JP2012255756A - Three-dimensional measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a three-dimensional measuring method for measuring the shape of an object to be measured with high accuracy.SOLUTION: In accordance with the height of an alignment mark 29 on a wafer, the focus height of a camera 8 is adjusted by a Z2-axis stage 9. In accordance with the focus height, the height of an alignment mark 10 for calibration is adjusted by a Z3-axis stage 11. In such a state, a center position of the alignment mark 10 for calibration is measured by the camera 8 and a probe 1, and distance offset (Xo, Yo) of two of the camera 8 and the probe 1 is determined with high accuracy. By using the offset, coordinates of the alignment mark 29 on the wafer measured by the camera 8 are transformed into a coordinate system of the probe 1 with high accuracy. A lens on the wafer is measured by the probe 1, thereby highly accurately determining a lens center while defining a position of the alignment mark 29 on the wafer as a criterion.

Description

  The present invention relates to a three-dimensional measurement method for measuring a shape using a probe or the like. In particular, the present invention is an alignment for determining a position on a wafer using a camera provided in a three-dimensional measuring machine in measurement of a wafer lens or the like in which a plurality of lenses are formed on the upper and lower surfaces of a wafer that transmits light. The present invention relates to a three-dimensional measurement method in which a mark is measured and a lens center position measured by a stylus or the like is obtained with high accuracy with respect to an alignment mark measured by this camera.

  Camera functions have been added to mobile devices such as mobile phones and PDAs, and the need for small and inexpensive cameras has increased dramatically. In addition, the performance of cameras used in these devices is also highly demanded for higher pixel count. Conventionally, lenses used in these mobile devices are produced by attaching 10 to 15 lens molds to a molding machine and producing 10 to 15 lenses per molding shot. It is not easy to respond to the increase in cost and reduce costs. Therefore, in these fields where cost reduction is required and the number of production is rapidly increasing, several hundred to several thousand lenses are molded on a single wafer, and a plurality of the molded wafers are laminated. Therefore, there is a need for a technique for manufacturing a wafer lens that is cut out and manufactured after lamination.

  The structure of the wafer lens is shown in FIG. Lenses 102 are molded on the upper and lower surfaces of the wafer 101, and a plurality of positioning alignment marks 103 are formed around the upper surface. The wafers 101 are stacked so that the alignment marks 103 overlap with each other, and the lens unit is completed.

  When the center position of the lens 102 is shifted between the front and back surfaces of one wafer 101 constituting the lens unit by stacking, or in the lens unit in which a plurality of wafers 101 are stacked, the lens center is based on the alignment mark on each wafer 101. When the position is deviated, the image is blurred and problems such as inability to obtain optical performance occur.

  On the other hand, Patent Document 1 discloses a method of aligning a glass substrate and a master for photo-curing resin molding when manufacturing this type of wafer lens. FIG. 14 shows the procedure of the alignment method described in Patent Document 1.

In FIG. 14, reference numeral 104 denotes an alignment mark of the master 109, and reference numeral 105 denotes an alignment mark of the glass substrate 107. Reference numeral 106 denotes a camera that measures the position of the alignment mark.

  The height of the camera 106 is raised and lowered to focus on the alignment marks 104 and 105 on the upper and lower surfaces, and the master 109 and the glass substrate 107 are overlapped so that the XY positions of the marks coincide.

  Specifically, as shown in FIG. 14, the camera 106 that can move only in the vertical direction focuses on the alignment mark 105 from above the glass substrate 107 (see (1) in FIG. 14). Thereafter, the camera 106 is moved upward to place the master 109 at a position between the camera 106 and the glass substrate 107, and while adjusting the height position of the camera 106, the focus position is adjusted to the alignment mark 104 of the master 109 or Match the vicinity (see (2) in FIG. 14).

  In this case, for example, if it is assumed that the alignment mark 105 whose focus position has been adjusted first and the alignment mark 104 whose focus position has been adjusted thereafter are in the state shown in the upper part of FIG. The master 109 is moved in the horizontal direction to a position where the alignment mark 104 matches the mark 105 (see the lower part of FIG. 15), and in this state, the master 109 is pressed against the glass substrate 107 previously coated with a photo-curing resin. (See (3) in FIG. 15), the lens is formed by light irradiation.

JP 2010-72665 A

  As described above, the deviation of the center position of the upper and lower lenses 102 of the wafer 101 in the wafer lens and the deviation of the lens center position on the basis of the alignment mark on the individual wafer 101 at the time of stacking deteriorate the optical performance. In order to eliminate or reduce these deviations, it is required to measure the lens center position with reference to the alignment mark with the highest possible accuracy.

  However, even if the alignment method described with reference to FIGS. 14 and 15 is applied to the measurement of the lens center position with reference to the alignment mark, it is difficult to measure with high accuracy. Hereinafter, this point will be described.

  By a moving mechanism (not shown) that moves the camera 106 in the Z direction on the wafer surface, the focus position of the alignment mark 105 on the front surface of FIG. 14 (1) is changed to the focus position of the alignment mark 104 on the back surface of FIG. When the Z direction of the camera is moved, a stage (not shown) that moves the camera 106 in the Z direction causes a positional shift in the X direction or the Y direction perpendicular to the paper surface. Therefore, if the Z height of the camera is changed, the alignment mark position cannot be measured with high accuracy.

