based on acquisition instruction information registered in advance by an observer from repeatedly acquired three-dimensional shape information. However, the three-dimensional shape numerical information is one-dimensional numerical information obtained from the three-dimensional shape information, for example, various parameters such as line width, step, or roughness at a predetermined position of the sample. Since the configuration of this embodiment is the same as that of the first embodiment, the description thereof is omitted.

  FIG. 6 shows a flow of measurement operation in the present embodiment. First, the observer acquires the three-dimensional shape information of the sample by a normal method. That is, instead of repeatedly performing the Z operation as described in the first embodiment, the Z-scan is performed once to acquire the three-dimensional shape information of the sample, and the three-dimensional image is displayed on the monitor 46 (S11). .

  Next, the observer uses an acquisition instruction means (for example, a mouse or the like) that is not shown in the computer 28, and uses the 3 dimensional shape information of the sample displayed on the monitor 46, such as the line width and step of a predetermined part. Specifies the position from which to obtain the dimension shape numerical information. This will be described with reference to FIGS. 7A and 7B.

  FIG. 7A shows an example of three-dimensional shape information in the present embodiment. FIG. 7B shows the predetermined cross-sectional information of FIG. 7A. For example, consider a case in which three-dimensional shape information as shown in FIG. 7A is acquired and displayed on the monitor 46. The observer designates the cross section as shown in FIG. 7B using the acquisition instruction unit as a position for measuring the period w of the periodic uneven shape or the height h from the valley to the mountain. Then, the three-dimensional shape information calculation means 42 stores the position designated by the observer as measurement acquisition instruction information. In this case, the direction and position of the cross section designated by the observer, the coordinates at which the period w and the height h are measured, and the measurement items are stored. Then, the observer designates the repeated acquisition count R (S12).

  Next, the operations from S2 to S9 shown in the first embodiment are executed (S13). That is, the three-dimensional shape information is acquired for the repeated acquisition times R specified by the observer. At this time, similarly to the first embodiment, the stage initial position is shifted by Δ / R in each Z scan every time the measurement is repeatedly performed.

  Next, based on the measurement acquisition instruction information set using the acquisition instruction means in S12 for the R pieces of three-dimensional shape information acquired in S13, R sets of three-dimensional shape numerical information are obtained for the number of repeated acquisitions. Is generated (S14).

Then, an average value of the generated three-dimensional shape numerical information is generated (S15). This will be described with reference to FIG.
FIG. 8 shows an example of a table relating to the three-dimensional shape numerical information in the present embodiment. The table is stored in the memory of the three-dimensional shape information calculation means 42. For example, the R sets of three-dimensional shape numerical information acquired in S14 are averaged, and the average value is generated as final three-dimensional shape numerical information and stored in the table. At this time, statistical information such as standard deviation can be simultaneously generated for the R sets of three-dimensional shape numerical information.

  As described above, in this embodiment, one piece of three-dimensional shape information is first acquired, and a position at which the three-dimensional shape numerical information is acquired is specified for the image using the acquisition instruction unit. Then, the three-dimensional shape information for the specified repeated measurement is acquired, and for each of the acquired three-dimensional shapes, the three-dimensional shape numerical information at the position set by the acquisition instruction means is acquired, and the average value is obtained. Is generated. As a result, it is possible to reduce the accidental error and the quantization error caused by performing Z scanning discretely, and also realize a measurement method with good operability for the observer.

  In the first and second embodiments, the stage 34 is moved as the Z direction moving method. However, the present invention is not limited to this, and other methods may be used. For example, the objective lens 32 may be moved, or the entire confocal optical system 16 may be moved.

  In the second embodiment, the three-dimensional shape information is acquired in S1 and the subsequent processing is performed based on the acquired three-dimensional shape information. However, even if the previously stored three-dimensional shape information is read and the subsequent processing is performed. good.

  In the first and second embodiments, a confocal microscope is used. However, the present invention is not particularly limited as long as the microscope can move in the Z direction and measure a three-dimensional shape. For example, a white interference microscope that can move in the Z direction may be used.

  According to the present invention, the relative position between the objective lens and the sample is sequentially changed, Z scanning for obtaining the maximum light intensity information of the sample and the relative position at that time is repeated a predetermined number of times, and the initial stage position is shifted. The position of taking in each light intensity information can be shifted from each other. As a result, the average value of the three-dimensional shape information or the average value of the three-dimensional shape numerical information output for each of the three-dimensional shape information is calculated by performing a random error and Z scanning discretely. It becomes possible to reduce the quantization error.

