TWI824071B - Central detection method - Google Patents

Abstract

[Issue] Suppress the situation where the outer edge of the wafer is incorrectly recognized when detecting the center of the wafer. [Solution] In this processing method, the detection area extending in the X-axis direction is shifted by 1 pixel at a time in the X-axis direction, and each time the detection area is shifted, the pixels of each pixel in the detection area are obtained. Light and dark values, and then calculate the slope value relative to the X-axis in the distribution of light and dark values. Then, based on this slope value, the coordinates of the edge of the wafer, that is, the peripheral coordinates, are determined.

Description

Central detection method

Field of invention The present invention relates to a center detection method.

Background of the invention In edge trimming, the chamfered portion of the outer periphery of the wafer is removed along the circumferential direction. In this edge trimming process, it is required that the width (radial length) of the removed portion in the outer periphery of the wafer be equal. Therefore, the center of the wafer is identified and the deviation between the center of the wafer and the rotation axis of the holding table is corrected (see Patent Document 1).

In addition, in order to identify the center of the wafer, there is the following method (see Patent Documents 2 and 3). That is, at least three places on the outer periphery of the wafer are photographed, and the outer peripheral edge pixels representing the pixels on the outer peripheral edge of the wafer in each photograph are found. Then, the wafer center is calculated from the coordinate values of the three outer peripheral pixels. In this method, when the outer peripheral edge of the wafer is found, the photographed image is binarized, and the boundary between the black pixels and the white pixels is identified as the outer peripheral edge pixel. Prior technical literature patent documents

Patent Document 1: Japanese Patent Application Publication No. 2006-93333 Patent Document 2: Japanese Patent No. 5486405 Patent Document 3: Japanese Patent Application Publication No. 2015-102389

Summary of the invention The problem to be solved by the invention When the photographic image is binarized, the pixels are classified into black or white based on a preset threshold, so the outer periphery of the wafer may be mistaken. In addition, if light is reflected at a portion close to the outer periphery of the wafer, the portion may be captured in a white color. Also, if a shadow appears on the portion outside the outer periphery of the wafer, it may occur. There are cases where this part is ingested in a black color. These situations may cause incorrect identification of the outer periphery of the wafer.

The object of the present invention is to suppress the following situation: when detecting the center of the wafer, the outer peripheral edge of the wafer is erroneously identified. means to solve problems

The center detection method of the present invention (this center detection method) is a center detection method for detecting the center of a disc-shaped workpiece, and includes the following steps: The holding step is to hold the disc-shaped workpiece on a holding table, which has a surface with a position specified by the X coordinate on the X axis and the Y coordinate on the Y axis and has a rotation axis; The outer peripheral image acquisition step is to position the imaging unit on the outer periphery of the disc-shaped workpiece to perform imaging by the imaging unit to obtain the peripheral image. The imaging unit is equipped with an imaging element, and the imaging element has an object facing A plurality of pixels are arranged in a direction parallel to the axis and in a direction parallel to the Y-axis, and the peripheral image includes a plurality of pixels arranged in the X-axis direction and the Y-axis direction; The detection area setting step is to set a linear detection area in the outer peripheral image. The number of pixels in the linear detection area is smaller than the number of pixels arranged in the X-axis direction of the outer peripheral image. 1/2; The threshold setting step is to obtain the light and dark values of each pixel in the detection area in a state where the pixels corresponding to the outer circumference of the disc-shaped workpiece in the outer peripheral image pass through the center of the detection area, and use the final The Xiaoping method calculates the slope value relative to the X-axis in the distribution of light and dark values, which is the detection area, as the reference slope value, and sets the slope value smaller than the reference slope value as the threshold value. The light and dark value of each pixel is represented by the position of each pixel in the X-axis direction; The slope value calculation step is to obtain the light and dark values of each pixel in the detection area while shifting the detection area in the outer peripheral image toward the disc-shaped workpiece and in the X-axis direction by 1 pixel at a time. The least squares method calculates the slope value relative to the X-axis in the distribution of light and dark values; In the outer peripheral coordinate determination step, when the slope value calculated in the slope value calculation step is equal to or greater than the threshold, the average value of the minimum value and the maximum value of the light and dark values of the pixels in the detection area is calculated, and the closest value is obtained. The coordinates of the pixels in the detection area based on the average light and dark value are used as the peripheral coordinates of the plate-shaped workpiece; and The center coordinate calculation step calculates the center of the disc-shaped workpiece based on three or more outer peripheral coordinates obtained by performing the slope value calculation step and the outer peripheral coordinate determination step three or more times at different positions of the disc-shaped workpiece. coordinates.

In this center detection method, in the slope value calculation step, the average value of the light and dark values of a plurality of pixels surrounding each pixel constituting the detection region may be used as the average value of each pixel constituting the detection region. Use light and dark values.

