Processing device

Abstract

To provide a processing device capable of measuring a shape of a processed workpiece with high accuracy.SOLUTION: A processing device comprises: a table which holds a workpiece on a holding surface which is orthogonal to a Z axis; a processing unit which processes the workpiece on the table; an imaging unit which images a surface of the workpiece in an optical interference manner; a driving unit which moves the imaging unit relative to the table along a Z-axis direction; and an imaging control unit which controls the driving unit and the imaging unit and scans and images the surface of the workpiece on the table along the Z-axis direction. The imaging control unit performs imaging according to an imaging condition determined for each part to be imaged when scanning the surface of the workpiece along the Z-axis direction to image the surface.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a processing apparatus, and more particularly to a processing apparatus equipped with an imaging unit that captures images using an optical interference method. [Background technology]

[0002] Dicing devices (blade dicers) are known as processing devices that cut a workpiece using a blade (extremely thin peripheral blade) attached to a spindle that rotates at high speed. Patent Document 1 describes a dicing device equipped with a white light interferometer inside the device. According to the dicing device described in Patent Document 1, the shape of the kerf can be measured inside the device using the white light interferometer. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-084201 Summary of the Invention [Problem to be solved by the invention]

[0004] In optical interference type shape measuring instruments such as white light interferometers, measurements are performed by capturing so-called interference images. An interference image is captured by splitting light emitted from a light source into two, irradiating one side onto a reference surface and the other onto the surface of the object to be measured, and receiving the light reflected from both sides with an image sensor.

[0005] When grooves are measured with a shape measuring device that uses optical interference, the reflected light from inside the groove becomes weak, reducing the S / N ratio and causing noise to be erroneously recognized. In addition, when grooves are measured with a shape measuring device that uses optical interference, the reflected light from inside the groove may cause spurious interference fringes. These spurious interference fringes can cause instability and false detection when extracting the actual shape.

[0006] The present invention has been made in view of the above circumstances, and has an object to provide a processing apparatus capable of measuring the shape of a processed workpiece with high accuracy. [Means for solving the problem]

[0007] In order to solve the above problems, a first embodiment of the processing apparatus includes a table that holds a workpiece on a holding surface perpendicular to the Z axis, a processing unit that processes the workpiece on the table, an imaging unit that images the surface of the workpiece using an optical interference method, a drive unit that moves the imaging unit relatively to the table along the Z axis direction, and an imaging control unit that controls the drive unit and the imaging unit to scan and image the surface of the workpiece on the table in the Z axis direction, and when scanning and imaging the surface of the workpiece in the Z axis direction, the imaging control unit images according to imaging conditions defined for each part to be imaged.

[0008] A processing device of a second aspect is the processing device of the first aspect, in which the imaging control unit irradiates illumination light with a light amount determined for each portion and images the surface of the workpiece.

[0009] A processing device of a third aspect is the processing device of the first aspect, wherein the imaging control unit adjusts gain with a setting determined for each portion and images the surface of the workpiece.

[0010] A fourth aspect of the processing device is the processing device of the first aspect, wherein the imaging control unit controls the exposure time with a setting determined for each portion, and images the surface of the workpiece.

[0011] A fifth aspect of the processing device is the processing device of the first aspect, wherein the imaging control unit performs gradation correction with settings determined for each portion and images the surface of the workpiece.

[0012] A sixth aspect of the processing device is the processing device of the first aspect, wherein the imaging control unit outputs a pixel value from each pixel within an output range determined for each portion, and images the surface of the workpiece.

[0013] In the seventh aspect, when imaging a groove machined in the surface of a workpiece in the processing apparatus of the first aspect, the processing apparatus divides the workpiece into a first portion including the surface of the workpiece, a third portion including the bottom surface of the groove, and a second portion between the first portion and the third portion, and imaging conditions are determined for each of the divided portions.

[0014] The machining apparatus of the eighth aspect is the machining apparatus of the first aspect, further comprising an image processing unit that processes an image captured by scanning the surface of the workpiece in the Z-axis direction to measure the shape of the surface of the workpiece.

[0015] A ninth aspect of the processing apparatus is a processing apparatus of any one of the first to eighth aspects, further comprising a work information acquisition unit that acquires work information including information on the cross-sectional shape of the work, and an imaging condition setting unit that sets imaging conditions based on the work information.

[0016] The processing apparatus of the 10th aspect is the processing apparatus of the 9th aspect, further comprising an output range setting unit that sets an output range of the image based on information on the cross-sectional shape of the workpiece, and the imaging control unit extracts and outputs an image of the range set by the output range setting unit from the captured image.

[0017] The processing device of an eleventh aspect is a processing device of the first to eighth aspects, further comprising an output range setting unit that analyzes a captured image and sets an output range, and the imaging control unit extracts and outputs an image of the range set by the output range setting unit from the captured image.

[0018] A twelfth aspect of the processing device is the eleventh aspect of the processing device, in which the output range setting unit analyzes the captured image to recognize the interface of the workpiece, and sets the recognized interface and a specified range as the output range.

[0019] In the processing device of the 13th aspect, in the processing device of the 12th aspect, when imaging a groove machined in the surface of a workpiece, the output range setting unit recognizes the surface of the workpiece, the inner wall surface of the groove, and the bottom surface of the groove, and sets a predetermined range including each of the recognized surfaces as the output range.

[0020] A processing apparatus according to a fourteenth aspect is the processing apparatus according to any one of the first to eighth aspects, wherein the imaging unit images the surface of the workpiece by a white light interference method. Effect of the Invention

[0021] According to the present invention, the shape of a machined workpiece can be measured with high accuracy. [Brief description of the drawings]

[0022] [Figure 1] FIG. 1 is a front view showing an embodiment of a dicing device to which the present invention is applied. [Diagram 2] FIG. 2 is a diagram showing a schematic configuration of the table. [Diagram 3] FIG. 3 is a diagram showing a schematic configuration of the first imaging unit. [Figure 4] FIG. 4 is a diagram showing a schematic configuration of the second imaging unit. [Diagram 5] FIG. 5 is a block diagram of a control system of the dicing device. [Figure 6] FIG. 6 is a block diagram of functions possessed by the image processing unit. [Figure 7] FIG. 7 is a block diagram of the electrical configuration of the first imaging unit. [Figure 8] FIG. 8 is a block diagram of the electrical configuration of the second imaging unit. [Figure 9] FIG. 9 is a conceptual diagram of light amount control of illumination light. [Figure 10] FIG. 10 is a block diagram of functions that the second camera control unit has regarding control of the amount of illumination light. [Figure 11] FIG. 11 is a flowchart showing a procedure for measuring the shape of a kerf. [Figure 12] FIG. 12 is a flowchart showing the procedure of the imaging process of the interference image. [Figure 13] FIG. 13 is a block diagram of functions related to gain control that the second camera control unit has. [Figure 14] FIG. 14 is a flowchart showing the procedure of the imaging process of the interference image. [Figure 15] FIG. 15 is a block diagram of functions that the second camera control unit has regarding control of the exposure time. [Figure 16] FIG. 16 is a flowchart showing the procedure of the imaging process of the interference image. [Figure 17] FIG. 17 is a block diagram of functions that the second camera control unit has regarding control of tone correction. [Figure 18] FIG. 18 is a diagram showing an example of tone correction settings. [Figure 19] FIG. 19 is a flowchart showing the procedure of the imaging process of the interference image. [Figure 20] FIG. 20 is a block diagram of functions that the second camera control unit has regarding control of output of pixel values. [Figure 21] FIG. 21 is a flowchart showing the procedure of the imaging process of the interference image. [Figure 22] FIG. 22 is a diagram showing an example of the measurement results of the kerf of the wafer using a white light interferometer. [Figure 23] FIG. 23 is a conceptual diagram of setting the output range of an image. [Figure 24] FIG. 24 is a block diagram of functions that the second camera control unit has regarding control of image output. [Diagram 25] FIG. 25 is a conceptual diagram of setting the output range of an image. [Figure 26] FIG. 26 is a conceptual diagram of frame setting. [Figure 27] FIG. 27 is a block diagram of the control system for image output. [Figure 28] FIG. 28 is a diagram showing an example of a case where an output range of an image is set using a multi-joint frame. [Figure 29] FIG. 29 is a diagram showing another example of the frame for setting the output range of the image. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings.