  When measuring the lens center positions on the front and back surfaces of one wafer 101 shown in FIG. 13, for example, the posture of the wafer 101 is first set so that the surface on which the alignment mark 103 is provided is the upper surface. After the Z height is adjusted so that the camera 106 is focused on the upper surface of the wafer 101 and the alignment mark 103 is measured, the center position of the lens is measured with a probe based on the alignment mark. After that, the posture of the wafer 101 is turned upside down, and the Z height of the camera is changed to the Z height position of the alignment mark 103 where the position has changed on the lower surface of the wafer 101 so as to focus. taking measurement.

  When the camera 106 is moved in the Z direction on the wafer surface by a moving mechanism (not shown), a stage (not shown) that moves the camera 106 in the Z direction causes a positional shift in the X direction or the Y direction perpendicular to the paper surface. An error occurs in that the center position of the X direction or the Y direction slightly shifts. As a result, the lens center position measured by the probe with the measured alignment mark reference is shifted between the front surface and the back surface. In this way, when calculating the front and back lens centers based on the alignment mark reference, the center of the front and back surfaces of the camera 106 can be accurately determined by a single wafer 101 due to a slight deviation in the center position of the camera 106 in the X or Y direction. It cannot be measured.

  The present invention solves the above-described conventional problems, and an object thereof is to provide a three-dimensional measurement method for measuring the shape of a measurement object with high accuracy.

  In order to achieve the above object, the three-dimensional shape measuring method of the present invention is a three-dimensional measuring method for acquiring surface shape data of a measured object by two or more surface detection means, and is provided separately from the measured object. The calibration alignment mark whose position in the XY direction does not move with respect to the object to be measured is made to coincide with the surface height of the object to be measured so that the height of the alignment mark for calibration is the same. Measured by the two or more surface detection means, calibrated offsets in the XY direction of the two or more surface detection means using the measurement results, and measured by the two or more surface detection means The surface shape of the object to be measured is obtained using the surface shape data of the object to be measured and the calibrated offset.

  Preferably, the surface detection means includes a stylus for surface shape measurement provided on the first Z direction moving means that moves in the Z direction on an XY stage that moves in the XY direction, and a Z direction on the XY stage. A camera for measuring an image in the XY plane provided in the moving second Z-direction moving unit, and photographing the calibration alignment mark matched with the surface height of the object to be measured with the camera. Measure with the stylus, calibrate the offset of the center position of the stylus and the camera from the results of the photographing and measurement, measure the surface shape of the object to be measured with the stylus, and measure the measurement result with the stylus and the calibration The surface shape of the object to be measured is obtained using the offset.

  More preferably, the measurement of the surface shape of the object to be measured by the stylus is collectively measured in the X direction of the lens surface so as to pass near the apex position of each lens surface on the object to be measured. The measurement data is measured at a time in the Y direction so as to pass through the vicinity of the vertex position of the lens, and the measurement data is divided for each lens according to the preset pitch in the X direction and the Y direction of the lens from the measurement data. For the data divided into, the shape, the XYZ position and the posture of the lens center are evaluated.

  For example, the object to be measured is a wafer lens in which a large number of lenses are formed on a thin plate.

  According to the three-dimensional measurement method of the present invention, the shape of the object to be measured can be measured with high accuracy. For example, the lens center positions on the front and back surfaces of the wafer lens can be measured with high accuracy on the basis of the alignment mark. In other words, even when an alignment mark is formed on the back side of the wafer and there is no unevenness on the front side of the wafer, the center of the lens on the wafer can be measured using the alignment mark that can only be measured by the camera on the wafer. Even when measuring the lens on the wafer with the alignment mark reference for assembly with high accuracy by measuring the lens position with the alignment mark reference using the calibrated offset value between the center position of the camera and the stylus. The position can be measured.

  This allows the lens on the wafer to measure the lens shape of each of the upper and lower surfaces of the wafer and calculate the positional deviation based on the alignment mark, thereby correcting the positional deviation of the optical axis center of the lens on the wafer based on the alignment mark. It becomes possible to measure with high accuracy.

  Furthermore, since the deviation of the lens center axes on the upper and lower surfaces from the alignment mark can be calculated individually, the deviation of the lens center axes on the upper and lower surfaces can be measured with high accuracy.

  Furthermore, the three-dimensional shape can be evaluated with high accuracy and high speed by separating the X- and Y-direction batch measurement data passing through the vicinity of the apex position of each lens surface on the object to be measured into individual lens data. be able to.

BRIEF DESCRIPTION OF THE DRAWINGS The whole apparatus of the three-dimensional measuring method in Embodiment 1 of this invention. The probe block diagram of the three-dimensional measuring method in Embodiment 1 of this invention. FIG. 3 is a detailed view of a three-dimensional measurement method according to Embodiment 1 of the present invention. 1 is a configuration diagram of a wafer lens and an alignment mark in Embodiment 1 of the present invention. The block diagram of the alignment mark for a calibration in Embodiment 1 of this invention. The whole measurement flow in Embodiment 1 of this invention. FIG. 3 is a cross-sectional view of an edge portion of the calibration alignment mark in the first embodiment of the present invention. FIG. 5 is a detailed positional relationship diagram of alignment marks in the first embodiment of the present invention. FIG. 5 is a detailed diagram of calibration alignment mark measurement according to Embodiment 1 of the present invention. FIG. 3 is a detailed view of the positional relationship of the wafer lens according to the first embodiment of the present invention. FIG. 3 is a detailed view of a wafer lens measurement scanning path according to the first embodiment of the present invention. FIG. 3 is a detailed diagram of measurement data obtained by a probe in Embodiment 1 of the present invention. FIG. 5 is a detailed view of lens center position calculation in Embodiment 1 of the present invention. 1 is a configuration diagram of a wafer lens described in Patent Document 1. FIG. The figure which shows the conventional three-dimensional measuring method described in patent document 1. FIG. The figure which shows the conventional three-dimensional measuring method described in patent document 1. FIG.