In this center detection method, in the detection area setting step, a plurality of detection areas arranged in the Y-axis direction may be set in the peripheral image, In this slope value calculation step, slope values are calculated for a plurality of detection areas, In the outer peripheral coordinate determination step, the outer peripheral coordinates of the disc-shaped workpiece are obtained for a plurality of detection areas, and the distance from each outer peripheral coordinate to the coordinate of the rotation axis center of the holding table is calculated to create the outer peripheral coordinates of the disk-shaped workpiece. The distance approximates the group of outer peripheral coordinates, and any outer peripheral coordinate of the group to which the largest number of outer peripheral coordinates belongs is determined as the outer peripheral coordinate of the disc-shaped workpiece. Invention effect

In this center detection method, the detection area extending in the X-axis direction is shifted by 1 pixel each time in the X-axis direction, and each time the detection area is shifted, the light and dark values of each pixel in the detection area are obtained. Furthermore, the outer peripheral coordinates of the disc-shaped workpiece are determined based on the slope value relative to the X-axis in the distribution of light and dark values. Therefore, in this center detection method, the outer peripheral coordinates of the disc-shaped workpiece can be determined without binarizing the captured image. Therefore, it is possible to avoid erroneous identification of the outer circumference of the disk-shaped workpiece due to errors in binarization processing.

Furthermore, when the average value of the brightness values of a plurality of pixels surrounding the pixel is used as the brightness value of a pixel in the detection area, even if the brightness value of the pixel in the detection area is affected by unnecessary light, etc. Deviating from the original value can still suppress the effect.

In addition, when a plurality of detection areas are used and the outer peripheral coordinates are set by majority voting, even if the light and dark value of one detection area is erroneously detected due to the influence of unnecessary light, etc., and as a result Obtaining wrong peripheral coordinates can still reduce the impact.

Form used to implement the invention The processing method of this embodiment (this processing method) is a method of processing a wafer, which is an example of a disk-shaped workpiece, and includes a center detection method for detecting the center of the wafer.

As shown in FIG. 1 , a device D is formed on the front surface Wa of a disk-shaped wafer W. The back surface Wb of the wafer W does not have the device D and is a surface to be ground by a grinding stone or the like.

As shown in FIG. 2 , on the edge WE of the wafer W, a chamfer portion is formed such that the cross-sectional shape from the front surface Wa to the back surface Wb becomes an arc shape. In addition, in FIG. 2 , the device D provided on the front surface Wa of the wafer W is omitted. In addition, in this embodiment, the color of the wafer W is black.

In this processing method, the front surface Wa side of the edge WE of the wafer W is removed (edge trimming process). Therefore, in this processing method, the cutting device 1 shown in Figure 3 is used. As shown in FIG. 3 , the dicing device 1 is a device that performs edge trimming processing on the wafer W held on the holding surface 302 of the holding table 30 by rotating and cutting the dicing blade 63 provided in the dicing unit 6 . The cutting device 1 includes a base 10 , a portal support 14 erected on the base 10 , and a control unit 7 that controls each component of the cutting device 1 .

An X-axis direction feed unit 11 is provided on the base 10 . The X-axis direction feed unit 11 moves the holding table 30 along the cutting feed direction (X-axis direction). The X-axis direction feed assembly 11 includes a pair of guide rails 111 extending in the X-axis direction, an X-axis table 113 mounted on the guide rails 111, a ball screw 110 extending parallel to the guide rails 111, and the ball screw 110 is rotated. of motor 112.

The pair of guide rails 111 are arranged on the upper surface of the base 10 parallel to the X-axis direction. The X-axis table 113 is slidably provided on a pair of guide rails 111 along these guide rails 111 . The holding part 3 is placed on the X-axis table 113 .

The ball screw 110 is screwed into a nut portion (not shown) provided on the lower surface side of the X-axis table 113 . The motor 112 is connected to one end of the ball screw 110 and drives the ball screw 110 to rotate. By rotationally driving the ball screw 110, the X-axis table 113 and the holding part 3 move along the guide rail 111 and along the cutting feed direction, that is, the X-axis direction.

The holding unit 3 has a holding table 30 for holding the wafer W. The holding table 30 has a θ table 31 which is a rotation axis that supports and rotates the holding table 30 .

The holding table 30 is a member for adsorbing and holding the wafer W shown in FIG. 1 , and is formed in a disc shape. The holding table 30 includes an adsorption part 300 made of a porous material, and a white frame 301 that supports the adsorption part 300.

The suction part 300 is connected to a suction source (not shown) and has a holding surface 302 that is an exposed surface. The holding surface 302 is circular slightly smaller than the wafer W, and is flush with the upper surface of the frame 301 . The suction part 300 attracts and holds the wafer W through the holding surface 302 . In this embodiment, the positions on the upper surface of the frame 301 and the holding surface 302 that form the surface of the holding table 30 are defined by the X coordinate on the X axis and the Y coordinate on the Y axis.