[0024] [First embodiment] Here, an example will be described in which the present invention is applied to a dicing device. As described above, the dicing device is a device that cuts a wafer with a blade attached to a spindle that rotates at high speed. The wafer to be processed is, for example, a semiconductor wafer. On the surface of the semiconductor wafer, a large number of ICs (Integrated Circuits) and the like are formed in lattice-like areas partitioned by streets. The dicing device cuts the wafer along the streets and separates the individual ICs and the like into chips. The dicing device is an example of a processing device. The wafer is an example of a workpiece.

[0025] [Device configuration] Fig. 1 is a front view showing an embodiment of a dicing device to which the present invention is applied. In Fig. 1, the X-axis, the Y-axis, and the Z-axis are mutually orthogonal axes. The plane including the X-axis and the Y-axis constitutes a horizontal plane.

[0026] The dicing device 1 shown in FIG. 1 is a so-called twin-spindle dicer. A twin-spindle dicer is a dicing device equipped with a pair of spindles. A twin-spindle dicer can employ processing forms such as dual cut and step cut. Dual cut is a processing method that performs full cut or half cut on two lines simultaneously using a pair of blades. Full cut is a processing method that cuts into the fixed material such as dicing tape and completely cuts the object to be processed. Half cut is a groove cutting process that cuts into the middle of the thickness of the object to be processed. Step cut is a processing method that performs half cut and full cut in two stages.

[0027] 1, the dicing apparatus 1 has a base 2 serving as a base. On the base 2, a table 10 for holding a wafer W, an X-axis feed mechanism 30X for feeding the table 10 in the X-axis direction, and the like are provided.

[0028] FIG. 2 is a diagram showing a schematic configuration of the table.

[0029] 2, the table 10 has a disk-like shape and holds the wafer W on a horizontal holding surface 10A. As an example, the table 10 holds the wafer W by vacuum suction.

[0030] 2, a wafer W to be processed is mounted on a dicing frame DF and held on a table 10. The wafer W has a disk shape and is mounted on the dicing frame DF via a dicing tape DT.

[0031] The table 10 is driven by a table drive unit 11 to rotate around the θ axis. The θ axis passes through the center of the table 10 and is parallel to the Z axis. The table drive unit 11 includes a motor and a rotational position detector (not shown). The rotational position detector is formed of, for example, a rotary encoder, and detects the rotational position of the table 10.

[0032] The X-axis feed mechanism 30X is a mechanism for moving the table 10 in the X-axis direction. As shown in FIG. 1, the X-axis feed mechanism 30X is composed of an X-axis guide rail 31X, an X-axis table 32X, an X-axis actuator 33X, an X-axis position detector (not shown), and the like. The X-axis guide rail 31X is provided on the base 2 along the X-axis direction. The X-axis table 32X is provided movably on the X-axis guide rail 31X. The X-axis actuator 33X moves the X-axis table 32X along the X-axis guide rail 31X. The X-axis actuator 33X is composed of, for example, a linear motor. The X-axis position detector (not shown) detects the position of the X-axis table 32X. The X-axis position detector is composed of, for example, a linear scale.

[0033] Table 10 and table drive unit 11 are disposed on X-axis table 32X. Table 10 moves in the X-axis direction by driving X-axis actuator 33X to move X-axis table 32X in the X-axis direction. In addition, the position of table 10 in the X-axis direction is detected by an X-axis position detector (not shown).

[0034] 1, a gate-shaped column 3 is provided on the base 2, straddling the X-axis guide rail 31X. The column 3 is provided with a first processing unit 40A and a second processing unit 40B for processing a wafer W on a table 10, a first imaging unit 50 and a second imaging unit 60 for imaging the wafer W on the table 10, a first Y-axis feed mechanism 30YA for feeding the first processing unit 40A and the first imaging unit 50 in the Y-axis direction, a second Y-axis feed mechanism 30YB for feeding the second processing unit 40B and the second imaging unit 60 in the Y-axis direction, a first Z-axis feed mechanism 30ZA for feeding the first processing unit 40A and the first imaging unit 50 in the Z-axis direction, a second Z-axis feed mechanism 30ZB for feeding the second processing unit 40B and the second imaging unit 60B in the Z-axis direction, and the like.

[0035] The first processing unit 40A cuts the wafer W with a blade 41 rotating at high speed. The first processing unit 40A includes a first spindle 42A to which the blade 41 is attached, a first spindle motor 43A to rotate the first spindle 42A, and the like. The first spindle 42A is disposed parallel to the Y axis. The first spindle 42A includes a blade attachment portion at its tip. The blade 41 is detachably attached to the blade attachment portion. The blade 41 attached to the first spindle 42A rotates at high speed around its axis by driving the first spindle motor 43A to rotate the first spindle 42A at high speed.

[0036] The second processing unit 40B cuts the wafer W with a blade 41 rotating at high speed. The second processing unit 40B includes a second spindle 42B to which the blade 41 is attached, a second spindle motor 43B to rotate the second spindle 42B, and the like. The second spindle 42B is disposed parallel to the Y axis. The second spindle 42B includes a blade attachment portion at its tip. The blade 41 is detachably attached to the blade attachment portion. The blade 41 attached to the second spindle 42B rotates at high speed around its axis by driving the second spindle motor 43B to rotate the second spindle 42B at high speed.

[0037] The first imaging unit 50 is an imaging unit for alignment. The first imaging unit 50 captures an enlarged image of a part of the surface of the wafer W on the table 10 from directly above.

[0038] FIG. 3 is a diagram showing a schematic configuration of the first imaging unit.

[0039] The first imaging unit 50 includes a first microscope section 51 and a first camera section 56. The first imaging unit 50 captures an image magnified by the first microscope section 51 with the first camera section 56.

[0040] The first microscope unit 51 includes an illumination unit 52, a beam splitter 53, an objective lens 54, and an imaging lens 55, etc.

[0041] The illumination unit 52 has an illumination light source 52A and an illumination lens 52B. The illumination unit 52 emits light (illumination light) emitted from the illumination light source 52A via the illumination lens 52B. For example, a halogen lamp, a metal halide lamp, a mercury lamp, a xenon lamp, a light emitting diode (LED), or the like is used as the illumination light source 52A.

[0042] The light emitted from the illumination unit 52 is irradiated onto the wafer W on the table 10 via the beam splitter 53 and the objective lens 54. Then, the light reflected by the wafer W is incident on the first camera unit 56 via the objective lens 54, the beam splitter 53, and the imaging lens 55.

[0043] The first camera unit 56 has an image pickup element 56A, and electronically picks up the image magnified by the first microscope unit 51. For the image pickup element 56A, for example, an area image sensor such as a complementary metal oxide semiconductor image sensor (CMOS image sensor) or a charge-coupled device image sensor (CCD image sensor) is used. The image picked up by the first camera unit 56 is output to the image processing unit 110.

[0044] The second imaging unit 60 is an imaging unit for measurement. The second imaging unit 60 images the surface of the wafer W by optical interference. In particular, in this embodiment, the surface of the wafer W is imaged by white light interference using white light as a light source. By scanning and imaging the surface of the wafer W in the Z-axis direction (height direction) by the white light interference, the shape of the surface of the wafer W can be measured. Since the surface of the wafer W is imaged by the white light interference, the second imaging unit 60 essentially constitutes a white light interferometer.

[0045] FIG. 4 is a diagram showing a schematic configuration of the second imaging unit.

[0046] The second imaging unit 60 includes a second microscope section 61 and a second camera section 68. The second imaging unit 60 captures an image (interference image) observed by the second microscope section 61 with the second camera section 68. The second imaging unit 60 captures an image of the surface of the wafer W from above the table 10 along the Z-axis direction.

[0047] The second microscope section 61 is configured with a white light interference microscope. In this embodiment, the second microscope section 61 is configured with a so-called Mirau interference type white light interference microscope. As shown in FIG. 4, the second microscope section 61 includes an illumination section 62, a first beam splitter 63, an objective lens 64, a glass plate 65, a second beam splitter 66, and an imaging lens 67. The glass plate 65 has a reference mirror 65A in the center. The objective lens 64, the glass plate 65, and the second beam splitter 66 configure an interference optical system.

[0048] The illumination unit 62 has an illumination light source 62A and an illumination lens 62B. The illumination unit 62 emits white light emitted from the illumination light source 62A via the illumination lens 62B. The illumination light source 62A may be, for example, a halogen lamp, an LED, or the like.

[0049] The white light emitted from the illumination unit 62 passes through the first beam splitter 63, the objective lens 64, and the glass plate 65 and enters the second beam splitter 66. The white light entering the second beam splitter 66 is separated by the second beam splitter 66 into a measurement light and a reference light.