  Embodiments of the present invention will be described below with reference to the drawings.

(Embodiment 1)
1) Explanation of Device Configuration of CMM Used FIG. 1 is a perspective view showing a schematic configuration of a shape measuring device as an embodiment for carrying out the shape measuring method of the present invention. In FIG. 1, a frequency-stabilized He-Ne laser 4 for measuring an XYZ coordinate position is disposed on a stone surface plate 14 on an XY stage 3, and the probe 1 is connected to the stone surface plate 14 via the Z1 axis stage 2. Is reflected on the X reference mirror 5, the Y reference mirror 6 and the Z reference mirror 7 having a high flatness of nanometer order fixed by an oscillation frequency stabilized He-Ne laser beam. XYZ coordinates can be measured with extremely high accuracy.

  These units are controlled by a computer 41 as a control calculation unit, and are configured to perform automatic operation in combination with a three-dimensional measuring machine.

  Further, the shape measuring apparatus calculates the offset distance between the camera 8 that measures the alignment mark position, the Z2 axis stage 9 that moves the camera 8 in the Z-axis direction, the center position of the tip of the probe 1, and the center position of the image of the camera 8. A calibration alignment mark 10 to be calibrated, a Z3-axis stage 11 for moving the calibration alignment mark 10 in the Z-axis direction, a wafer lens 12 as a measurement object, and a wafer chuck 13 on which the wafer lens 12 is installed. . In FIG. 1, the clockwise rotation in the positive direction of the X axis is A, the clockwise rotation in the positive direction of the Y axis is B, and the clockwise rotation in the positive direction of the Z axis is C.

  FIG. 2 shows the configuration of the probe used in this shape measuring apparatus. In FIG. 2, the stylus 18 is supported by a micro air slider 15, and the movable part of the micro air slider 15 is supported by a micro spring 16. The deflection of the microspring is measured by the focus detection laser 19 irradiated on the mirror of the micro air slider 15 so that the weak atomic force acting on the tip of the measurement object 17 and the stylus 18 is constant. The entire position in the Z direction is feedback controlled by a linear motor (not shown). At the same time, the position in the Z direction is measured by irradiating the mirror 22 with the frequency-stabilized He-Ne laser light 21 for the displacement in the Z direction. In this state, the entire probe unit is scanned in the XY directions to measure the shape of the measurement surface. With this configuration that can reduce the weight of the movable part of the micro air slider 15, which is a movable part to which the stylus 18 is attached, it is possible to measure with high accuracy with nanometer accuracy up to a highly inclined surface of a maximum of 75 °.

  This shape measuring apparatus obtains an XYZ coordinate data string on the measurement object by scanning the probe 1 in the X and Y directions on the surface of the measurement object 17, and measures the probe 1 by the computer 41 as a control calculation unit (not shown). The column of the Z coordinate data at the XY coordinate position is calculated and the shape of the measurement object 17 is measured.

2) Principle of Measurement Method FIG. 3 is a principle diagram showing the principle of the measurement method of the present invention. In FIG. 3, a Z1-axis stage 2 that drives the probe 1 in the Z-axis direction in order to perform measurement with the Z-direction probe 1 with high accuracy is installed on an XY stage 3 that scans the probe 1 in the XY direction. The XY stage 3 is provided with a Z2 axis stage 9 for driving the alignment mark measuring camera 8 in the Z axis direction. The alignment mark measuring camera 8 is installed at a position offset in the XY direction with respect to the probe 1. The distance (inter-center distance) Xo between the center of the probe 1 and the center of the camera 8 due to this offset is a slight amount of the position in the XY direction due to the movement of the camera 8 in the Z-axis direction by the position on the Z2-axis stage 9 Changes due to deviation.

  Furthermore, a wafer chuck 13 on which a wafer lens 12 (see also FIG. 4), which is a measurement object of a circular substrate having a diameter of about 200 mm, is installed, a γ stage 23 that rotates the entire wafer chuck 13 around the Z axis, and a wafer It is constituted by a recognition device (not shown) for observing an alignment mark 29 for specifying an XY position on the wafer lens 12 installed on the chuck 13 and detecting a pattern of the mark.

  Further, a calibration alignment mark 10 for calibrating the center position offset between the measurement probe 1 and the camera 8 is installed on the Z3-axis stage 11 movable in the Z direction on the stone surface plate 24 of FIG. As will be described later in detail, the calibration is performed by measuring the calibration alignment mark 10 with the probe 1 and the camera 8.

  A wafer lens 12 that is an object to be measured is shown in FIG. On the wafer lens 12, lenses 34 are formed in a lattice pattern at predetermined substantially constant pitches in the X and Y directions. Specifically, the lens 34 is formed at the same position on both the A surface and the B surface of the wafer lens 12. Further, two or more alignment marks 29 are formed at predetermined positions on the A surface of the wafer lens 12.

  5A and 5B show the calibration alignment mark 10. In the present embodiment, the calibration alignment mark 10 uses a chromium mask substrate for semiconductor manufacturing in which a chromium film 26 of about 0.1 μm is deposited on a glass substrate 25. In the center of the chromium film 26, a square-shaped etching portion 27 having a size of about 1 mm square in the vertical and horizontal directions is provided. In this etching part 27, the chromium film 26 is removed and the glass substrate 25 is exposed.