The holding table 30 is supported by a θ table 31 arranged on the bottom side of the holding table 30 . The θ table 31 is provided on the upper surface of the X-axis table 113 so as to be rotatable in the XY plane. Thereby, the θ table 31 can support the holding table 30 and rotationally drive the holding table 30 in the XY plane.

On the rear side (-X direction side) of the base 10, the gate-shaped support 14 is erected and installed across the X-axis direction feed assembly 11. A cutting part moving mechanism 13 for moving the cutting part 6 is provided on the front surface (the surface on the +X-axis direction side) of the portal support 14 . The cutting part moving mechanism 13 allows the cutting part 6 to index and feed in the Y-axis direction and to cut and feed in the Z-axis direction. The cutting unit moving mechanism 13 includes a Y-axis direction moving unit 12 that moves the cutting unit 6 in the indexing feed direction (Y-axis direction), and a Z unit that moves the cutting unit 6 in the cutting feed direction (Z-axis direction). Axis direction moving component 16.

The Y-axis direction moving component 12 is arranged on the front surface of the portal support 14 . The Y-axis direction moving unit 12 moves the Z-axis direction moving unit 16 and the cutting part 6 back and forth in the Y-axis direction. The Y-axis direction is parallel to the holding surface direction (horizontal direction) and orthogonal to the X-axis direction.

The Y-axis direction moving assembly 12 includes a pair of guide rails 121 extending in the Y-axis direction, a Y-axis table 123 mounted on the guide rails 121, a ball screw 120 extending parallel to the guide rails 121, and a device for rotating the ball screw 120. Motor 122.

The pair of guide rails 121 are arranged on the front surface of the portal support 14 parallel to the Y-axis direction. The Y-axis table 123 is slidably provided on a pair of guide rails 121 along these guide rails 121 . The Z-axis direction moving unit 16 and the cutting unit 6 are mounted on the Y-axis table 123 .

The ball screw 120 is screwed to a nut portion (not shown) provided on the back side of the Y-axis table 123 . The motor 122 is connected to one end of the ball screw 120 and drives the ball screw 120 to rotate. By rotationally driving the ball screw 120, the Y-axis table 123, the Z-axis direction moving assembly 16, and the cutting part 6 move along the guide rail 121 in the indexing feed direction, that is, in the Y-axis direction.

The Z-axis direction moving unit 16 reciprocates the cutting part 6 in the Z-axis direction (vertical direction). The Z-axis direction is orthogonal to the X-axis direction and the Y-axis direction. That is, the Z-axis direction is a direction orthogonal to the holding surface 302 of the holding table 30 .

The Z-axis direction moving assembly 16 includes a pair of guide rails 161 extending in the Z-axis direction, a support member 163 placed on the guide rails 161, a ball screw 160 extending parallel to the guide rails 161, and a motor 162 that rotates the ball screw 160. .

The pair of guide rails 161 are arranged on the Y-axis table 123 parallel to the Z-axis direction. The support member 163 is provided on a pair of guide rails 161 so as to be slidable along these guide rails 161 . The cutting part 6 is attached to the lower end of the support member 163 .

The ball screw 160 is screwed into a nut portion (not shown) provided on the back side of the support member 163 . The motor 162 is connected to one end of the ball screw 160 and drives the ball screw 160 to rotate. By rotationally driving the ball screw 160, the support member 163 and the cutting part 6 move along the guide rail 161 in the cutting feed direction, that is, the Z-axis direction.

As shown in an enlarged version of FIG. 3 , the cutting unit 6 includes a housing 61 provided at the lower end of the support member 163 , a rotation shaft 60 extending in the Y-axis direction, a cutting blade 63 installed on the rotation shaft 60 , and A motor (not shown) that drives the rotating shaft 60 . The rotating shaft 60 is rotatably supported by the housing 61 . As the motor rotates and drives the rotating shaft 60, the cutting blade 63 also rotates at high speed.

In addition, the cutting unit 6 is provided with a photographing unit 65 on the front surface of the housing 61 . The photographing assembly 65 is installed at the front of the housing 61 . The imaging unit 65 includes an imaging element having a plurality of pixel portions (light receiving portions) arranged in a direction parallel to the X-axis and a direction parallel to the Y-axis. That is, the plurality of pixel portions (light-receiving portions) of the imaging element are arranged two-dimensionally along the X-axis direction and the Y-axis direction. The imaging unit 65 is configured to set the lower part of the imaging unit 65 along the −Z-axis direction as an imaging area and photograph this area. The imaging area of the imaging unit 65 can be set by the X-axis direction feed unit 11, the cutting unit moving mechanism 13, and the θ stage 31. The imaging component 65 can photograph, for example, the wafer W placed on the holding surface 302 and its vicinity. In the present embodiment, the image captured by the capturing unit 65 is, for example, a 256-gradation gray scale image.