[0050] The measurement light passes through the second beam splitter 66 and is incident on the surface of the wafer W. Then, the measurement light reflected from the surface of the wafer W is incident on the second camera unit 68 via the second beam splitter 66, the glass plate 65, the objective lens 64, the first beam splitter 63, and the imaging lens 67.

[0051] The reference light is reflected by the second beam splitter 66 and enters the glass plate 65. The reference light that entered the glass plate 65 is reflected by the reference mirror 65A and enters the second beam splitter 66 again. The reference light is then reflected again by the second beam splitter 66 and enters the second camera unit 68 via the glass plate 65, the objective lens 64, the first beam splitter 63, and the imaging lens 67.

[0052] Interference light is generated by overlapping the measurement light reflected by the surface of the wafer W and incident on the second camera unit 68 with the reference light reflected by the reference mirror 65A and incident on the second camera unit 68. The interference optical system consisting of the objective lens 64, the glass plate 65, and the second beam splitter 66 is designed so that the optical path lengths of the measurement light and the reference light are equal when the focus is on the measurement target.

[0053] The second camera unit 68 has an image pickup element 68A, and electronically picks up an image (interference image) generated by the second microscope unit 61. For example, an area image sensor such as a CMOS image sensor or a CCD image sensor is used for the image pickup element 68A. The image (interference image) picked up by the second camera unit 68 is output to the image processing unit 110.

[0054] When the second imaging unit 60 is moved along the Z-axis direction (height direction), the optical path length of the measurement light reflected on the surface of the wafer W changes. The interference intensity of the interference lights incident on the imaging element 68A is maximized when the optical path lengths of the interference lights coincide with each other. Therefore, by reading the position in the Z-axis direction where the interference intensity is maximized for each pixel, the position of the unevenness on the surface of the wafer W can be measured.

[0055] The first Y-axis feed mechanism 30YA is a mechanism for moving the first processing unit 40A and the first imaging unit 50 in the Y-axis direction. As shown in FIG. 1, the first Y-axis feed mechanism 30YA is composed of a first Y-axis guide rail 31YA, a first Y-axis table 32YA, a first Y-axis actuator 33YA, and a first Y-axis position detector (not shown), etc. The first Y-axis guide rail 31YA is laid on the column 3 along the Y-axis direction. The first Y-axis table 32YA is provided so as to be movable on the first Y-axis guide rail 31YA. The first Y-axis actuator 33YA moves the first Y-axis table 32YA along the first Y-axis guide rail 31YA. The first Y-axis actuator 33YA is composed of, for example, a linear motor. The first Y-axis position detector (not shown) detects the position of the first Y-axis table 32YA. The first Y-axis position detector is composed of, for example, a linear scale.

[0056] The second Y-axis feed mechanism 30YB is a mechanism for moving the second processing unit 40B and the second imaging unit 60 in the Y-axis direction. As shown in FIG. 1, the second Y-axis feed mechanism 30YB is composed of a second Y-axis guide rail 31YB, a second Y-axis table 32YB, a second Y-axis actuator 33YB, and a second Y-axis position detector (not shown), etc. The second Y-axis guide rail 31YB is laid on the column 3 along the Y-axis direction. The second Y-axis table 32YB is provided so as to be movable on the second Y-axis guide rail 31YB. The second Y-axis actuator 33YB moves the second Y-axis table 32YB along the second Y-axis guide rail 31YB. The second Y-axis actuator 33YB is composed of, for example, a linear motor. The second Y-axis position detector (not shown) detects the position of the second Y-axis table 32YB. The second Y-axis position detector is composed of, for example, a linear scale.

[0057] The first Z-axis feed mechanism 30ZA is a mechanism for moving the first processing unit 40A and the first imaging unit 50 in the Z-axis direction. The first Z-axis feed mechanism 30ZA is composed of a first Z-axis guide rail 31ZA, a first Z-axis table 32ZA, a first Z-axis actuator 33ZA, a first Z-axis position detector (not shown), and the like. The first Z-axis guide rail 31ZA is laid on the first Y-axis table 32YA along the Z-axis direction. The first Z-axis table 32ZA is provided so as to be movable on the first Z-axis guide rail 31ZA. The first Z-axis actuator 33ZA moves the first Z-axis table 32ZA along the first Z-axis guide rail 31ZA. The first Z-axis actuator 33ZA is composed of, for example, a linear motor. The first Z-axis position detector (not shown) detects the position of the first Z-axis table 32ZA. The first Z-axis position detector is composed of, for example, a linear scale.

[0058] The second Z-axis feed mechanism 30ZB is a mechanism for moving the second processing unit 40B and the second imaging unit 60 in the Z-axis direction. The second Z-axis feed mechanism 30ZB is composed of a second Z-axis guide rail 31ZB, a second Z-axis table 32ZB, a second Z-axis actuator 33ZB, a second Z-axis position detector (not shown), and the like. The second Z-axis guide rail 31ZB is laid on the second Y-axis table 32YB along the Z-axis direction. The second Z-axis table 32ZB is provided so as to be movable on the second Z-axis guide rail 31ZB. The second Z-axis actuator 33ZB moves the second Z-axis table 32ZB along the second Z-axis guide rail 31ZB. The second Z-axis actuator 33ZB is composed of, for example, a linear motor. The second Z-axis position detector (not shown) detects the position of the second Z-axis table 32ZB. The second Z-axis position detector is composed of, for example, a linear scale.

[0059] The first processing unit 40A and the first imaging unit 50 are attached to the first Z-axis table 32ZA via the bracket 44A. As a result, when the first Z-axis table 32ZA is moved, the first processing unit 40A and the first imaging unit 50 move in the Z-axis direction (height direction). Also, when the first Y-axis table 32YA is moved, the first processing unit 40A and the first imaging unit 50 move in the Y-axis direction (horizontal direction). Also, by detecting the position of the first Z-axis table 32ZA in the Z-axis direction with a first Z-axis position detector (not shown), the position of the blade 41 attached to the first spindle 42A in the Z-axis direction and the position of the first imaging unit 50 in the Z-axis direction are detected. Also, by detecting the position of the first Y-axis table 32YA in the Y-axis direction with a first Y-axis position detector (not shown), the position of the blade 41 attached to the first spindle 42A in the Y-axis direction and the position of the first imaging unit 50 in the Y-axis direction are detected.

[0060] The second machining unit 40B and the second imaging unit 60 are attached to the second Z-axis table 32ZB via the bracket 44B. As a result, when the second Z-axis table 32ZB is moved, the second machining unit 40B and the second imaging unit 60 move in the Z-axis direction (height direction). Also, when the second Y-axis table 32YB is moved, the second machining unit 40B and the second imaging unit 60 move in the Y-axis direction (horizontal direction). Also, by detecting the position of the second Z-axis table 32ZB in the Z-axis direction with a second Z-axis position detector (not shown), the position of the blade 41 attached to the second spindle 42B in the Z-axis direction and the position of the second imaging unit 60 in the Z-axis direction are detected. Also, by detecting the position of the second Y-axis table 32YB in the Y-axis direction with a second Y-axis position detector (not shown), the position of the blade 41 attached to the second spindle 42B in the Y-axis direction and the position of the second imaging unit 60 in the Y-axis direction are detected.

[0061] FIG. 5 is a block diagram of a control system of the dicing device.

[0062] The dicing device 1 includes a system controller 100 that functions as an overall control unit, an image processing unit 110 that processes images captured by the first imaging unit 50 and the second imaging unit 60, an operation unit 120 that allows an operator to perform various operations, a display unit 130 that displays various information, a communication unit 140 that communicates with external devices, and the like. The form of communication is not particularly limited. It may be wired communication or wireless communication. It may also be a form of communication via a network.

[0063] The system controller 100 is composed of a computer equipped with a processor, memory, etc. That is, the computer functions as the system controller 100 by executing a predetermined program. The processor is, for example, a CPU (Central Processing Unit) or the like. The memory includes a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, etc.

[0064] Regarding the operation of table 10, the system controller 100 controls the table drive unit 11 to control the rotation of table 10. The system controller 100 also controls the X-axis feed mechanism 30X to control the feed of table 10 in the X-axis direction. More specifically, the system controller 100 controls the X-axis actuator 33X based on the output of an X-axis position detector (not shown) to control the feed of table 10 in the X-axis direction.