3) Outline of Measurement Method FIG. 6 shows an overall flow of measurement by the three-dimensional measurement method of the present invention.

  Measurement is started in step 201, and a recipe which is a kind of computer program indicating a procedure for automatically performing a series of measurements is set in the computer 41. The recipe includes the design shape of the wafer lens 12 to be measured, the wafer size, the pitch between the lens centers in the X and Y directions in the design of the lens 34 on the wafer lens 12, the lens arrangement on the wafer lens 12, and the probe 1 at the time of measurement. The measurement speed, the position of the two alignment marks 29 and other information necessary for measurement are input, measurement conditions are set and measurement preparations are performed, and then measurement is started.

  The wafer lens 12 to be measured in step 202 is set on the wafer chuck 13 of the shape measuring apparatus so that the A surface (the surface on which the alignment mark 29 is provided) side is up. The wafer lens 12 is manually set in alignment with the wafer chuck 13 of the shape measuring device, and is sucked by a vacuum suction mechanism (not shown).

  Thereafter, in step 203, the XY position of the camera 8 is moved by a drive mechanism (not shown) of the XY stage 3 so as to be positioned above the alignment mark 29 on the A surface of the wafer lens 12 installed in step 202. Thereafter, the Z2 axis stage 9 for adjusting the camera height is set so that the focus height of the camera 8 becomes the height L1 of the alignment mark 29 on the wafer on the surface of the wafer lens 12 as the object to be measured shown in FIG. Are adjusted in the Z direction while monitoring the image of the camera 8 on the monitor (not shown) of the computer 41.

  In step 204, the camera 8 is moved by the XY stage 3 so as to be positioned above the calibration alignment mark 10 while maintaining the focus height position of the camera 8 adjusted in step 203. Next, the height of the Z3-axis stage 11 is adjusted so that the camera 8 is focused on the calibration alignment mark 10 while maintaining the state where the camera 8 is fixed and is not moved in any of the XYZ directions. As a result of this adjustment, the height L2 of the mark surface of the calibration alignment mark 10 coincides with the height L1 of the alignment mark 29 on the surface (A surface) of the wafer lens 12 that is the object to be measured.

  In step 205, the calibration alignment mark 10 is measured by the camera 8 adjusted in the procedure of steps 201 to 204, and the calibration alignment mark 10 is also measured by the probe 1, and the probe 1 is measured based on the measurement result. Measurement of the center-to-center distance Xo of the camera 8 (calibration of the center position of the probe 1 and the camera 8) is executed.

  In step 206, using the camera 8 and the XY stage 3, the positions of the two alignment marks 29 provided on the wafer lens 12 are measured, and the center positions of the alignment marks 29 are stored. The rotational deviation angle γ of the wafer lens 12 with respect to the coordinates (fixed XYZ coordinate system) of the shape measuring apparatus is measured from the two measured alignment mark positions.

  In step 207, the wafer lens 12 is rotated by the measured rotational deviation angle γ by the γ stage 23 provided below the wafer chuck 13 so as to be parallel to the X reference mirror 5 and the Y reference mirror 7 of the shape measuring apparatus. adjust.

  In step 208, it is confirmed whether the rotational deviation angle γ is within a specified range with respect to a predetermined value set in advance. If it is within the range, the process proceeds to step 209. If it is out of the range, the procedure from step 206 is repeated.

  In step 209, the entire area of the A surface of the wafer lens 12 is measured in two strokes in the X direction and the Y direction in a single stroke so that the probe 1 passes through the center of each lens on the wafer lens 12.

  In step 210, the measurement data in the two directions X and Y measured in step 209, the positions of the two alignment marks 29 obtained in step 206, and the centers of the probe 1 and camera 8 obtained in step 205 are obtained. Using the distance, the center of each lens 34 on the wafer lens 12 is calculated based on the alignment mark 29.

  In steps 211 to 219, the same processing is repeated for the B surface of the wafer lens 12.

  First, in step 211, the wafer lens 12 measured in the above step is attached to the wafer chuck 13 so that the front and back of the wafer lens 12 are reversed with the Y axis as a rotation axis and the B surface is on the upper side.

  In step 212, the focus height of the camera 8 is adjusted by the Z2 axis stage 9 to the position of the alignment mark 29 located on the lower side.

  In step 213, the Z3-axis stage 11 is adjusted such that the focus of the camera 8 is aligned with the calibration alignment mark 10 while maintaining the focus height position of the camera 8 adjusted by the Z2-axis stage 9 in step 212.

  In step 214, the calibration alignment mark 10 is used to measure the center-to-center distance XoB between the center position of the camera 8 and the probe 1 slightly shifted in the XY direction as a result of the adjustment in the procedure of steps 211 to 213. .

  In steps 215 to 218, processing similar to that in steps 206 to 209 is performed in the installation state.

  In step 219, the measurement data in the two directions X and Y measured in step 218, the positions of the two alignment marks 29 obtained in step 215, and the center of probe 1 and camera 8 obtained in step 214. Using the distance, the center of each lens 34 on the wafer lens 12 is calculated based on the alignment mark 29.

  In step 220, the center deviation of the front and back lenses 34 of the wafer lens 12 is calculated from the center position of each lens 34 on the A surface and the center position of each lens 34 on the B surface.

  In step 221, the measurement ends.

2) Details of Measurement Method Details of the procedure described in FIG. 6 will be sequentially described below.