The control unit 7 is provided with a memory 71 that stores various data and programs. The control unit 7 executes various processes and integrally controls each component of the cutting device 1 . For example, detection results from various detectors (not shown) can be input to the control unit 7 . In addition, the control part 7 controls the X-axis direction feed assembly 11 (motor 112), the cutting part moving mechanism 13 (motor 122 and motor 162) and the θ table 31 to determine the cutting blade 63 of the cutting part 6. The position of the wafer W and the imaging area of the imaging component 65. In addition, the control unit 7 controls the motor of the cutting unit 6 to perform cutting processing on the wafer W, and controls the imaging unit 65 to perform imaging of the imaging area.

Hereinafter, the present processing method using the cutting device 1 shown in FIG. 3 will be described. (1) Maintain steps In this step, the user places the wafer W on the holding surface 302 of the holding table 30 . Furthermore, the control unit 7 controls a suction source (not shown) to transmit the suction force to the holding surface 302 so that the holding surface 302 suctions and holds the wafer W. Furthermore, as mentioned above, the positions on the upper surface of the frame 301 and the holding surface 302 that constitute the surface of the holding table 30 are defined by the X coordinate and the Y coordinate.

(2) Peripheral image acquisition steps In this step, the control unit 7 first controls the X-axis direction feeding assembly 11, the cutting unit moving mechanism 13 and the θ table 31, and as shown in Figure 4, sets the imaging area of the imaging assembly 65 to be in this area. The frame 301 includes the edge WE of the wafer W and the holding table 30 . After that, the control part 7 controls the imaging component 65 to photograph the imaging area. Thereby, an initial image including a plurality of pixels arranged in the X-axis direction and the Y-axis direction as shown in FIG. 5 can be formed.

In this initial image, the pixel FP corresponding to the upper left side of the frame 301 which is a background member is represented by white corresponding to the color of the frame 301 . On the other hand, the pixels WP from the upper right to the lower left corresponding to the wafer W are represented by black corresponding to the color of the wafer W. In addition, the gray pixel SP formed at the boundary between the black pixel WP and the white pixel FP is a shadow formed by the wafer W on the frame 301 . In addition, the white line formed in the black pixel WP is the boundary line B set in the imaging unit 65 .

Next, the control unit 7 controls the X-axis direction feed unit 11, the cutting unit moving mechanism 13, and the θ stage 31, sets the imaging area of the imaging unit 65 so that the boundary line B overlaps the gray pixel SP, and controls the imaging unit 65 to capture this shooting area. Thereby, an image including a plurality of pixels arranged in the X-axis direction and the Y-axis direction and capturing the edge WE of the wafer W and the outer peripheral image of the frame 301 of the holding table 30 is formed as shown in FIG. 6 .

(3) Detection area setting procedure In this step, as shown in FIG. 6 , the control unit 7 sets a linear detection area (linear sensing area) LS on the peripheral image. This detection area LS is composed of a plurality of pixels arranged in one column along the X-axis direction, for example. The number of pixels forming the detection area LS is less than 1/2 of the number of pixels arranged in the X-axis direction in the peripheral image (that is, smaller than the X-axis in the imaging element (imaging area) of the imaging unit 65 1/2 of the number of pixel portions in the direction), and is, for example, 20 pixels.

(4) Threshold setting steps In this step, the control unit 7 sets the position of the detection area LS on the outer peripheral image to a position where the pixels corresponding to the outer periphery of the wafer W pass through the center of the detection area LS (that is, as if The position where the detection area LS2 is equally divided) (reference position). In this state, the control unit 7 obtains the brightness value of each pixel in the detection area LS. Furthermore, each time the detection area LS is shifted by 1 pixel in the X-axis direction, the slope value with respect to the X-axis in the distribution of light and dark values can be calculated, and the slope value of the detection area LS can be maximized. The position is set as the reference position, and the distribution of the light and dark values is represented by the light and dark value of each pixel in the detection area LS as the position of each pixel in the X-axis direction.

When the detection area LS is located at the reference position, the left side of the detection area LS is the pixel FP corresponding to the frame 301, the right side is the pixel WP corresponding to the wafer W, and the center is the gray pixel SP. Therefore, the relationship between the position of each pixel in the detection area LS and the light and dark values output from each pixel (distribution of light and dark values) is formed as shown in FIG. 7 . Then, the control unit 7 creates the regression line RL by the least squares method, calculates the slope value relative to the X-axis in the distribution of light and dark values in the detection area LS located at the reference position, and sets this slope value as the reference Slope value θ0.

In addition, the control unit 7 sets a slope value smaller than the reference slope value θ0 and close to the reference slope value θ0 as the threshold value. The threshold value is a value obtained by multiplying the reference slope value θ0 by 0.8, for example. Furthermore, the threshold value can be obtained for one detection area on the peripheral image. Furthermore, one threshold value may be used for at least three peripheral images obtained by photographing at least three locations described below, or a different threshold value may be used for each peripheral image.

