Intersection detection method

Abstract

To provide an intersection region detection method which can automatically detect a region where a division scheduled line intersects and display the region on display means.SOLUTION: An intersection region detection method comprises the steps of: holding a wafer 10 on a chuck table; positioning the wafer immediately below imaging means 11; imaging the wafer with the imaging means and detecting a plurality of straight line regions with an image processing algorithm provided in control means 100 and executing Hough transformation; detecting two edges close to each other in a prescribed range in a straight line region having the largest number of straight line components from the plurality of detected straight line regions as a pair of first edges defining the division scheduled line 14 in a prescribed direction; and detecting the straight line region intersecting the pair of first edges and detecting the two edges close to each other in a prescribed range in the straight line region having the largest number of straight line components from the plurality of detected straight line regions as a pair of second edges defining the division scheduled line intersecting the pair of first edges.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an intersection area detection method for detecting an area where a plurality of devices are formed on a wafer whose surface is partitioned by dividing lines defined by a pair of edges, by imaging means. [Background technology]

[0002] Wafers with IC, LSI, and other devices formed on their surfaces, partitioned by planned division lines, are then separated into individual device chips using dicing equipment and laser processing equipment, and are used in electrical devices such as mobile phones and personal computers.

[0003] The dicing device includes at least a chuck table for holding a wafer, cutting means for cutting the wafer held on the chuck table, X-axis movement means for moving the chuck table in the X-axis direction, Y-axis movement means for moving the chuck table in the Y-axis direction perpendicular to the X-axis direction, rotation means for rotating the chuck table in a plane defined by the X-axis and Y-axis, imaging means facing the chuck table for imaging the wafer held on the chuck table and detecting the area to be cut, display means for displaying the image captured by the imaging means, and control means, and can divide the wafer into individual device chips with high precision (see, for example, Patent Document 1).

[0004] Furthermore, similar to the above-mentioned dicing device, a laser processing device can also use an imaging means to detect the intended dividing lines to be laser processed, position a concentrator that focuses the laser beam, and divide the wafer into individual device chips with high precision (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 05-315444 [Patent Document 2] Japanese Patent Application Laid-Open No. 2004-322168 Summary of the Invention [Problem to be solved by the invention]

[0006] The control means in the above-mentioned dicing device and laser processing device pre-records the characteristic pattern of the device near the planned dividing line as a target pattern, and performs pattern matching between the image of the device captured by the imaging means and the target pattern to detect the position information of the planned dividing line to be processed.

[0007] In addition, the present applicant has proposed a technology for automatically setting a target pattern or a technology for automatically creating a map based on the area to be processed (Patent Application No. 2021-74026, Patent Application No. 2021-98677).

[0008] However, in any of the above techniques, it is necessary to identify the area where the dividing lines formed on the surface of the wafer intersect before detecting the dividing lines to be processed by performing pattern matching between the image of the device captured by the imaging means and the target pattern.In contrast, in the conventional technique, an operator is required to display the image captured by the imaging means on a display means and manually search for the intersecting area while referring to the image, which is an unbearable hassle.

[0009] The present invention has been made in consideration of the above-mentioned facts, and its main technical object is to provide an intersection area detection method that can automatically detect an area where planned division lines intersect and display it on a display means. [Means for solving the problem]

[0010] In order to solve the above-mentioned main technical problem, according to the present invention, there is provided an intersection area detection method for detecting an area where a plurality of devices are formed on a surface partitioned by planned dividing lines specified by a pair of edges by an imaging means, the method comprising: placing a wafer on the chuck table of a processing apparatus comprising: a chuck table for holding a wafer; processing means for processing the wafer held on the chuck table; imaging means facing the chuck table and imaging the wafer held on the chuck table; X-axis moving means for moving the chuck table in an X-axis direction relative to the imaging means; Y-axis moving means for moving the chuck table in a Y-axis direction perpendicular to the X-axis direction relative to the imaging means; a wafer positioning step of positioning the wafer held on the chuck table directly under the imaging means; a straight line region detecting step of imaging the wafer with the imaging means and detecting a plurality of straight line regions by an image processing algorithm for performing a Hough transform provided in the control means; a first edge detecting step of detecting, from the plurality of straight line regions detected in the straight line region detecting step, two edges that are adjacent to each other within a predetermined range and have the most straight line components, as a pair of first edges that define a planned dividing line in a predetermined direction; and a second edge detecting step of detecting, from the plurality of straight line regions detected in the straight line region detecting step, a straight line region that intersects with the pair of first edges, and detecting, from the plurality of straight line regions detected, two edges that are adjacent to each other within a predetermined range and have the most straight line components, as a pair of second edges that intersect with the pair of first edges and define the planned dividing line. In the linear region detection step, if the linear components constituting the linear region having the most linear components are below an allowable value, the X-axis movement means and the Y-axis movement means are operated to change the imaging region to be imaged by the imaging means. A method for detecting an intersection region is provided.