[0065] Regarding the operation of the first machining unit 40A, the system controller 100 controls the first Y-axis feed mechanism 30YA to control the feed (index feed) of the first machining unit 40A in the Y-axis direction. More specifically, based on the output of a first Y-axis position detector (not shown), the system controller 100 controls the first Y-axis actuator 33YA to control the feed of the first machining unit 40A in the Y-axis direction. In addition, the system controller 100 controls the first Z-axis feed mechanism 30ZA to control the feed of the first machining unit 40A in the Z-axis direction. More specifically, based on the output of a first Z-axis position detector (not shown), the system controller 100 controls the first Z-axis actuator 33ZA to control the feed of the first machining unit 40A in the Z-axis direction. Furthermore, the system controller 100 controls the first spindle motor 43A to control the drive of the first spindle 42A.

[0066] Regarding the operation of the second machining unit 40B, the system controller 100 controls the second Y-axis feed mechanism 30YB to control the feed in the Y-axis direction of the second machining unit 40B. More specifically, based on the output of a second Y-axis position detector (not shown), the system controller 100 controls the second Y-axis actuator 33YB to control the feed in the Y-axis direction of the second machining unit 40B. In addition, the system controller 100 controls the second Z-axis feed mechanism 30ZB to control the feed in the Z-axis direction of the second machining unit 40B. More specifically, based on the output of a second Z-axis position detector (not shown), the system controller 100 controls the second Z-axis actuator 33ZB to control the feed in the Z-axis direction of the second machining unit 40B. Furthermore, the system controller 100 controls the second spindle motor 43B to control the drive of the second spindle 42B.

[0067] By controlling the feed in the X-axis direction of the table 10, the feed in the cutting direction (cutting feed) is controlled. Moreover, by controlling the feed in the Y-axis direction of the first machining unit 40A and the second machining unit 40B, the feed in the direction perpendicular to the cutting direction (index feed) is controlled. Furthermore, by controlling the feed in the Z-axis direction of the first machining unit 40A and the second machining unit 40B, the feed in the cutting direction (cutting feed) is controlled. And by controlling the feed in the cutting direction, the cutting depth is controlled.

[0068] Moreover, the system controller 100 controls the first imaging unit 50 and the second imaging unit 60, and controls the imaging of the wafer W. The control of imaging will be described later.

[0069] The image processing unit 110 is configured with a computer equipped with a processor, a memory, and the like. That is, the computer executes a predetermined program to function as the image processing unit 110. Note that the computer constituting the image processing unit 110 and the computer constituting the system controller 100 may be the same. That is, the same computer may function as both the system controller 100 and the image processing unit 110.

[0070] FIG. 6 is a block diagram of functions possessed by the image processing unit.

[0071] As shown in FIG. 6, the image processing unit 110 has the functions of an alignment unit 110A, a kerf check unit 110B, and a kerf shape measuring unit 110C.

[0072] The alignment section 110A performs alignment processing in cooperation with the system controller 100. Alignment is an operation for determining the position of a street. The position of the street is determined, for example, based on an alignment mark. The alignment mark is a mark (also called an alignment target) that is attached to the surface of the wafer W for alignment purposes. The street is a cuttable area on the wafer W. The image captured by the first imaging unit 50 is processed, and the alignment mark is detected from within the image. Note that this type of technology is well known, so detailed explanations are omitted.

[0073] The kerf check section 110B performs a kerf check based on an image captured by the first imaging unit 50. The kerf check is a process for confirming whether the blade 41 is processing the correct position on the wafer W in an appropriate state. The kerf check includes detection of cutting deviation, chipping, and kerf width. The kerf check is performed at a preset position at a preset frequency. Note that the kerf check using an image is a known technique, and therefore a detailed description thereof will be omitted.

[0074] The kerf shape measuring section 110C processes the image (interference image) captured by the second imaging unit 60 to measure the kerf shape. Specifically, it processes a group of images (for example, a group of images having different focal planes with respect to the object) captured by scanning the second imaging unit 60 in the Z-axis direction to measure the three-dimensional shape of the measurement object. The measurement object is a kerf machined on the surface of the wafer W. As described above, the second imaging unit 60 is composed of a white light interferometer. Therefore, by scanning the measurement object in the Z-axis direction and capturing images with the second imaging unit 60, the three-dimensional shape of the measurement object can be measured from the obtained image group. Note that the process itself of obtaining the three-dimensional shape of the measurement object from the image group captured by the white light interferometer is a known technique. Therefore, a detailed description thereof will be omitted. The kerf shape measuring section 110C generates and outputs three-dimensional shape data of the measurement object.

[0075] FIG. 7 is a block diagram of the electrical configuration of the first imaging unit.

[0076] As shown in FIG. 7, the first camera section 56 includes an image sensor 56A, an analog signal processing section 56B, an ADC (Analog to Digital Converter) 56C, a digital signal processing section 56D, and a first camera control section 56E.

[0077] The analog signal processor 56B performs predetermined analog signal processing such as correlated double sampling and gain adjustment on the image signal output from the imaging element 56A.

[0078] The ADC 56C converts the analog image signal, which has been subjected to predetermined signal processing by the analog signal processor 56B, into a digital image signal.

[0079] The digital signal processor 56D performs various signal processing. The processing performed by the digital signal processor 56D includes processing for generating image data based on the signal output from the imaging element 56A. The processing for generating image data includes processing such as tone correction.

[0080] The first camera control unit 56E controls each unit of the first camera unit 56 and the first microscope unit 51, and controls imaging by the first imaging unit 50. The first camera control unit 56E is composed of a computer equipped with a processor, a memory, and the like.

[0081] The control performed by the first camera control unit 56E includes AF (Autofocus) control. The AF control is performed in cooperation with the system controller 100. That is, scanning is performed in the Z-axis direction to detect the position where the focus is on the surface of the wafer W (focus position). The focus position is detected, for example, by a contrast method (so-called contrast AF). By detecting the focus position with respect to the surface of the wafer W, the height position of the surface of the wafer W (position in the Z-axis direction) can be detected.

[0082] The control of the first microscope unit 51 performed by the first camera control unit 56E is the control of the illumination unit 52. More specifically, it is the control of the illumination light source 52A of the illumination unit 52. The first camera control unit 56E controls the amount of light (brightness) of the illumination light emitted from the illumination light source 52A.

[0083] FIG. 8 is a block diagram of the electrical configuration of the second imaging unit.

[0084] As shown in FIG. 8, the second camera section 68 includes an imaging element 68A, an analog signal processing section 68B, an ADC 68C, a digital signal processing section 68D, and a second camera control section 68E.

[0085] Analog signal processing section 68B performs predetermined analog signal processing such as correlated double sampling and gain adjustment on the image signal output from imaging element 68A.

[0086] The ADC 68C converts the analog image signal, which has been subjected to predetermined signal processing by the analog signal processor 68B, into a digital image signal.

[0087] The digital signal processor 68D performs various signal processing. The processing performed by the digital signal processor 68D includes processing for generating image data based on the signal output from the imaging element 68A. The processing for generating image data includes processing such as tone correction.

[0088] The second camera control unit 68E controls each part of the second camera unit 68 and the second microscope unit 61, and controls imaging by the second imaging unit 60. The second camera control unit 68E is composed of a computer equipped with a processor, a memory, and the like.

[0089] The imaging is controlled in cooperation with the system controller 100. As described above, imaging by the second imaging unit 60 is performed by scanning in the Z-axis direction. For this reason, the system controller 100 controls movement in the Z-axis direction, and the second camera control unit 68E controls imaging at each position. Therefore, in this embodiment, the second camera control unit 68E and the system controller 100 constitute an imaging control unit.

[0090] The control of the second microscope unit 61 performed by the second camera control unit 68E is the control of the illumination unit 62. More specifically, it is the control of the illumination light source 62A of the illumination unit 62. The second camera control unit 68E controls the amount of light (brightness) of the illumination light emitted from the illumination light source 62A. In particular, in this embodiment, the measurement object is divided into multiple parts along the scanning direction, and the amount of illumination light is controlled to change depending on the part to be imaged. A part is a part that occupies a certain position within the entire scanning range. Each part is set as a part that can be imaged under the same conditions. In this embodiment, the measurement object is divided into multiple parts along the scanning direction as parts that can be imaged with the same amount of illumination light.

[0091] Hereinafter, control of the amount of illumination light depending on the part to be imaged (image-taking part) will be described.

[0092] When measuring grooves with a shape measuring device that uses optical interference such as a white light interferometer, the reflected light from inside the grooves may be weak, resulting in a small S / N ratio between the desired interference fringes and noise. This may result in erroneous recognition of noise.

[0093] Therefore, in this embodiment, when imaging is performed by scanning in the Z-axis direction, the amount of illumination light is changed according to the imaging site (measurement site).