(5) Steps to calculate the slope value and determine the peripheral coordinates In the slope value calculation step, the control unit 7 first sets the position of the detection area LS on the outer peripheral image as shown in FIG. 8 so that all the pixels in the detection area LS correspond to the pixels FP of the frame 301 Location. Thereafter, as shown by arrow A in FIG. 8 , the control unit 7 moves (shifts) the detection area LS toward the pixel WP corresponding to the wafer W and in the X-axis direction by 1 at a time on the peripheral image. Amount of pixels.

Then, as shown in FIG. 9 , the control unit 7 acquires the light and dark values of each pixel in the detection area LS each time the linear sensing area LS is moved (shifted) ( S1 ). In addition, the control unit 7 calculates the slope value relative to the X-axis in the distribution of light and dark values by the least squares method (S2).

For example, when the detection area LS is set so that all pixels are positioned corresponding to the pixel FP of the frame 301, the distribution of light and dark values in the detection area LS and the regression line RL become as shown in FIG. 10 A straight line roughly parallel to the X-axis. On the other hand, when the detection area LS is set to the right direction beyond the reference position, that is, when the detection area LS is set at a position where a large number of pixels become the pixels WP corresponding to the wafer W. , the distribution of light and dark values in the detection area LS and the regression line RL are formed as shown in Figure 11.

Furthermore, every time the slope value is calculated, the control unit 7 determines whether the calculated slope value is equal to or greater than the threshold value set in the threshold value setting step (S3), as shown in FIG. 9 . When the control unit 7 determines that the calculated slope value is less than the threshold value (S3; No), the control unit 7 returns to S1 and continues the movement of the detection area LS and the calculation of the slope value.

On the other hand, when the control unit 7 determines that the calculated slope value is equal to or greater than the threshold value (S3; Yes), the control unit 7 calculates the minimum value and the maximum value of the light and dark values output from each pixel of the detection area LS. Average (S4). Then, the control unit 7 identifies a pixel having a brightness value closest to the calculated average value among the pixels in the detection area LS, and uses the coordinates of this pixel as the outer peripheral coordinates of the wafer W (the coordinates of the wafer W). The coordinates of edge WE (refer to Figure 1) are obtained (S5).

For example, when it is determined that the slope value is equal to or greater than the threshold value, a distribution of light and dark values as shown in FIG. 12 can be obtained. The control unit 7 calculates the average value of the light and dark values by dividing the sum of the maximum value Bmax and the minimum value Bmin of the light and dark values by 2. Then, the control unit 7 identifies the pixel (pixel Xe in FIG. 12 ) of the detection area LS that has a light and dark value closest to this average value ((maximum value Bmax + minimum value Bmin)/2). Then, the control unit 7 specifies the X coordinate and Y coordinate of the specified pixel based on the position of the detection area LS in the imaging area, and obtains them as the outer peripheral coordinates of the wafer W.

(6) Center coordinate calculation steps Next, the control unit 7 controls the θ stage 31 to rotate the wafer W by a predetermined angle together with the holding stage 30 holding the wafer W. Then, the control unit 7 executes the above-mentioned outer peripheral image acquisition step, detection region setting step, threshold setting step, slope value calculation step, and outer peripheral coordinate determination step. In this manner, the control unit 7 rotates the wafer W by a predetermined angle and performs the following steps at least three times (that is, more than three times): the peripheral image acquisition step, the detection area setting step, and the threshold setting step. , the step of calculating the slope value and the step of determining the peripheral coordinates. That is, the control unit 7 acquires at least three peripheral images (that is, three or more peripheral images) and sets a threshold corresponding to each peripheral image. Then, after executing the slope value calculation step, the control unit 7 executes the outer peripheral coordinate determining step using the set threshold value. Thereby, the control unit 7 obtains the outer peripheral coordinates of at least three locations (that is, three or more locations). Then, the center coordinates of the wafer W are calculated based on these three or more peripheral coordinates.

As a method of calculating the center coordinates, any known method may be used. For example, the control unit 7 obtains the perpendicular bisectors of two straight lines connecting two adjacent points among the selected three points. Then, the control unit 7 can calculate the intersection point of the two perpendicular bisectors as the center of the wafer W. Furthermore, the control unit 7 may perform the peripheral image acquisition step, the detection region setting step, the slope value calculation step, and the like without performing the threshold setting step by rotating the wafer W by a predetermined angle. The peripheral coordinate determination step is to obtain the peripheral coordinates of at least three locations. In this case, the control unit 7 may use the threshold set in the first threshold setting step in the second and subsequent outer peripheral coordinate determining steps.