[0011] It is preferable to further include a display step of rotating the chuck table to position the pair of first edges and the pair of second edges parallel to the X-axis direction and the Y-axis direction and displaying them on the display means. It is also preferable that when the first edge detection step and the second edge detection step are performed, the imaging means is set to a low magnification, and when the display step is performed, the imaging means is set to a higher magnification than when the first edge detection step and the second edge detection step are performed. 。 [Effects of the Invention]

[0012] The intersection area detection method of the present invention includes a wafer holding step of holding a wafer on the chuck table of a processing apparatus including a chuck table for holding a wafer, processing means for processing the wafer held on the chuck table, imaging means facing the chuck table and imaging the wafer held on the chuck table, X-axis moving means for moving the chuck table in the X-axis direction relative to the imaging means, Y-axis moving means for moving the chuck table in the Y-axis direction perpendicular to the X-axis direction relative to the imaging means, rotation means for rotating the chuck table in a plane defined by the X-axis and Y-axes, display means for displaying an image captured by the imaging means, and control means, and a wafer positioning step of positioning the wafer held on the chuck table directly below the imaging means. a first edge detection step of detecting, from the plurality of detected straight line regions, two edges that are adjacent within a predetermined range and have the most straight line components, as a pair of first edges that define a planned dividing line in a predetermined direction; and a second edge detection step of detecting, from the plurality of detected straight line regions, two edges that are adjacent within a predetermined range and have the most straight line components, as a pair of second edges that define the planned dividing line in a predetermined direction, in the straight line region detection step. In the linear region detection step, if the linear components constituting the linear region having the most linear components are below an allowable value, the X-axis movement means and the Y-axis movement means are operated to change the imaging region to be imaged by the imaging means.As a result, the coordinates of the area where the planned division line along the X-axis direction and the planned division line along the Y-axis direction intersect can be automatically detected by the action of the control means, and the area can be displayed on the display means, thereby eliminating the problem that the operator has to manually operate the X-axis moving means and Y-axis moving means to search for the intersecting area where the planned division lines intersect and display it on the display means, which is troublesome. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a perspective view of a laser processing apparatus suitable for carrying out the intersection area detection method of the present embodiment. [Figure 2] 2 is a perspective view showing an aspect in which a linear region detection step is carried out by the laser processing apparatus shown in FIG. 1. FIG. [Figure 3] FIG. 10 is a conceptual diagram showing an image displayed on a display means when a first edge detection step is performed. [Figure 4] FIG. 10 is a conceptual diagram showing an image displayed on a display means when a second edge detection step is performed. [Figure 5] 1A is a conceptual diagram showing a state in which an image captured by an imaging means is displayed on a display means at a low magnification, and FIG. 1B is a conceptual diagram showing a state in which an image captured by an imaging means is displayed on a display means at a high magnification. DETAILED DESCRIPTION OF THE INVENTION

[0014] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an intersection area detection method configured based on the present invention will be described in detail with reference to the accompanying drawings.

[0015] 1 shows a perspective view of a dicing apparatus 1 suitable for carrying out the intersection area detection method according to the present invention, and a wafer 10 that is diced by the dicing apparatus 1. In addition to the configuration shown in Fig. 1, the actual dicing apparatus 1 is equipped with a housing that covers the entire apparatus, a wafer cassette for carrying the wafer 10 into the dicing apparatus 1, a transport means, and the like, which are omitted from Fig. 1 for convenience of explanation.