[0094] Fig. 9 is a conceptual diagram of light amount control of illumination light. Fig. 9 shows an example of measuring the shape of a kerf C machined on the surface of a wafer W.

[0095] Consider the case where the second imaging unit 60 is moved vertically downward (Z-axis scanning) to image the cut groove C with respect to the wafer W on the table 10. The imaging starts from the imaging start position Z0. The imaging start position Z0 is set at a position of a predetermined height H from the surface of the wafer W. In FIG. 9, the position Z1 is the position of the surface of the wafer W. Also, the position Z2 is the position of the bottom surface of the cut groove C. The position ZE is the imaging end position. The imaging end position is set at a position of a predetermined distance from the position of the bottom surface of the cut groove C.

[0096] In the present embodiment, the measurement target is divided into three parts along the scanning direction, and the light amount of the illumination light is set for each part. Specifically, the part up to the surface of the wafer W is the first part Pz1 (Z1 ≦ Pz1 ≦ Z0), the part from the surface of the wafer W to the bottom surface of the cut groove C is the second part Pz2 (Z2 < Pz2 < Z1), and the part after the bottom surface of the cut groove C is the third part Pz3 (Pz3 ≦ Z2), and the light amount of the illumination light is set for each part.

[0097] Let the light amount of the illumination light when imaging the first part Pz1 be the first light amount LV1, the light amount of the illumination light when imaging the second part Pz2 be the second light amount LV2, and the light amount of the illumination light when imaging the third part Pz3 be the third light amount LV3.

[0098] The imaging of the first part Pz1 is no different from the case of imaging a measurement target in normal air. Therefore, the first light amount LV1 is set to the same light amount as in the case of imaging a measurement target in normal air.

[0099] The imaging of the second part Pz2 is the imaging inside the cut groove C. Therefore, the second light amount LV2 is set to a value higher than the first light amount LV1 (LV1 < LV2).

[0100] The third part Pz3 is an image of the bottom surface of the cutting groove C. Therefore, it is preferable to set the third light quantity LV3 according to the state of the bottom surface of the cutting groove C. When the bottom surface is a smooth surface, reflected light may be obtained from the bottom surface. Therefore, in this case, the third light quantity LV3 is set to a value between the first light quantity LV1 and the second light quantity LV2 (LV1 < LV3 < LV2). When the bottom surface is not a smooth surface (when there is almost no reflected light from the bottom surface, or when it is darker than the inner wall surface of the cutting groove C), or when the depth is deep even if it is a smooth surface (when it is equal to or greater than the threshold value), the amount of reflected light decreases compared to the second part Pz2. Therefore, in this case, it is set to a value higher than the second light quantity LV2 (LV2 < LV3).

[0101] The second camera control unit 68E acquires information including the processed shape information of the cutting groove C (information on the cross-sectional shape of the cutting groove C) and information on the state of the bottom surface of the cutting groove C (workpiece information), and based on the acquired workpiece information, performs setting of parts and setting of the light quantity of the illumination light for each part. Then, based on the set parts and light quantity, the illumination unit 52 is controlled.

[0102] FIG. 10 is a block diagram of the functions of the second camera control unit regarding the control of the light quantity of the illumination light.

[0103] As shown in FIG. 10, the second camera control unit 68E has the functions of a workpiece information acquisition unit 68E1, an imaging condition setting unit 68E2, and an illumination control unit 68E3.

[0104] The workpiece information acquisition unit 68E1 acquires workpiece information. As described above, the workpiece information includes the processed shape information of the cutting groove C and the information on the state of the bottom surface of the cutting groove C. The workpiece information acquisition unit 68E1 acquires the workpiece information from the system controller 100. The system controller 100 acquires the workpiece information from the operator. The operator inputs the workpiece information via the operation unit 120 or the communication unit 140.

[0105] Information on the processed shape of the kerf C (so-called cross-sectional profile) is obtained, for example, by a simulation or the like. If the blade 41 to be used and the cutting depth are known, the cross-sectional shape (theoretical cross-sectional shape) of the processed kerf C can also be known. Therefore, the processed shape of the kerf C is obtained by a simulation or the like, and the information is used as information on the processed shape.

[0106] The system controller 100 may be configured to acquire information on the blade 41 to be used and information on the cutting depth from the operator, and automatically calculate information on the machining shape of the kerf C based on the acquired information.

[0107] In addition, it is also possible to measure the kerf C machined under the same conditions and use the results as information on the machined shape of the target.

[0108] Alternatively, the workpiece surface may be found by AF, and the vicinity of the third portion Pz3 may be scanned with the third light amount LV3 to find the third portion Pz3 first, and then the section of the second portion Pz2 may be determined.

[0109] The imaging condition setting unit 68E2 sets imaging conditions based on the work information. In this embodiment, the amount of illumination light irradiated from the illumination unit 62 is set for each part to be imaged. Specifically, the kerf C is divided into three parts (first part Pz1, second part Pz2, and third part Pz3) along the Z-axis direction, and the amount of illumination light (first light amount VL1, second light amount VL2, and third light amount VL3) during imaging is set for each part.

[0110] Information on the setting conditions of the light intensity determined for each part is stored in, for example, a memory.

[0111] The illumination control unit 68E3 controls the amount of illumination light emitted from the illumination unit 62 according to the light amount setting conditions by the imaging condition setting unit 68E2. More specifically, it controls the illumination light source 62A of the illumination unit 62 to control the amount of illumination light emitted from the illumination light source 62A. The illumination control unit 68E3 controls the illumination light source 62A based on the imaging position information so that the illumination light is emitted at the light amount set for each part.

[0112] The illumination control unit 68E3 acquires imaging position information from the system controller 100. The system controller 100 detects the imaging position based on the detection result of the second Z-axis position detector. In this embodiment, the imaging position is the position where the focus is on the measurement target (focus position). The imaging position is synonymous with the measurement position. Since the focus position is known in terms of the configuration, the imaging position can be obtained from the position of the second imaging unit 60 in the Z-axis direction.

[0113] [Effect] [Wafer processing] The wafer W is cut along the streets to be divided into chips.

[0114] First, alignment is performed. As described above, alignment is the operation of determining the position of the street. After alignment, processing begins.

[0115] First, the blade 41 is set to a predetermined height position and rotated at high speed. The height position of the blade 41 is set to a position that provides a predetermined cutting depth. After the blade 41 is set, a cutting feed (feed in the X-axis direction) is given to the wafer W. This causes the wafer W to be cut along the streets. After cutting, an index feed (feed in the Y-axis direction) is given to the blade 41. This makes it possible to cut the next street. In this manner, the cutting feed and index feed are alternately performed to cut the wafer W. When cutting of all the streets in the first direction is completed, the wafer W is rotated 90° and the streets in the second direction are cut.

[0116] The dicing device 1 of the present embodiment is equipped with two spindles (first spindle 42A and second spindle 42B), and is therefore capable of processing two locations simultaneously.

[0117] The cutting method is not particularly limited, and may be any method such as full cut (cutting into the dicing tape DT and completely cutting the wafer) or half cut (grooving into the middle of the wafer thickness).

[0118] [Measuring the shape of the groove] Next, a case where the shape of a machined kerf is measured will be described.

[0119] FIG. 11 is a flowchart showing a procedure for measuring the shape of a kerf.

[0120] First, workpiece information is acquired (step S1). As described above, the workpiece information includes information on the machined shape of the kerf C and information on the state of the bottom surface of the kerf C.

[0121] Next, imaging conditions are set based on the acquired workpiece information (step S2). In this embodiment, the amount of illumination light when imaging each part (the first part Pz1, the second part Pz2, and the third part Pz3) is set.

[0122] Next, an interference image is captured under the set imaging conditions (step S3).

[0123] FIG. 12 is a flowchart showing the procedure of the imaging process of the interference image.

[0124] First, the portion of the kerf to be measured is moved to the measurement position (step S3_A0). That is, it is moved to a position directly below the second imaging unit 60 (on the imaging optical axis).

[0125] Next, the second imaging unit 60 moves to the imaging start position (step S3_A1). As described above, the imaging start position Z0 is set at a position at a predetermined height H from the surface of the wafer W (see FIG. 9). The position of the surface of the wafer W is measured during the alignment process.

[0126] After moving to the imaging start position, the light amount of the illumination light is set to the first light amount LV1 (step S3_A2). Then, imaging of the interference image is started (step S3_A3). That is, while moving the second imaging unit 60 vertically downward at a constant pitch, imaging of the interference image is performed at constant intervals. More specifically, the interference image is captured by alternately repeating movement at a constant pitch and imaging.