(7) Edge removal step In this step, the control unit 7 determines the offset amount between the center of the wafer W calculated in the center coordinate calculation step and the center of the holding table 30 , that is, the rotation axis center (hereinafter, the center offset amount). Furthermore, the control unit 7 causes the rotating dicing blade 63 to cut into the edge WE in the front surface Wa of the wafer W while rotating the holding table 30 via the θ table 31 shown in FIG. 3 .

At this time, the control unit 7 moves the dicing blade 63 in the Y-axis direction based on the center deviation amount and the outer peripheral coordinates of the wafer W. Thereby, the control unit 7 can cut the edge WE of the wafer W in the circumferential direction at substantially the same width as the center position of the wafer W. In addition, regarding the position control of the cutting blade 63 in response to the center deviation amount, a known method such as that described in Patent Document 1 can be used. In this way, the control unit 7 removes the front surface Wa side of the edge WE of the wafer W (edge trimming). The cutting blade 63 used has, for example, a flat tip.

As mentioned above, in this processing method, the detection area LS extending in the X-axis direction is moved (shifted) by 1 pixel each time in the X-axis direction, and each time the detection area LS is moved (shifted), the detection is obtained Using the light and dark values of each pixel in the area LS, the slope value relative to the X-axis in the light and dark value distribution is calculated. Then, based on this slope value, the coordinates of the edge WE of the wafer W, that is, the outer peripheral coordinates are determined. Therefore, in this processing method, the outer peripheral coordinates can be determined without binarizing the captured image. Therefore, misidentification of edge WE based on errors in binarization processing can be avoided.

Furthermore, in this embodiment, in the slope value calculation step, the light and dark values of each pixel constituting the detection area LS are obtained. At this time, as shown in FIG. 13 , the control unit 7 may set a pixel group PG for each pixel constituting the detection area LS. The pixel group PG is composed of a plurality of pixels (for example, 20 pixels in the X-axis direction) surrounding each pixel. , 20 pixels along the Y-axis direction). Furthermore, the control unit 7 may use an average value of the brightness values of a plurality of pixels included in the pixel group PG as the brightness value output from each pixel.

In this regard, the brightness value of the pixel in the detection area LS may deviate from the original value due to the influence of unnecessary light (unnecessary light) irradiated to the frame 301 and the wafer W from the outside. In the above-described configuration, since the average value of the light and dark values of the pixel group surrounding the pixel is used, the above-mentioned influence of unnecessary light, etc. can be suppressed.

In addition, in this embodiment, the control unit 7 calculates the slope value using one detection region LS in the slope value calculation step. Instead of this, a plurality (n) of detection areas LS (LS1 to LSn) arranged offset in the Y-axis direction as shown in FIG. 14 may be used.

In this case, in the slope value calculation step, the control unit 7 calculates slope values for each of the plurality of detection areas LS. Furthermore, in the outer peripheral coordinate determination step, the control unit 7 acquires the outer peripheral coordinates of the wafer W for each of the plurality of detection areas LS.

Furthermore, the control unit 7 calculates the distance (first distance) from the outer peripheral coordinates to the coordinates of the rotation axis center holding the center of the table 30 based on the outer peripheral coordinates. Thereby, the distance to the coordinate of the rotation axis center corresponding to each outer peripheral coordinate is calculated. Then, the control unit 7 creates a group of peripheral coordinates in the form of the calculated distance approximation. That is, the control unit 7 divides the outer peripheral coordinates into groups according to the calculated distance. Then, the control unit 7 determines any one of the outer peripheral coordinates in the group to which the largest number of outer peripheral coordinates belongs as the outer peripheral coordinates of the wafer W.

For example, in the example shown in FIG. 15 , the outer peripheral coordinates are divided into a group G1 corresponding to a certain distance D1 and a group G2 corresponding to a distance D2 shorter than the distance D1 . In this example, the control unit 7 determines any one (it may be one or a plurality) of the outer peripheral coordinates belonging to the group G1 as the outer peripheral coordinate of the wafer W. Furthermore, in FIG. 15 , the distance between each outer peripheral coordinate is represented by the position in the Y-axis direction of the detection area LS (LS1 to LSn) corresponding to each outer peripheral coordinate.

In this regard, as mentioned above, due to the influence of unnecessary light irradiated on the frame 301 and the wafer W, there is a possibility that the light and dark values of the pixels in the detection area LS are erroneously detected, and an incorrect outer periphery is determined. coordinates as the result. In the method shown in FIG. 14 , a plurality of detection areas LS are used to determine the outer peripheral coordinates by majority voting. Therefore, the influence of erroneous detection of light and dark values that occurs in one detection area LS can be suppressed. becomes smaller, it is possible to suppress the influence of unnecessary light as mentioned above.