[0016] The dicing apparatus 1 is equipped with a table mechanism 7 including a holding means 20 equipped with a chuck table 24 that holds the wafer 10 in an X-axis / Y-axis plane defined by the X-axis direction indicated by the arrow X in the figure and the Y-axis direction indicated by the arrow Y in the figure, an imaging means 11 that can image the wafer 10 held on the chuck table 24, and a cutting means 120 that cuts the wafer 10 held on the chuck table 24. In this embodiment, the wafer 10 is supported on an annular frame F via an adhesive tape T.

[0017] The table mechanism 7 is disposed on the stationary base 2 and includes a holding means 20 for holding the wafer 10, and an X-axis moving means 30 for supporting the holding means 20 and moving the chuck table 24 and the imaging means 11 relatively in the X-axis direction. Furthermore, a spindle support mechanism 8 is disposed on the stationary base 2 and includes a Y-axis moving means 50 for moving the chuck table 24 relatively in the Y-axis direction perpendicular to the X-axis direction with respect to the imaging means 11, and a spindle unit 9 is attached to the spindle support mechanism 8 and is disposed so as to be movable in the Z-axis direction indicated by arrow Z1 in the figure, which is perpendicular to the X-axis and Y-axis directions.

[0018] The holding means 20 includes a rectangular X-axis direction movable base 21 movably supported on a pair of guide rails 2a, 2a arranged parallel to each other along the X-axis direction on the stationary base 2, a cylindrical support member 22 arranged on the X-axis direction movable base 21, a rectangular cover plate 23 covering the periphery of the upper part of the cylindrical support member 22, a chuck table 24 arranged on the cylindrical support member 22, and a clamp mechanism 26 arranged between the cylindrical support member 22 and the chuck table 24 and gripping the frame F when the wafer 10 is supported on the chuck table 24. A motor (not shown) is built into the cylindrical support member 22 and functions as a rotation means for rotating the chuck table 24 on a plane defined by the X-axis and Y-axis. A pair of guide grooves 21a, 21a that slidably contact the guide rails 2a, 2a are formed on the underside of the X-axis direction moving base 21, and the guide rails 2a, 2a and the guide grooves 21a, 21a form a sliding portion.

[0019] The Y-axis movement means 50 includes a drive source 51 formed, for example, by a pulse motor, and a ball screw 52 that converts the rotation of the drive source 51 into linear motion and transmits it to a female screw portion (not shown) formed on the underside of the support portion 42 of the Y-axis direction movement base 40. By driving the ball screw 52 in the forward and reverse directions by the drive source 51, the Y-axis direction movement base 40 is moved in the Y-axis direction along the guide rails 2c, 2c.

[0020] The spindle unit 9 is mounted on the mounting wall 43 of the Y-axis direction movable base 40 and includes a Z-axis direction movable base 121, a spindle housing 122 supported by the Z-axis direction movable base 121, and cutting means 120 supported by the spindle housing 122, having a rotation shaft 122a in the Y-axis direction, and rotatably supporting a cutting blade 123 at the tip of the rotation shaft 122a. A spindle motor 125 that rotates and drives the rotation shaft 122a is disposed on the rear end side of the spindle housing 122. The Z-axis direction movable base 121 is supported on the mounting wall 43 of the Y-axis direction movable base 40 so as to be movable along a pair of guide rails 44, 44 disposed in the Z-axis direction (cutting feed direction). The Y-axis direction movable base 40 is provided with a pair of guide grooves 121a (only one of which is shown in the figure) that slidably abut on the guide rails 44, 44. The Y-axis direction moving base 40 includes a drive source 124 made up of a pulse motor and a ball screw (not shown) that converts the rotation of the drive source 124 into linear motion and transmits it to a female screw (not shown) formed in the Z-axis direction moving base 121, and the drive source 124 and the ball screw constitute a Z-axis feed means that feeds the cutting means 120 in the Z-axis direction, thereby moving the cutting means 120 in the Z-axis direction. As shown in the figure, the imaging means 11 is integrated with a spindle housing 122, and the cutting blade 123 of the cutting means 120 is positioned in the X-axis direction of a predetermined position (usually the center position) of the area imaged by the imaging means 11.