[0127] After the imaging starts, it is determined whether the imaging position has reached the second portion Pz2 (step S3_A4). If it is determined that the imaging position has not reached the second portion Pz2, imaging continues to be performed at the first light amount LV1. On the other hand, if it is determined that the imaging position has reached the second portion Pz2, the light amount of the illumination light is switched to the second light amount (step S3_A5). Thereafter, imaging is performed at the second light amount LV2.

[0128] After the amount of illumination light is switched to the second amount of light LV2, it is determined whether the imaging position has reached the third portion Pz3 (step S3_A6). If it is determined that the third portion Pz3 has not been reached, imaging continues at the second amount of light LV2. On the other hand, if it is determined that the third portion Pz3 has been reached, the amount of illumination light is switched to the third amount of light (step S_A7). Thereafter, imaging is performed at the third amount of light LV3.

[0129] After that, it is determined whether the imaging position has reached the imaging end position (step S3_A8). If it is determined that the imaging end position has not been reached, imaging is continued with the third light amount LV3. On the other hand, if it is determined that the imaging end position has been reached, imaging is ended (step S3_A9).

[0130] This series of steps completes the capturing of the interference images by the second imaging unit 60. The interference images obtained by the capturing are output to the system controller 100 in sequence.

[0131] The system controller 100 outputs the captured image to the image processing unit 110, which processes the image (step S4). That is, the system controller 100 generates three-dimensional shape data of the target.

[0132] As described above, according to the dicing device 1 of the present embodiment, when capturing an interference image, the amount of illumination light is changed according to the portion to be captured. This allows the interference image to be captured with appropriate brightness. This also allows the shape of the kerf to be measured with high accuracy.

[0133] In the above example, the region to be imaged is divided into three regions, but the division of the region to be imaged is not limited to this. The region may be divided into more detailed regions. By dividing the region into more detailed regions, the amount of illumination light can be controlled more precisely.

[0134] Furthermore, when measuring the kerf C, it is preferable to set the amount of illumination light according to the width of the kerf C (kerf width).

[0135] [Second embodiment] In this embodiment, when an interference image is captured by the second imaging unit 60, the gain is automatically adjusted according to the part to be imaged.

[0136] Note that, except for the imaging conditions of the interference image, the configuration is the same as that of the dicing apparatus 1 of the first embodiment. Therefore, only the points related to the control of imaging of the interference image will be described below.

[0137] The gain adjustment is performed by the analog signal processing unit 68 B. The second camera control unit 68 E controls the analog signal processing unit 68 B so that the gain adjustment is performed according to the settings determined for each part.

[0138] FIG. 13 is a block diagram of functions related to gain control that the second camera control unit has.

[0139] As shown in FIG. 13, the second camera control unit 68E has the functions of a work information acquisition unit 68E1, an imaging condition setting unit 68E2, and a gain control unit 68E4.

[0140] The function of the workpiece information acquisition unit 68E1 is the same as that in the first embodiment, that is, it acquires workpiece information.

[0141] The imaging condition setting unit 68E2 sets imaging conditions based on the work information. In this embodiment, the gain adjustment setting is determined for each part to be imaged. The part setting is the same as that in the first embodiment (see FIG. 9).

[0142] The gain adjustment setting when imaging the first region Pz1 is "gain setting 1," the gain adjustment setting when imaging the second region Pz2 is "gain setting 2," and the gain adjustment setting when imaging the third region Pz3 is "gain setting 3."

[0143] Imaging of the first portion Pz1 is the same as imaging of a measurement target in normal air, and therefore the gain setting 1 is set to the same value as the gain when imaging a measurement target in normal air.

[0144] The imaging of the second portion Pz2 is an imaging of the inside of the kerf C. Therefore, the gain setting 2 is set to a value higher than the gain setting 1 (gain setting 1<gain setting 2).

[0145] The third region Pz3 is an image of the bottom surface of the kerf C. Therefore, it is preferable to set the gain setting 3 according to the state of the bottom surface of the kerf C. When the bottom surface is smooth (when there is light reflected from the bottom surface), the gain setting 3 is set to a value between the gain setting 1 and the gain setting 2 (gain setting 1<gain setting 3<gain setting 2). On the other hand, when the bottom surface is not smooth (when there is almost no light reflected from the bottom surface, or when it is darker than the inner wall surface of the kerf C), or when the depth of the kerf C is deep (when it is equal to or greater than the threshold value), the gain setting 3 is set to a value higher than the gain setting 2 (gain setting 2<gain setting 3).

[0146] Information on the gain settings determined for each part is stored in, for example, a memory.

[0147] The gain control unit 68E4 controls the analog signal processing unit 68B in accordance with the gain setting by the imaging condition setting unit 68E2. That is, the gain control unit 68E4 controls the analog signal processing unit 68B so that the gain adjustment is performed with the gain setting determined for each part.

[0148] FIG. 14 is a flowchart showing the procedure of the imaging process of the interference image.

[0149] First, the portion of the kerf to be measured is moved to the measurement position (step S3_B0).

[0150] Next, the second imaging unit 60 moves to the imaging start position (step S3_B1).

[0151] After moving to the imaging start position, the gain adjustment setting is set to "Gain setting 1" (Step S3_B2), and imaging of an interference image is started (Step S3_B3).

[0152] After imaging starts, it is determined whether the imaging position has reached the second portion Pz2 (step S3_B4). If it is determined that the imaging position has not reached the second portion Pz2, the interference image is continuously imaged with the gain setting 1. That is, gain adjustment is performed with the gain setting 1, and the interference image is imaged. On the other hand, if it is determined that the second portion Pz2 has been reached, the gain adjustment setting is switched to "gain setting 2" (step S3_B5). Thereafter, gain adjustment is performed with the gain setting 2, and the interference image is imaged.

[0153] After the gain adjustment setting is switched to gain setting 2, it is determined whether the imaging position has reached the third portion Pz3 (step S3_B6). If it is determined that the imaging position has not reached the third portion Pz3, the interference image is captured while continuing with gain setting 2. That is, gain adjustment is performed with gain setting 2, and the interference image is captured. On the other hand, if it is determined that the third portion Pz3 has been reached, the gain adjustment setting is switched to "gain setting 3" (step S_B7). Thereafter, gain adjustment is performed with gain setting 3, and the interference image is captured.

[0154] Thereafter, it is determined whether the imaging position has reached the imaging end position (step S3_B8). If it is determined that the imaging end position has not been reached, the interference image is continuously captured with the gain setting 3. That is, the gain adjustment is performed with the gain setting 3, and the interference image is captured. On the other hand, if it is determined that the imaging end position has been reached, the imaging is terminated (step S3_B9).

[0155] As described above, according to the dicing device 1 of the present embodiment, the gain adjustment setting can be switched depending on the part to be imaged. This makes it possible to acquire an image with appropriate brightness. This also makes it possible to measure the shape of the kerf with high accuracy.

[0156] In this example, the gain adjustment is performed on an analog signal (so-called analog gain adjustment), but the gain adjustment may be performed on a digital signal (so-called digital gain adjustment).

[0157] In the above example, the region to be imaged is divided into three regions, but the manner of dividing the region to be imaged is not limited to this. The region may be divided into more detailed regions. By dividing the region into more detailed regions, the gain can be adjusted more precisely.

[0158] In addition, when measuring the kerf C, it is preferable to set a gain according to the width of the kerf C.

[0159] [Third embodiment] In this embodiment, when an interference image is captured by the second imaging unit 60, the exposure time is automatically adjusted according to the part to be imaged.

[0160] Note that, except for the imaging conditions of the interference image, the configuration is the same as that of the dicing apparatus 1 of the first embodiment. Therefore, only the points related to the control of imaging of the interference image will be described below.

[0161] As an example, in the present embodiment, imaging element 68A has a function of a so-called electronic shutter. Second camera control unit 68E controls imaging element 68A so that imaging is performed at an exposure time determined for each part.

[0162] FIG. 15 is a block diagram of functions that the second camera control unit has regarding control of the exposure time.

[0163] As shown in FIG. 15, the second camera control unit 68E has the functions of a work information acquisition unit 68E1, an imaging condition setting unit 68E2, and an exposure control unit 68E5.

[0164] The function of the workpiece information acquisition unit 68E1 is the same as that in the first embodiment, that is, it acquires workpiece information.

[0165] The imaging condition setting unit 68E2 sets imaging conditions based on the work information. In this embodiment, the exposure time during imaging is determined for each part to be imaged. The setting of the part is the same as that in the first embodiment (see FIG. 9).