  In step 205, the positional relationship between the alignment mark 29 on the wafer lens 12 adjusted in the procedure of steps 201 to 204 and the calibration alignment mark 10 is used to determine the distance between the center of the probe 1 and the camera 8 using the calibration alignment mark 10. Measure the distance XoA.

  The details of the calibration procedure for calibrating the center positions of the probe and the camera using the calibration alignment mark will be described below.

  First, the center position (Xc, Yc) of the calibration alignment mark 10 is calculated by the camera 8 by the following procedure.

  After the adjustment of the Z3-axis stage 11 in step 204, the camera 8 is moved by the XY stage 3 to the mark center position of the calibration alignment mark 10.

  Thereafter, an image as shown in FIG. 7 is obtained by the camera 8 as the etching part 27 of the calibration alignment mark 10.

  Here, the calculation of the center position of the square-shaped etching part 27 is performed by the following procedures 1 to 5.

  Procedure 1: In the X-axis direction of the etching unit 27, intersection positions X1L and X1R with the etching unit 27 are obtained from the data of the X measurement line 30, and intersection points X2L and X2R with the etching unit 27 are obtained from the data of the X measurement line 31. Ask.

  Procedure 2: For the Y-axis direction of the etching part 27, the intersection positions Y1D and Y1U with the etching part 27 are obtained from the data of the Y measurement line 32, and the intersection positions Y2D and Y2U with the etching part 27 are obtained from the data of the Y measurement line 33. Ask.

  Procedure 3: Y-direction measurement center, which is an average line of the vertical line of the etching portion 27 connecting the intersection position X1L and the intersection position X2L and the vertical line of the etching portion 27 connecting the intersection position X1R and the intersection position X2R Line 35 is determined.

  Procedure 4: X-direction measurement center, which is an average line of a horizontal line of the etching part 27 connecting the intersection point position Y1D and the intersection point position Y2D and a horizontal line of the etching part 27 connecting the intersection point position Y1U and the intersection point position Y2U Line 34 is determined.

  Procedure 5: A mark center position (Xc, Yc) that is an intersection of the X direction measurement center line 34 and the Y direction measurement center line 35 is obtained.

The details of the calculation procedure of the mark center position (Xc, Yc), which is the center position of the etching portion 27, will be described below. In the following calculation, a grayscale image of the chrome film 6 including the entire etching portion 5 is extracted from the CCD data of the camera 8.
Here, the measurement data is extracted with the two measurement cross sections on the X axis as the X measurement line 30 and the X measurement line 31. As shown in FIG. 8, these data have uneven shapes due to shading. The center portion of the shading of the shading image is set as a threshold, and the edge position is extracted, that is, the vector notation in the XY directions shown in FIG. In terms of position, the intersection position of the threshold of the X measurement line 30 is expressed as X1L = (X1Lx, X1y), X1R = (X2Lx, X2y). Similarly, threshold crossing position vectors in the X measurement line 32 are represented as X2L and X2R. Similarly, the intersection positions Y1U, Y1D, Y2U, and Y2D of a total of four thresholds of the Y measurement line 32 and the Y measurement line 33 in the two directions in the Y-axis direction are obtained.

  Thereafter, the midpoint is calculated from the following equation using the symbols shown in FIG.

  Further, the following vector XV is defined.

  If this vector XV is used and t is a scalar, the equation of the X measurement center line 34 is expressed by the following equation (1).

Similarly, the equation of the Y measurement center line 35 is expressed as the following equation (2) using the scalar quantity s.

  Since the mark center is the intersection of the lines 34 and 35 represented by equations XL and YL, the mark center can be obtained by calculating t and s from the equations (1) = (2).

  The equation XL, that is, the equation (1) and the equation YL, that is, the equation (2) can be transformed into the following equations (1) 'and (2)', respectively.

  The following formulas (3) and (4) are obtained from the formula (1) ′ = (2) ′.

  When the equations (3) and (4) are solved for t, the following equation (5) is obtained.

  Substituting the value of t obtained by Equation (5) into Equation (1), the intersection, that is, the mark center (Xc, Yc) = Xc = t * XV + X1C is obtained. The mark center (Xc, Yc) can also be obtained by substituting the value of s obtained by solving equations (3) and (4) for s into equation (2).

  The alignment mark position within the range of the camera image is calculated by detecting edges in each of the X-axis direction and the Y-axis direction by the above calculation procedure.

  In this way, by calculating the center of the edge position of the alignment mark 10 in the X direction and the Y direction, even if the alignment mark for calibration is attached slightly tilted with respect to the X or Y direction of the measuring machine, it is highly accurate. The center (Xc, Yc) of the calibration alignment mark can be obtained.

  Next, the probe is moved to the calibration alignment mark position, and the probe is moved along the same path as the four lines of the X measurement line 30, the X measurement line 31, the Y measurement line 32, and the Y measurement line 33 measured by the camera. Then, two lines are scanned in each of the X direction and the Y direction, and the heights of the X measurement line 30, the X measurement line 31, the Y measurement line 32, and the Y measurement line 33 of the step shape portion of the calibration mark in each direction are measured. Then, the intersection positions X1L, X1R, X2L, X2R, Y1U, Y1D, Y2U, and Y2D of the threshold are sequentially calculated. Using this data, two X measurement center lines 34 and Y measurement center lines 35 passing through the midpoints of the intersection positions in the X direction and the Y direction are calculated in the same manner as the calculation procedure of the camera, and the calculated X direction Y The intersection coordinates of the center line of the two straight lines are calculated, and the alignment mark position (Xa, Ya) in the probe measurement is calculated by the same calculation formula as the calculation procedure in the camera.