[0021] The dicing apparatus 1 is provided with a control means 100. The control means 100 is connected to the imaging means 11, the spindle motor 125 disposed at the rear end of the spindle housing 122, the drive source 31 of the X-axis moving means 30, the drive source 51 of the Y-axis moving means 50, the drive source 124 constituting the Z-axis feed means, a rotation means disposed on the cylindrical support member 22 and rotating the chuck table in a plane defined by the X-axis and Y-axis, and the like, as well as a display means 15 and the like.

[0022] The control means 100 is configured by a computer and includes a central processing unit (CPU) that performs calculations according to a control program, a read-only memory (ROM) that stores the control program, etc., a readable / writable random access memory (RAM) that temporarily stores the results of calculations performed by the control means 100 and image information captured by the imaging means 11, etc., an input interface, and an output interface (details not shown in the figure). Furthermore, although not shown in the figure, the X-axis direction moving base 21, the Y-axis direction moving base 40, the Z-axis direction moving base 121, and the above-mentioned rotation means are provided with position detection means, and the above-mentioned drive sources 31, 51, 124, and the rotation means are controlled based on pulse signals instructed by the control means 100, so that the positions of the X-axis direction moving base 21, the Y-axis direction moving base 40, the Z-axis direction moving base 121, and the rotation means are controlled to desired positions with high precision. The control means 100 also stores a control program for implementing the intersection area detection method described below, and the control program includes an image processing algorithm that performs a Hough transform to detect multiple straight line areas from a captured image.

[0023] The dicing apparatus 1 has roughly the configuration as described above, and the intersection area detection method carried out based on the present invention will be described below.

[0024] In carrying out the intersection area detection method of this embodiment, first, the wafer 10 to be detected is placed on the chuck table 24 of the dicing apparatus 1 shown in FIG. 1, and a suction means (not shown) is activated to generate negative pressure and hold the wafer by suction (wafer holding step). In the wafer holding step, a transport means is operated to carry the wafer 10 from a wafer cassette containing multiple wafers 10 to a temporary placement table (both not shown), and the wafer 10 is physically aligned on the temporary placement table. Then, the wafer 10 is transported from the temporary placement table by the transport means (not shown) and placed on the chuck table 24, whereby the wafer 10 can be automatically placed and held by suction so that the predetermined dividing lines 14 are aligned substantially along the X-axis direction and so that the dividing lines 14 perpendicular to the predetermined dividing lines 14 are aligned substantially along the Y-axis direction.

[0025] Next, based on an instruction signal from the control means 100, the X-axis moving means 30 and Y-axis moving means 50 are operated to position the wafer 10 held on the chuck table 24 directly below the imaging means 11, as shown in FIG. 2 (wafer positioning step). As shown in the figure, the wafer 10 in this embodiment is a semiconductor wafer having a plurality of devices 12 formed on its surface 10a, which is partitioned by planned division lines 14. The planned division lines 14 are defined by a pair of adjacent edges, and their width is, for example, 50 μm. For convenience of explanation, the chuck table 24, clamping mechanism 26, etc. are omitted from FIG. 2.

[0026] As described above, once the wafer 10 has been positioned directly below the imaging means 11 in the wafer positioning step, the imaging means 11 is used to image the wafer 10. The image of the wafer 10 thus captured is transmitted to the control means 100 and displayed on the display means 15, as shown in Fig. 2. In this embodiment, as described above, the intended dividing line 14 in a predetermined direction is placed and held so as to be aligned roughly along the X-axis direction, but the following description will be given assuming that there is a slight deviation in the rotational direction.

[0027] A known image processing algorithm for performing a Hough transform stored in the control means 100 is applied to the image of the wafer 10 captured by the imaging means 11 and displayed on the display means 15, thereby performing a linear region detection process to detect multiple linear regions. In the linear region detection process of this embodiment, first, as shown in FIG. 3(a), multiple linear regions, for example, L1 to L9, are detected in a predetermined direction, i.e., within a predetermined angular range along the X-axis direction in the figure, and the linear region with the most linear components is identified from the detected multiple linear regions. In the image shown in FIG. 3(a), the linear region with the most linear components is identified as linear region L2. Note that, for convenience of explanation, FIG. 3(a) typically shows only a portion of the detected linear regions, and the linear regions other than the detected linear regions L1 to L9 are indicated by dashed lines.