[0166] The exposure time when imaging the first portion Pz1 is the "first time T1," the exposure time when imaging the second portion Pz2 is the "second time T2," and the exposure time when imaging the third portion Pz3 is the "third time T3."

[0167] The imaging of the first portion Pz1 is the same as the imaging of a measurement target in normal air, and therefore the first time T1 is set to the same exposure time as the imaging of a measurement target in normal air.

[0168] The imaging of the second portion Pz2 is an imaging of the inside of the kerf C. Therefore, the second time T2 is set to be longer than the first time T1 (first time T1<second time T2).

[0169] The third portion Pz3 is an image of the bottom surface of the kerf C. Therefore, the third time T3 is preferably set according to the condition of the bottom surface of the kerf C. When the bottom surface is smooth (when there is light reflected from the bottom surface), the third time T3 is set to a time between the first time T1 and the second time T2 (first time T1<third time T3<second time T2). On the other hand, when the bottom surface is not smooth (when there is almost no light reflected from the bottom surface, or when it is darker than the inner wall surface of the kerf C), or when the depth of the kerf C is deep (when it is greater than or equal to a threshold value), the third time T3 is set to a time longer than the second time T2 (second time T2<third time T3).

[0170] Setting information on the exposure time determined for each part is stored in, for example, a memory.

[0171] The exposure control unit 68E5 controls the driving of the imaging element 68A according to the exposure time set by the imaging condition setting unit 68E2. That is, the exposure control unit 68E5 controls the driving of the imaging element 68A so that each part is exposed for the exposure time set for that part.

[0172] FIG. 16 is a flowchart showing the procedure of the imaging process of the interference image.

[0173] First, a portion of the kerf to be measured is moved to a measurement position (step S3_C0).

[0174] Next, the second imaging unit 60 moves to the imaging start position (Step S3_C1).

[0175] After moving to the imaging start position, the exposure time is set to the first time T1 (step S3_C2), and imaging of an interference image is started (step S3_C3).

[0176] After the imaging starts, it is determined whether the imaging position has reached the second portion Pz2 (step S3_C4). If it is determined that the imaging position has not reached the second portion Pz2, the exposure time is continued at the first time T1 and an interference image is captured. On the other hand, if it is determined that the imaging position has reached the second portion Pz2, the exposure time is switched to the second time T2 (step S3_C5). Thereafter, an interference image is captured with the exposure time set to the second time T2.

[0177] After the exposure time is switched to the second time T2, it is determined whether the imaging position has reached the third portion Pz3 (step S3_C6). If it is determined that the imaging position has not reached the third portion Pz3, the exposure time continues at the second time T2 and an interference image is captured. On the other hand, if it is determined that the imaging position has reached the third portion Pz3, the exposure time is switched to the third time T3 (step S_C7). Thereafter, an interference image is captured with the exposure time set to the third time T3.

[0178] Thereafter, it is determined whether the imaging position has reached the imaging end position (step S3_C8). If it is determined that the imaging end position has not been reached, the interference image is continuously captured for the third time T3. On the other hand, if it is determined that the imaging end position has been reached, imaging is terminated (step S3_C9).

[0179] As described above, according to the dicing device 1 of the present embodiment, the exposure time can be switched depending on the part to be imaged. This makes it possible to acquire an image with appropriate brightness. This also makes it possible to measure the shape of the kerf with high accuracy.

[0180] In the above example, the case where the part to be imaged is divided into three parts has been described as an example, but the manner of dividing the part to be imaged is not limited to this. It is also possible to configure the part to be divided more finely. By dividing the part more finely, the exposure time can be controlled more finely.

[0181] In addition, when measuring the kerf C, it is preferable to set the exposure time according to the width of the kerf C.

[0182] [Fourth embodiment] In this embodiment, when an interference image is captured by the second imaging unit 60, the gradation is automatically corrected according to the part to be imaged.

[0183] Note that, except for the imaging conditions of the interference image, the configuration is the same as that of the dicing apparatus 1 of the first embodiment. Therefore, only the points related to the control of imaging of the interference image will be described below.

[0184] The gradation correction is performed in digital signal processing unit 68D. Second camera control unit 68E controls digital signal processing unit 68D so that the gradation correction is performed according to the settings determined for each part.

[0185] FIG. 17 is a block diagram of functions that the second camera control unit has regarding control of tone correction.

[0186] As shown in FIG. 17, the second camera control unit 68E has the functions of a work information acquisition unit 68E1, an imaging condition setting unit 68E2, and a gradation control unit 68E6.

[0187] The function of the workpiece information acquisition unit 68E1 is the same as that in the first embodiment, that is, it acquires workpiece information.

[0188] The imaging condition setting unit 68E2 sets imaging conditions based on the work information. In this embodiment, the setting of the gradation correction is determined for each part to be imaged. The setting of the part is the same as that in the first embodiment (see FIG. 9).

[0189] The gradation correction setting when imaging the first portion Pz1 is "gradation setting 1," the gradation correction setting when imaging the second portion Pz2 is "gradation setting 2," and the gradation correction setting when imaging the third portion Pz3 is "gradation setting 3."

[0190] Fig. 18 is a diagram showing an example of gradation correction settings. Fig. 18 shows an example in which the image to be captured is an 8-bit image. That is, it shows an example in which the gradation of the image is expressed in 256 gradations.

[0191] The imaging of the first portion Pz1 is the same as when imaging a measurement target in normal air. Therefore, the gradation setting 1 is set to the same setting as the gradation correction setting when imaging a measurement target in normal air. Fig. 18(A) shows an example of the gradation conversion characteristic of the gradation setting 1. In this example, the setting is set to have a linear relationship between input and output.

[0192] The imaging of the second portion Pz2 is an imaging of the inside of the cutting groove C. Therefore, the gradation setting 2 is the setting for imaging a dark area. Specifically, the setting is to amplify dark input data. FIG. 18(B) shows an example of the gradation conversion characteristics of the gradation setting 2. In this example, the setting is such that dark input data is amplified. In the example shown in FIG. 18(B), the setting is such that an input of 100 results in a maximum (255).

[0193] The third portion Pz3 is an image of the bottom surface of the cut groove C. Therefore, it is preferable to set the gradation setting 3 according to the state of the bottom surface of the cut groove C. When the bottom surface is smooth (when there is reflected light from the bottom surface), the setting is intermediate between the gradation setting 1 and the gradation setting 2. FIG. 18(C) shows an example of the gradation conversion characteristic of the gradation setting 3 in this case. In the example shown in the figure, the setting is set to the maximum (255) at the input of 150. On the other hand, when the bottom surface is not smooth (when there is almost no reflected light from the bottom surface, or when it is darker than the inner wall surface of the cut groove C), or when the depth of the cut groove C is deep (when it is above the threshold), the setting is set to capture an area darker than the gradation setting 2. FIG. 18(D) shows an example of the gradation conversion characteristic of the gradation setting 3 in this case. In the example shown in the figure, the setting is set to the maximum (255) at the input of 50.

[0194] FIG. 18 shows an example in which the captured image is an 8-bit image, but if the captured image is a 14-bit image, the image is expressed in 16384 gradations (0 to 16383).

[0195] The gradation correction is performed by using a table (gradation conversion table) such as a lookup table (LUT), etc. In this case, a gradation conversion table to be used is determined for each part to be imaged.

[0196] Information on the tone correction settings (tone conversion table settings to be used) determined for each part is stored in, for example, a memory.

[0197] The gradation control unit 68E6 controls the digital signal processing unit 68D so that gradation correction is performed for each part according to the gradation correction settings determined by the imaging condition setting unit 68E2.

[0198] FIG. 19 is a flowchart showing the procedure of the imaging process of the interference image.

[0199] First, a portion of the kerf to be measured is moved to a measurement position (step S3_D0).

[0200] Next, the second imaging unit 60 moves to the imaging start position (Step S3_D1).

[0201] After moving to the imaging start position, the tone correction setting is set to "tone setting 1" (step S3_D2), and imaging of an interference image is started (step S3_D3).

[0202] After imaging starts, it is determined whether the imaging position has reached the second portion Pz2 (step S3_D4). If it is determined that the imaging position has not reached the second portion Pz2, the interference image is captured by continuing with gradation setting 1. That is, gradation correction is performed with gradation setting 1, and the interference image is captured. On the other hand, if it is determined that the second portion Pz2 has been reached, the gradation correction setting is switched to "gradation setting 2" (step S3_D5). Thereafter, gradation correction is performed with gradation setting 2, and the interference image is captured.