  From these measurement results, the camera offset position (XoA, YoA) = (Xc, Yc) − (Xa, Ya) on the probe reference when measuring the A surface of the wafer lens 12 is calculated.

  Thereafter, the offset position of the probe 1 with respect to the camera 8 in the lens measurement on the wafer lens 12 surface is referred to this value, and the offset value is subtracted to convert the alignment mark position obtained by the camera 8 into the probe reference. calculate.

  As described above, when the height of the alignment mark 29 on the wafer lens 12 changes, the same mark is placed on both the camera 8 and the probe 1 with the calibration alignment mark 10 adjusted to the alignment mark height on the wafer. By measuring, an accurate offset value between the camera 8 and the probe 1 can be obtained.

  As a result, even if the straightness in the XY direction of the Z height adjusting Z2 axis stage that is adjusted during the focus adjustment of the camera 8 is shifted, the influence of measurement errors such as an optical center position shift caused by the focus height shift of the camera is affected. It is possible to perform highly accurate measurement without receiving it.

  In FIG. 9, Xw and Yw are wafer lens coordinate systems. Xm and Ym are coordinate systems of the shape measuring apparatus in which the wafer lens 12 is installed. However, in this state, as shown in FIG. 9, the coordinate system of the Xw and Yw wafer lenses 12 and the coordinate system of the Xm and Ym shape measuring devices cannot be installed in parallel and there is a slight deviation. State.

  In step 206, the positions of the two alignment marks 29 provided on the wafer lens 12 are measured using the camera 8 and the XY stage 3, and the center positions XaA1 and XaA2 of the alignment marks are stored. From the measured two alignment mark positions, the rotational position deviation γ of the wafer lens 12 is measured with respect to the coordinates of the three-dimensional measuring machine. The calculation procedure can be calculated by the alignment mark center calculation procedure by the camera 8 described in step 205.

  In the following description, it is assumed that the two alignment marks 29 on the wafer are configured in the vicinity of the Y-axis as shown in FIG. The difference between the XY position vectors of the two alignment marks 29 is obtained. As shown in FIG. 10, when the distance in the Y direction between the two alignment marks 29 is YLd, and the deviation in the X direction between the two marks is dX, The rotational position deviation about the Z axis is calculated as γ = atan (dX / YLd).

  Therefore, in step 207, the wafer lens 12 is rotated by a γ stage 23 provided under the wafer chuck 13 for the measured γ deviation, and the XY coordinates of the wafer lens are the X reference mirror 5 and the Y reference mirror of the three-dimensional measuring machine. Adjust to be parallel to 7.

  In step 208, it is confirmed whether the deviation of the value of γ from a predetermined value is within a specified range. If it is within the range, the process proceeds to step 209. If it is out of the range, the procedure from step 206 is repeated.

  In step 209, the entire wafer lens is measured in two strokes in the X direction and the Y direction in a single stroke so that the probe 1 passes through the center of each lens on the wafer lens 12.

  From the measured positions of the two alignment marks 29, the center position of the wafer lens 12 is calculated, and an NC path of a measurement path for measuring all the lenses 34 to be measured in a single stroke in the X and Y directions is created. As shown in FIG. 10, the X direction, the Y direction, and the probe 1 are sequentially scanned in a single stroke so as to pass through the vicinity of the apex position of the lens 34 on the entire surface of the wafer lens 12 on the XY axis of all the lenses on the wafer. Obtain measurement data.

  The arrangement of the lenses 34 of the wafer lens 12 measured here needs to be arranged in a lattice pattern at equal intervals in the X direction and the Y direction. Measurement is performed by continuously scanning the lenses 34 arranged in a lattice pattern. The design formulas for all lens shapes are the same.

  In step 210, the measurement data in the two directions X and Y measured in step 209, the positions XaA1 and XaA2 of the two alignment marks 29 obtained in step 206, the probe 1 and the center of the camera 8 are measured. Using the distance XoA, the center of each lens on the wafer lens 12 is calculated based on the alignment mark 29.

  This procedure will be described below. In the one-stroke writing form shown in FIG. 10, the entire lens surface is measured in the X direction, and the entire lens surface is measured in the Y direction. The measurement data in the X direction and the Y direction is divided for each lens based on the coordinates of the measuring machine according to the pitch in the X direction and the Y direction of the array.

  FIG. 9 shows the relationship between the lens position (i, j) and the indexes i and j indicating the lens position. The index of i, j is (i, j) = (0, 0), where i = + 1 for the first lens in the X + direction and j for the first lens in the Y + direction, where the center lens of the wafer lens is (i, j) = (0, 0). = + 1.

  FIG. 11 illustrates a data division procedure. For consecutive measurement data, a virtual center position X [i] = i * Xp is set for every i integral multiples of the design pitch, and a range of ± Xp / 2, which is a distance half the lens array pitch from this center. Is divided from the original one-stroke data as the i-th lens data.

  Similarly, a virtual center position Y [j] = j * Yp in the Y direction is set, and data in a range of ± Yp / 2, which is a distance half the lens arrangement pitch, is obtained from this center as data for the jth lens. Is divided from the original stroke data.

  Then, one set of data in the X direction and Y direction at the same intersection is stored in a memory of a computer (not shown) as measurement data for one lens. Using this divided data, the cut out data is aligned for each lens, and the center position, shape deviation, PV (Peak to Valley), RMS (Root Mean Square), etc., which are the maximum deviation from the design formula, are evaluated. .