[0028] In the linear area detection process, if the linear components constituting the linear area L2 fall below a predetermined tolerance, more specifically, if the surface 10a of the wafer 10 is imaged by the imaging means 11 and the linear components of the linear area with the most linear components detected by the Hough transform are less than 70% of a predetermined length (for example, the length in the X-axis direction of the area of the imaged image), they are not recognized as edges constituting the planned dividing line 14, and it is determined that the imaged area of the surface 10a of the wafer 10 is not suitable for implementing the intersection area detection method, and for example, the Y-axis moving means 50 is operated to change the imaging area imaged by the imaging means 11 by a predetermined width.

[0029] In this embodiment, the linear region L2 detected in FIG. 3(a) is determined to be 70% or more of the length of the captured image in the X-axis direction and is recognized as a first edge constituting the planned division line 14. Here, the planned division line 14 is formed by two adjacent edges spaced 50 μm apart, as described above. It is determined whether or not there is an edge constituting the planned division line 14 in combination with the linear region L2 in the captured image area shown in FIG. 3(a). More specifically, although it is recognized that a linear region L3 exists at a position adjacent to the linear region L2 within a predetermined range (e.g., within 60 μm) of the linear region L2, the linear component of the linear region L3 is less than 70% of the length of the captured image in the X-axis direction. Therefore, it cannot be determined that the linear region L3, together with the linear region L2 recognized as the first edge, constitutes the planned division line 14. Therefore, the control unit 100 determines that the captured image area is inappropriate and, for example, operates the Y-axis moving unit 50 to change the captured image area by a predetermined width, as shown in FIG. 3(b).

[0030] After the above-mentioned change in the imaging area is performed and the image shown in Figure 3(b) is obtained, the straight line area detection process is performed again to detect a straight line area L2 from the detected multiple straight line areas, which has the most straight line components and whose straight line components exceed the predetermined tolerance of 70%, and a straight line area L3 that is close to the straight line area L2 within a predetermined range and whose straight line components exceed the predetermined tolerance of 70%, and detect both of them as a pair of first edges that define the planned division line 14 in a predetermined direction (first edge detection process).

[0031] After the first edge detection step is performed, an image processing algorithm that performs the Hough transform is used to detect straight line regions that extend along the direction intersecting the pair of first edges, i.e., the Y-axis direction, within a predetermined angle range. As shown in Fig. 4(a), two edges (straight line regions L10, L11) that have the most straight line components and are adjacent to each other within a predetermined range (within 60 µm) are detected as a pair of second edges that define the planned division line 14 intersecting the pair of first edges (second edge detection step). If the number of straight line regions found to detect the pair of second edges is less than a predetermined tolerance, i.e., if the number of straight line regions is less than 70% of the length of the captured image in the Y-axis direction, the captured image is determined to be inappropriate, and the second edge detection step is performed by, for example, changing the captured image region by a predetermined amount in the X-axis direction until the number of detected straight line components exceeds the tolerance.

[0032] As described above, by carrying out the linear region detection process including the first edge detection process and the second edge detection process, the coordinates of the region where the planned division line 14 along the X-axis direction and the planned division line 14 along the Y-axis direction intersect (the position indicated by point P in Figure 4(a)) are properly detected, and it becomes possible to display this region on the display means 15. This eliminates the need for the operator to manually operate the X-axis movement means and Y-axis movement means to search for the intersecting region where the planned division lines 14 intersect and display it on the display means 15, thereby solving the problem of it being too troublesome.

[0033] As described above, after the linear region detection step is performed and a pair of first edges (linear regions L2, L3) and a pair of second edges (linear regions L10, L11) are detected, the chuck table 24 is rotated in the direction indicated by arrow R in FIG. 2, and the pair of first edges and the pair of second edges are positioned parallel to the X-axis direction and the Y-axis direction, respectively, and displayed on the display means 15 (display step).