In addition, in this embodiment, the wafer W shown in FIGS. 1 and 2 is used as the disk-shaped workpiece. Instead of this, a bonded substrate may be used as a disk-shaped workpiece. The bonded substrate includes a disc-shaped support substrate and a wafer bonded to the support substrate with a smaller diameter than the support substrate. At this time, in this processing method, the outer peripheral coordinates of the wafer to which the substrate is bonded can be obtained and the center coordinates thereof can be calculated.

At this time, in this processing method, since the slope value in the distribution of light and dark values in the detection area LS is used, the outer peripheral edge of the support substrate, which is a different color from the wafer, can be distinguished from the outer peripheral edge of the wafer. identify. Therefore, like the example shown in this embodiment, the outer peripheral edge of the wafer in the bonded substrate can be cut in the circumferential direction at a substantially same width from the center position of the wafer.

In addition, in this embodiment, the control unit 7 controls the X-axis direction feeding unit 11, the cutting unit moving mechanism 13 and the θ stage 31 from the initial image shown in FIG. 5 to change the imaging area of the imaging unit 65. It is set so that the boundary line B overlaps the gray pixel SP. At this time, for example, the control unit 7 can also cause the control unit 7 to overlap the boundary line B with the gray pixel SP according to the user's instruction of visually observing the position of the boundary line B and the pixel SP. Alternatively, the user can also directly control the X-axis direction feeding component 11, the cutting part moving mechanism 13, and the driving device of the θ table 31 to overlap the boundary line B with the gray pixel SP.

Alternatively, the light amount adjustment step of adjusting the light amount of the imaging unit 65 may be performed after the boundary line B is superimposed on the gray pixel SP. This step can also be implemented based on the visual results of the user, or can be implemented by the control unit 7 .

For example, the control unit 7 stores a plurality of types (for example, 100 types) of light intensity patterns in the memory 71 in advance. The light intensity patterns are distinguished by, for example, the ratio of epi-illumination and oblique illumination, and their respective illumination intensities.

Furthermore, as shown in FIG. 16 , the control unit 7 sets the first light amount inspection area RA in the black pixel WP corresponding to the wafer W, and sets the second light amount inspection area RA in the white pixel FP corresponding to the frame 301 Area RB. Furthermore, the control unit 7 sets the first inspection region RC1 which is adjacent to the boundary line B and includes the first light amount inspection region RA. Similarly, the control unit 7 sets the second inspection area RC2 which is adjacent to the boundary line B and includes the second light amount inspection area RB. The size of the first inspection area RC1 and the second inspection area RC2 is, for example, 10×10 pixels.

The control unit 7 changes the first inspection area RC1 and the second inspection area RC2 along the boundary line B with respect to the light amount pattern of one imaging unit 65, and acquires the plurality of first inspection areas RC1 and the second inspection area RC2. Light and dark values. The control unit 7 performs such acquisition of light and dark values for a plurality of light intensity patterns. Then, the control unit 7 specifies a light amount pattern in a form in which the difference between the light and dark values in the first inspection area RC1 and the second inspection area RC2 becomes the largest, and uses this light amount pattern to perform the operation performed by the imaging unit 65 Photograph and implement each step of this processing method.

This makes it possible to increase the difference in light and dark values on both sides of the boundary line B. Therefore, in the threshold value setting step, the slope value calculation step, and the peripheral coordinate determination step, the light and dark values in each pixel of the detection area LS can be made smaller. The difference becomes larger. As a result, the change in the slope value (the angle between the regression line RL and the X-axis) according to the position of the detection area LS in the distribution of light and dark values as shown in FIG. 7 and others can be increased. Thereby, the accuracy of the comparison between the slope value and the threshold value can be improved.

1: Cutting device 3:Maintenance Department 6: Cutting Department 7:Control Department 71:Memory 10:Abutment 11: X-axis direction feed component 110, 120, 160: Ball screw 111, 121, 161: Guide rail 112, 122, 162: Motor 113:X-axis workbench 12: Y-axis direction moving component 123:Y-axis workbench 13: Cutting part moving mechanism 14: Door type pillar 16:Z-axis direction moving component 163:Supporting member 30: Keep the workbench 31:θ workbench 300: Adsorption part 301:frame 302:Keep the surface 60:Rotation axis 61: Shell 63:Cutting blade 65: Shooting components A:arrow B:junction line D: device D1, D2: distance G1, G2: Group FP: white pixels (pixels) LS, LS1~LSn: detection area (linear sensing area) WP: black pixels (pixels) RA: 1st light intensity inspection area RB: 2nd light intensity inspection area RC1: 1st inspection area RC2: 2nd inspection area RL: regression line S1~S5: steps SP: gray pixels (pixels) PG: pixel group W:wafer Wa:front Wb: back WE: edge X, Y, ±X, ±Y, ±Z: direction