  From the data obtained by measuring the XZ and YZ sections on the XY axes of the lens 34 using the data for each of the divided lenses 34, the center position of the lens with respect to the measuring machine, the measurement data at each measurement point, and the design In order to minimize the sum of squares of the shape difference, the measurement data point sequence is arranged in the X, Y, Z, A rotation direction around the X axis, and B rotation direction around the Y axis in the design coordinate system. The lens center position of all the lenses 34 on the wafer lens 12 is calculated by sequentially moving the movement amount for all the lenses 34 on the wafer lens 12. The design formula of each lens shape is the same on one side of the wafer surface.

  In FIG. 12, the measurement data point sequence (Xk, Yk, Zk) at the (i, j) position on the wafer and the design shape. In FIG. 12, for easy explanation, a measurement data point sequence in the XZ section and a design shape are displayed.

In FIG. 12, using the data of each point of the measurement data point sequence and the design shape represented by Z = f (x, y), Zfk is changed to Zfk = according to each position of (Xk, Yk) of the measurement data. f (Xk, Yk) was calculated from the difference between Zk of the measurement data and this value, ZDK = calculates Zk-Zfk, rotation of the Y-axis around which the square sum for all measured data, ShigumaZdk ² is minimum dBa, a translational movement amount dXa in the X-axis direction, and a movement amount dZa in the Z-axis direction are calculated.

  The description of FIG. 12 is an explanatory diagram on the XZ plane, but the same procedure is applied to all data of the XZ cross section and the YZ cross section, and the design shape Z = f (X (Xk, Yk, Zk) is used in the same procedure. Xk, Yk) is calculated as dXa, dYa, dZa, dAa, and dBa by calculating the amount of movement that minimizes the square sum of squares. , DYa is calculated. The sum of this deviation and the virtual center position determined in 4-4) when dividing according to each lens from the one-stroke writing data of the lens at the (i, j) position. The lens center position is sequentially calculated by changing the values of i and j for all the lenses, using the following formula.

  The lens center position data point sequence is not shown as (Xf [i, j], Yf [i, j]) (i and j are integer values representing the lens lattice position on the XY plane on the wafer). Store in computer memory.

  The measurement center of each lens was installed by rotating the γ-axis stage so that the coordinates of the alignment mark on the wafer were as close as possible to the coordinates of the measuring machine. Due to accuracy limitations, it is difficult to perform alignment by rotating the γ-axis so that the positional deviation dX in the X direction of the two alignment marks is 1 μm or less.

  Therefore, using the values of the alignment mark position deviations YL2 and dX2 measured by the camera, the calculated center position of all the lenses is set so that the alignment mark is used as a reference for all the lens center XY position data points. (Xf [i, j], Yf [i, j]) (where i and j are integer values representing the lens position of the lens on the XY plane on the wafer) are rotated by an angle γ2 = atan (dX2 / YLd2) and aligned. The lens center position based on the mark is calculated. According to the above procedure, the lens center position can be obtained with high accuracy with respect to the alignment mark coordinates obtained by the camera.

  Here, FIG. 9 shows the relationship between the lens center position (Xf [i, j], Yf [i, j]) and the indexes i and j indicating the lens position. The lens center position is represented by (Xf [i, j], Yf [i, j]) as a position based on the alignment mark on the wafer. In this array, the indices i and j are (i, j) = (0, 0) at the center of the wafer lens, i = + 1 for the first lens in the X + direction, and the first lens in the Y + direction. J = + 1.

  In step 211, the wafer lens 12 measured in the above step is attached to the wafer chuck 13 so that the front and back surfaces of the wafer lens 12 are reversed with the Y axis as the rotation axis and the B surface is on the measurement side by the probe 1.

  After that, in step 212, the wafer front and back surfaces of the wafer lens 11 installed in step 202 are reversed and installed with the Y axis as the rotation axis, so that the alignment mark formed on the back surface of the substrate of the transparent wafer lens. In the measurement of steps 202 to 210, the alignment mark 29 on the probe 1 side is moved to the position on the wafer chuck 13 side and the position on the wafer chuck side is measured due to the rotation of the front and back of the wafer. The focus height of the camera 8 is adjusted by the Z-axis stage 9 at the position of the alignment mark 29 to be performed.

  Thereafter, in step 213, the camera 8 is moved to the position of the alignment mark 10 for calibration while maintaining the focus height position of the camera 8 adjusted by the Z-axis stage 9 in step 212. Is inverted with the Y axis as the rotation axis, and the Z3 axis stage 11 is adjusted so that the focus of the camera 8 matches the calibration alignment mark 10 by the Z height corresponding to the changed wafer thickness. .

  In step 214, the center position of the camera 8 slightly shifted in the X and Y directions by the adjustment in the procedure of steps 211 to 213 is determined by using the calibration alignment mark 10 and the distance between the center of the probe 1 and the camera 8. Measure XoB.

  This calibration procedure is calibrated by a calibration procedure similar to the calibration procedure shown in step 205.

  In steps 215 to 218, processing similar to that in steps 206 to 209 is performed in the installation state.

  In step 219, the measurement data in the two directions X and Y measured in step 218, the positions XaB1 and XaB2 of the two alignment marks 29 obtained in step 215, the probe 1 and the center distance XoB of the camera 8 are obtained. The center of each lens on the wafer lens 12 is calculated based on the alignment mark 29.

  This calculation procedure is calculated by a calculation procedure similar to the calculation procedure shown in the description of step 210.