[0034] Furthermore, in this embodiment, when performing the above-mentioned display process, the X-axis moving means 30 and the Y-axis moving means 50 are operated to move the intersection area indicated by the point P where the planned division line 14 along the X-axis direction and the planned division line 14 along the Y-axis direction intersect to the center of the display means 15 as shown in FIG. 5(a), and then the imaging magnification by the imaging means 11 is set to a high magnification, and the intersection area is enlarged and displayed as shown in FIG. 5(b). That is, in this embodiment, the magnification when imaging the wafer 10 by the imaging means 11 in the linear region detection process is set to a relatively low magnification to make it easier to detect the linear region, and after detecting the intersection area, when displaying the intersection area on the display means 15, the magnification is set to a higher magnification than that set in the linear region detection process, and a function of detecting a target pattern (Patent Application No. 2021-74026) or a function of automatically creating a map (Patent Application No. 2021-98677) is executed based on the image of the wafer 10 captured by the imaging means 11.

[0035] In the above embodiment, an example is shown in which the intersection area detection method implemented based on the present invention is applied to a dicing device 1 that cuts the workpiece using a cutting blade, but the present invention is not limited to this, and can also be applied, for example, to a laser processing device that processes using a laser beam with a wavelength that is absorbent or transparent to the planned division line 14 of the wafer 10. [Explanation of symbols]

[0036] 1: Dicing equipment 2: Base 2a: Guide rail 2c: Guide rail 8: Spindle support mechanism 9: Spindle unit 10: Wafer 11: Imaging means 12: Device 14: Planned division line 15:Display means 20: Holding means 21:X-axis direction movement base 21a: Guide groove 22: Cylinder support member 24: Chuck table 26: Clamping mechanism 30:X-axis movement means 40: Y-axis movement base 42: Support part 43: Mounting wall 50: Y-axis movement means 51: Power source 52: Ball screw 100: Control means 120:Cutting means 121: Z-axis movement base 122: Spindle housing 122a: Rotating axis 123: Cutting blade 124: Power source 125: Spindle motor L1~L11: Straight line area

Claims

1. 1. An intersection area detection method for detecting an area where a plurality of devices are formed on a surface of a wafer partitioned by dividing lines defined by a pair of edges, by imaging means, the area including the dividing lines being identified by a pair of edges, the method comprising: a wafer holding step of holding a wafer on the chuck table of a processing apparatus including: a chuck table for holding a wafer; processing means for processing the wafer held on the chuck table; imaging means facing the chuck table and imaging the wafer held on the chuck table; X-axis moving means for moving the chuck table in the X-axis direction relative to the imaging means; Y-axis moving means for moving the chuck table in the Y-axis direction perpendicular to the X-axis direction relative to the imaging means; rotation means for rotating the chuck table in a plane defined by the X-axis and Y-axes; display means for displaying an image captured by the imaging means; and control means; a wafer positioning step of positioning the wafer held on the chuck table directly below the imaging means; a linear region detection step of capturing an image of the wafer with the imaging means and detecting a plurality of linear regions using an image processing algorithm that performs a Hough transform and is provided in the control means; a first edge detection step of detecting, from the plurality of detected straight line regions in the straight line region detection step, two edges that are adjacent to each other within a predetermined range and that are in a straight line region having the most straight line components, as a pair of first edges that define a planned division line in a predetermined direction; a second edge detection step of detecting, in the straight line region detection step, straight line regions intersecting the pair of first edges, and detecting, from the detected plurality of straight line regions, two edges that are straight line regions having the most straight line components and are adjacent to each other within a predetermined range, as a pair of second edges that intersect with the pair of first edges and define a planned division line; and when the linear components constituting the linear region having the most linear components are below an allowable value in the linear region detection step, the X-axis movement means and the Y-axis movement means are operated to change the imaging region to be imaged by the imaging means.

2. 2. The method for detecting an intersection area according to claim 1, further comprising a display step of rotating the chuck table to position the pair of first edges and the pair of second edges parallel to the X-axis direction and the Y-axis direction and displaying them on the display means.

3. 3. The intersection area detection method according to claim 2, wherein when the first edge detection step and the second edge detection step are performed, the imaging means is set to a low magnification, and when the display step is performed, the imaging means is set to a higher magnification than when the first edge detection step and the second edge detection step are performed.

Images