FIG. 1 is a perspective view showing a wafer according to this embodiment. FIG. 2 is a cross-sectional view of the wafer shown in FIG. 1 . FIG. 3 is a perspective view showing a cutting device for processing wafers. FIG. 4 is an explanatory diagram showing an implementation aspect of the positioning step. FIG. 5 is an explanatory diagram showing an example of an initial image obtained in the positioning step. FIG. 6 is an explanatory diagram showing an example of a peripheral image obtained in the positioning step. FIG. 7 is a graph showing an example of the relationship between the position of each pixel in the detection area (linear detection area) and the brightness value output from each pixel. FIG. 8 is an explanatory diagram showing an example of movement of the detection area in the slope value calculation step. FIG. 9 is a flowchart showing the operations of the slope value calculation step and the outer peripheral coordinate determination step. FIG. 10 is a graph showing an example of the distribution of light and dark values and the regression line of the detection area when the detection area is set at a position such that all pixels in the detection area correspond to the pixels of the frame. 11 is a graph showing an example of the distribution of light and dark values and the regression line of the detection area when the detection area is set at a position such that a large number of pixels in the detection area correspond to pixels on the wafer. FIG. 12 is a graph showing an example of the distribution of light and dark values when it is determined that the slope value is equal to or greater than the threshold value. FIG. 13 is an explanatory diagram showing a modified example of calculation of the brightness value of the detection area. FIG. 14 is an explanatory diagram showing a modification using a plurality of detection areas. FIG. 15 is a graph showing the distribution of a plurality of peripheral coordinates obtained in the modification of FIG. 14 . FIG. 16 is an explanatory diagram showing the light amount adjustment procedure.

A:arrow

B:junction line

FP: white pixels

LS: Detection area (linear sensing area)

WP: black pixels

SP: gray pixels

X, Y: direction

Claims (3)

A center detection method is to detect the center of a disc-shaped workpiece. The center detection method includes the following steps: The holding step is to hold the disc-shaped workpiece on a holding table, which has a surface with a position specified by the X coordinate on the X axis and the Y coordinate on the Y axis and has a rotation axis; The outer peripheral image acquisition step is to position the imaging unit on the outer periphery of the disc-shaped workpiece to perform imaging by the imaging unit to obtain the peripheral image. The imaging unit is equipped with an imaging element, and the imaging element has an object facing A plurality of pixel portions are arranged in a direction parallel to the axis and in a direction parallel to the Y-axis, and the peripheral image includes a plurality of pixels arranged in the X-axis direction and the Y-axis direction; The detection area setting step is to set a linear detection area in the outer peripheral image. The number of pixels in the linear detection area is smaller than the number of pixels arranged in the X-axis direction of the outer peripheral image. 1/2; The threshold setting step is to obtain the light and dark values of each pixel in the detection area in a state where the pixels corresponding to the outer circumference of the disc-shaped workpiece in the outer peripheral image pass through the center of the detection area, and use the final The Xiaoping method calculates the slope value relative to the X-axis in the distribution of light and dark values, which is the detection area, as the reference slope value, and sets the slope value smaller than the reference slope value as the threshold value. The light and dark value of each pixel is represented by the position of each pixel in the X-axis direction; The slope value calculation step is to obtain the light and dark values of each pixel in the detection area while shifting the detection area in the outer peripheral image toward the disc-shaped workpiece and in the X-axis direction by 1 pixel at a time. The least squares method calculates the slope value relative to the X-axis in the distribution of light and dark values; In the outer peripheral coordinate determination step, when the slope value calculated in the slope value calculation step is equal to or greater than the threshold, the average value of the minimum value and the maximum value of the light and dark values of the pixels in the detection area is calculated, and the closest value is obtained. The coordinates of the pixels in the detection area based on the average light and dark values are used as the peripheral coordinates of the disc-shaped workpiece; and The center coordinate calculation step calculates the circumferential coordinates of the disc-shaped workpiece based on three or more outer peripheral coordinates obtained by performing the slope value calculation step and the outer peripheral coordinate determining step three or more times at different positions of the disc-shaped workpiece. Center coordinates. The center detection method of Claim 1, wherein in the step of calculating the slope value, an average value of the brightness and darkness values of a plurality of pixels surrounding each pixel constituting the detection region is used as the brightness and darkness of each pixel constituting the detection region. value to use. The center detection method of claim 1 or 2, wherein in the detection area setting step, a plurality of the detection areas arranged in the Y-axis direction are set in the peripheral image, In this slope value calculation step, slope values are calculated for a plurality of detection areas, In the outer peripheral coordinate determination step, the outer peripheral coordinates of the disk-shaped workpiece are obtained for a plurality of detection areas, and the distance from each outer peripheral coordinate to the coordinate of the rotation axis center of the holding table is calculated. This distance approximates the group of outer peripheral coordinates, and any outer peripheral coordinate of the group to which the largest number of outer peripheral coordinates belongs is determined as the outer peripheral coordinate of the disc-shaped workpiece.