1 shows a configuration of a confocal scanning optical microscope system in a first embodiment. The operation | movement flow in the repeated measurement mode in 1st Embodiment is shown. The relationship between the position Z in the Z direction and the light intensity I output from the photodetector 40 when the position is fixed in the XY direction and the position is moved discretely in the Z direction in the first embodiment is shown. It is explanatory drawing of the estimation method of the height in the modification of 1st Embodiment. The IZ curve approximated with the quadratic function for several times in the modification of 1st Embodiment, and its peak position are shown. The flow of the measurement operation | movement in 2nd Embodiment is shown. An example of the three-dimensional shape information in 2nd Embodiment is shown. The predetermined cross-sectional information of FIG. 7A is shown. An example of the table regarding the three-dimensional shape numerical information in this embodiment is shown. 1 shows a schematic configuration of a general confocal microscope.

Explanation of symbols

1 point light source 2 half mirror 3 objective lens 4 sample 5 pinhole 6 photodetector 16 confocal optical system 18 laser light source 20 mirror 22 half mirror 24 two-dimensional scanning mechanism 26 scan control unit 28 computer 30 revolver 32 objective lens 36 sample 38 Pinhole 40 Photodetector 42 Three-dimensional shape information calculation means 44 Z direction movement control circuit 46 Monitor

Claims (8)

A three-dimensional shape measuring apparatus for measuring a three-dimensional shape of a specimen from a plurality of two-dimensional images picked up by a microscope capable of moving the position of the specimen relative to an objective lens at least in the optical axis direction. ,
Moving means for moving the sample position relative to the objective lens relative to the optical axis;
Movement control means for controlling the movement means to move the sample position relative to the objective lens in a discrete manner;
Number of times acquisition means for acquiring the number of times of generating the three-dimensional shape information of the specimen;
Obtaining discrete light intensity information of observation light from the specimen according to the change in the specimen position, obtaining the specimen position for obtaining the maximum light intensity based on the light intensity information, obtaining the maximum intensity Three-dimensional shape information generating means for performing the number of times the processing for generating the three-dimensional shape information based on the sample position is acquired by the number of times acquisition means;
With
When the three-dimensional shape information is generated by the number of times acquired by the number-of-times acquisition unit, the movement control unit mutually obtains the sample positions for acquiring the light intensity information discretely every time the three-dimensional shape information is generated. The three-dimensional shape measuring apparatus, wherein the moving means is controlled so as to be different from each other. The three-dimensional shape information generating means generates the number of times of three-dimensional shape information generated by making the sample positions from which the light intensity information is discretely obtained different for each generation of the three-dimensional shape information. The three-dimensional shape measuring apparatus according to claim 1, wherein the three-dimensional shape information of the specimen is averaged. The three-dimensional shape measuring apparatus further includes:
Measurement object acquisition means for acquiring setting information of a predetermined measurement object set based on the three-dimensional shape information generated in advance;
Measurement means for measuring a three-dimensional shape corresponding to the measurement object from the three-dimensional shape information obtained by the three-dimensional shape information generation means based on the setting information of the measurement object;
The three-dimensional shape measuring apparatus according to claim 1, comprising: The movement control unit is configured to determine a deviation amount when the sample positions for discretely acquiring the light intensity information are different for each generation of the three-dimensional shape information according to the number of times acquired by the number of times acquisition unit. The three-dimensional shape measuring apparatus according to claim 1, wherein the three-dimensional shape measuring apparatus is changed. The three-dimensional shape measuring apparatus according to claim 4, wherein the deviation amount is calculated by dividing one movement step amount by the movement control unit by the number of times acquired by the number-of-times acquisition unit. . The shift amount when the position where the light intensity information is discretely acquired by the movement control unit is different for each generation of the three-dimensional shape information can be arbitrarily set. The three-dimensional shape measuring apparatus as described. The three-dimensional shape information generating means extracts a predetermined number of light intensity information from the plurality of light intensity information when acquiring the sample position for obtaining the maximum light intensity, and extracts the predetermined number of light intensity information. The maximum value on a suitable change curve is calculated, the sample position corresponding to the maximum value is estimated, and the estimated sample position is acquired as three-dimensional shape information. Three-dimensional shape measuring device. Method of operating a three-dimensional shape measuring apparatus for measuring a three-dimensional shape of a specimen from a plurality of two-dimensional images picked up by a microscope capable of moving the position of the specimen relative to the objective lens at least in the optical axis direction Because
Relatively moving the sample position relative to the objective lens in the direction of the optical axis;
Obtain the number of times to generate the three-dimensional shape information of the sample,
Obtain discrete light intensity information of the observation light from the specimen according to the change in the specimen position, obtain the specimen position for obtaining the maximum light intensity based on the light intensity information, and based on the specimen position The process of generating the three-dimensional shape information is performed the acquired number of times,
When generating the three-dimensional shape information for the acquired number of times, the sample position is moved so that the sample positions at which the light intensity information is discretely acquired are different for each generation of the three-dimensional shape information. A method for operating a three-dimensional shape measuring apparatus.