  In step 220, the center position of each lens on the A surface and the wafer front and back are reversed with the Y axis as the rotation axis based on the alignment mark 29 obtained in step 210, and the alignment mark 29 obtained in step 219 is used. The center deviation of the front and back lenses of the wafer lens 12 is calculated from the center position of each lens on the B surface by calculation processing.

  (Xf [i, j], Yf [i, j]) (i, j are lens gratings on the XY plane on the wafer). An integer value representing the position), and the lens center XY position data point sequence on the back side of the same wafer lens obtained by the procedure of step 219, measured by rotating 180 degrees on the Y axis, [I, j], Yb [i, j]) (where i and j are integer values representing the grating position of the lens on the XY plane on the wafer) The lens position can be specified by using the value -i in which the polarity is reversed by reversing the polarity on the back side, and the center position deviation between the front and back surfaces of the lens is determined by the reference of each lens on the back surface. Center position deviation is calculated by the following formula It is.

  By the above procedure, it is possible to obtain the lens center position shift between the front and back surfaces of all lenses at positions corresponding to the front and back surfaces of the wafer.

  The characteristics of the wafer lens 12 are evaluated based on the results of the lens center position data on the front surface in step 220, the lens center position data on the back surface in step 219, and the center misalignment between the front and back lenses in step 220.

  In step 221, the measurement ends.

  In actual wafer lens manufacturing, there are wafer lenses of various thicknesses, and the mark height on the wafer changes according to the product type. The heights of the Z2-axis stage 9 for calibration and the Z3-axis stage 11 for calibration alignment mark height adjustment are adjusted.

  Further, when the lens position of one surface (A surface or B surface) of the wafer lens 12 of the same type is repeatedly measured based on the alignment mark 29, the measurement can be performed according to the steps 201 to 210. In this case, when measuring the second and subsequent wafer lenses 12 by repeated measurement, the measurement can be shortened and measured by skipping steps 203 to 205.

  In this way, the calibration alignment mark is measured by the camera, the lens shape and the lens center position are measured by the probe, and further, the wafer height or the alignment mark on the wafer is displayed by using the calibrated offset value. Depending on the height of whether the camera is on the back or on the back, the Z height adjustment Z2 axis stage that is adjusted during the focus adjustment of the camera 8 is affected by the deviation in straightness in the X and Y directions, and the optical caused by the focus height deviation of the camera. High-precision measurement can be performed without being affected by measurement errors such as center position deviation.

  The present invention is not limited to the above embodiment, and various modifications can be made. For example, with respect to the center distance between the probe 1 and the camera 8 in the measurement of the A surface of the wafer lens 12 provided with the alignment mark 29, it is not necessary to use the calibration alignment mark as in steps 204 and 205 of FIG. The center distance between the probe 1 and the camera 8 may be calibrated by performing the same processing as steps 204 and 205 using the 12 alignment marks 29.

  In the three-dimensional measurement method of the present invention, a three-dimensional shape is measured with high accuracy using a probe, and the offset with respect to the probe position is measured with high accuracy by using a camera as a reference based on the alignment mark position. Therefore, the present invention can be applied to a three-dimensional measurement application in which a plurality of lenses formed on a single substrate, such as a wafer lens and a lens array, are measured at high speed and with high accuracy.

  In the above-described embodiment, the embodiment using the wafer lens as the measurement object has been described. However, the present invention can be applied to a mold for manufacturing a wafer lens instead of the wafer lens, or a lens array.

DESCRIPTION OF SYMBOLS 1 Probe 2 Z1 axis stage 3 XY stage 8 Camera 9 Z2 axis stage 10 Calibration alignment mark 11 Z3 axis stage 12 Wafer lens 29 Alignment mark

Claims (4)

In a three-dimensional measurement method for acquiring surface shape data of an object to be measured by two or more surface detection means,
The height in the Z direction of the calibration alignment mark whose position in the XY direction does not move with respect to the object to be measured provided separately from the object to be measured matches the surface height of the object to be measured.
The calibration alignment marks having the same height are measured by the two or more surface detection means,
Using the measurement results to calibrate the offsets in the XY direction of the two or more surface detection means,
A three-dimensional shape measuring method, wherein the surface shape of the object to be measured is obtained by using the surface shape data of the object to be measured measured by the two or more surface detecting means and the calibrated offset. The surface detecting means includes a stylus for measuring a surface shape provided on a first Z-direction moving means that moves in the Z direction on an XY stage that moves in the XY direction, and a first that moves in the Z direction on the XY stage. A camera for measuring an image in the XY plane provided in the Z-direction moving means of No. 2,
The calibration alignment mark that matches the surface height of the object to be measured is photographed by the camera and measured by the stylus, and the offset between the stylus and the center position of the camera is calibrated from the photographing and measurement results. And
Measure the surface shape of the object to be measured with the stylus,
The three-dimensional shape measurement method according to claim 1, wherein a surface shape of the object to be measured is obtained using a measurement result by the stylus and the calibrated offset. The measurement of the surface shape of the object to be measured by the stylus is performed in a lump in the X direction of the lens surface so as to pass near the apex position of each lens surface on the object to be measured, and further in the vicinity of the apex position of the lens surface Measure in a lump in the Y direction so that
From the measurement data, the measurement data is divided for each lens according to the preset pitches of the X direction and Y direction of the lens, and the shape, the XYZ position and the posture of the lens center are divided for the data divided for each lens. The three-dimensional shape measuring method according to claim 2, wherein evaluation is performed.   The shape measurement method according to claim 1, wherein the object to be measured is a wafer lens in which a plurality of lenses are formed on a thin plate.

Images