[0203] After the gradation correction setting is switched to gradation setting 2, it is determined whether the imaging position has reached the third portion Pz3 (step S3_D6). If it is determined that the third portion Pz3 has not been reached, the interference image is captured while continuing with gradation setting 2. That is, gradation correction is performed with gradation setting 2, and the interference image is captured. On the other hand, if it is determined that the third portion Pz3 has been reached, the gradation correction setting is switched to "gradation setting 3" (step S_D7). Thereafter, gradation correction is performed with gradation setting 3, and the interference image is captured.

[0204] Thereafter, it is determined whether the imaging position has reached the imaging end position (step S3_D8). If it is determined that the imaging end position has not been reached, the interference image is continuously captured with the gradation setting 3. That is, the gradation correction is performed with the gradation setting 3, and the interference image is captured. On the other hand, if it is determined that the imaging end position has been reached, the imaging is terminated (step S3_D9).

[0205] As described above, according to the dicing device 1 of the present embodiment, the gradation correction setting can be switched depending on the part to be imaged. This makes it possible to acquire an image with appropriate brightness. This also makes it possible to measure the shape of the kerf with high accuracy.

[0206] In the above example, the case where the part to be imaged is divided into three parts has been described as an example, but the manner of dividing the part to be imaged is not limited to this. It is also possible to configure the part to be divided in more detail. By dividing it in more detail, the gradation of the image to be imaged can be adjusted in more detail.

[0207] In addition, when measuring the kerf C, it is preferable to change the setting of the gradation correction according to the width of the kerf C.

[0208] [Fifth embodiment] In this embodiment, when an interference image is captured by the second imaging unit 60, the output range of pixel values ​​(luminance values) is switched depending on the part to be imaged.

[0209] Note that, except for the imaging conditions of the interference image, the configuration is the same as that of the dicing apparatus 1 of the first embodiment. Therefore, only the points related to the control of imaging of the interference image will be described below.

[0210] The process of adjusting the output range of pixel values ​​is performed by digital signal processor 68 D. Second camera controller 68 E controls digital signal processor 68 D so that pixel values ​​are output within the output range determined for each part.

[0211] FIG. 20 is a block diagram of functions that the second camera control unit has regarding control of output of pixel values.

[0212] As shown in FIG. 20, the second camera control unit 68E has the functions of a work information acquisition unit 68E1, an imaging condition setting unit 68E2, and a pixel output control unit 68E7.

[0213] The function of the workpiece information acquisition unit 68E1 is the same as that in the first embodiment, that is, it acquires workpiece information.

[0214] The imaging condition setting unit 68E2 sets imaging conditions based on the work information. In this embodiment, the output range of pixel values ​​is determined for each part to be imaged. The part is set in the same manner as in the first embodiment (see FIG. 9).

[0215] The output range setting when imaging the first portion Pz1 is "output setting 1," the output range setting when imaging the second portion Pz2 is "output setting 2," and the output range setting when imaging the third portion Pz3 is "output setting 3."

[0216] The imaging of the first portion Pz1 is the same as when imaging a measurement target in normal air. Therefore, the output setting 1 is the same as when imaging a measurement target in normal air. For example, in the case of an 8-bit image (an image with pixel values ​​of 0 to 255), a pixel value of 100 or more is output (pixels with pixel values ​​less than 100 output pixel values ​​as 0 or NULL).

[0217] The imaging of the second portion Pz2 is imaging of the inside of the kerf C. Therefore, output setting 2 is a setting for imaging a dark area. For example, in the case of an 8-bit image, the setting is to output pixel values ​​equal to or greater than 0 and less than 100 (pixels with pixel values ​​equal to or greater than 100 are output as 0 or NULL).

[0218] The third portion Pz3 is an image of the bottom surface of the kerf C. Therefore, it is preferable to set the output setting 3 according to the state of the bottom surface of the kerf C. When the bottom surface is smooth (when there is reflected light from the bottom surface), the output setting 3 is set to an intermediate setting between the output setting 1 and the output setting 2. For example, in the case of an 8-bit image, the setting is set to output a pixel value of 50 or more and less than 150 (pixels with pixel values ​​less than 50 and 150 or more output pixel values ​​as 0 or NULL). On the other hand, when the bottom surface is not smooth (when there is almost no reflected light from the bottom surface, or when it is darker than the inner wall surface of the kerf C), or when the depth of the kerf C is deep (when it is equal to or greater than the threshold value), the setting is set to capture an area darker than the output setting 2. For example, in the case of an 8-bit image, the setting is set to output a pixel value of 0 or more and less than 50 (pixels with pixel values ​​of 50 or more output pixel values ​​as 0 or NULL).

[0219] The output range of pixel values ​​is controlled by, for example, filter processing. In this case, a filter to be used is set for each part. Information on the setting of the output range of pixel values ​​determined for each part (setting of the filter to be used) is stored in, for example, a memory.

[0220] The pixel output control unit 68E7 controls the digital signal processing unit 68D so that pixel values ​​are output according to the settings determined by the imaging condition setting unit 68E2. By limiting the output range of pixel values, image data is generated in which the pixel values ​​of the limited pixels are set to 0 or NULL.

[0221] FIG. 21 is a flowchart showing the procedure of the imaging process of the interference image.

[0222] First, a portion of the kerf to be measured is moved to a measurement position (step S3_E0).

[0223] Next, the second imaging unit 60 moves to the imaging start position (step S3_E1).

[0224] After moving to the imaging start position, the output range is set to "output setting 1" (step S3_E2), and imaging of an interference image is started (step S3_E3).

[0225] After imaging starts, it is determined whether the imaging position has reached the second portion Pz2 (step S3_E4). If it is determined that the imaging position has not reached the second portion Pz2, the interference image is continuously captured with output setting 1. That is, pixel values ​​are output within the output range determined by output setting 1. On the other hand, if it is determined that the imaging position has reached the second portion Pz2, the output range setting is switched to "output setting 2" (step S3_E5). Thereafter, pixel values ​​are output within the output range determined by output setting 2.

[0226] After the output range setting is switched to output setting 2, it is determined whether the imaging position has reached the third portion Pz3 (step S3_E6). If it is determined that the imaging position has not reached the third portion Pz3, the interference image is continuously captured with output setting 2. That is, pixel values ​​are output within the output range determined by output setting 2. On the other hand, if it is determined that the third portion Pz3 has been reached, the output range setting is switched to "output setting 3" (step S_E7). Thereafter, pixel values ​​are output within the output range determined by output setting 3.

[0227] Thereafter, it is determined whether the imaging position has reached the imaging end position (step S3_E8). If it is determined that the imaging end position has not been reached, the interference image is continuously captured with the output setting 3. That is, pixel values ​​are output within the output range determined by the output setting 3. On the other hand, if it is determined that the imaging end position has been reached, imaging is terminated (step S3_E9).

[0228] As described above, according to the dicing device 1 of the present embodiment, the output range of pixel values ​​is limited according to the part to be imaged. This makes it possible to suppress the inclusion of noise components in the image. This also makes it possible to measure the shape of the kerf with high accuracy.

[0229] In the above example, the region to be imaged is divided into three regions, but the division of the region to be imaged is not limited to this. The region may be divided in more detail. By dividing the region in more detail, the output range of pixel values ​​can be controlled in more detail.

[0230] Furthermore, when measuring the kerf C, it is preferable to set an output range of pixel values ​​according to the width of the kerf C.

[0231] [Sixth embodiment] 22 is a diagram showing an example of the measurement results of the kerf of the wafer by the white light interferometer, in which the three-dimensional shape data is projected onto a two-dimensional cross section (the YZ cross section of the kerf C).

[0232] As shown in Figure 22, when a kerf C machined on the surface of a wafer is measured with a white light interferometer, noise may occur near the surface of the wafer W. Also, interference fringes may be reflected inside the groove near the surface. Furthermore, ghost-like interference fringes may occur inside the groove.

[0233] These noises and interference fringes that occur in areas where there is nothing can cause instability and false detection when extracting the actual shape.

[0234] Therefore, in this embodiment, only the area necessary for shape extraction is extracted from the captured interference image, and the extracted image is output as a captured image to the system controller 100. In other words, the area unnecessary for measurement is cut off and output. This makes it possible to eliminate unstable factors and erroneous detection factors when extracting the original shape.

[0235] FIG. 23 is a conceptual diagram of setting the output range of an image.

[0236] Fig. 23 shows an example of measuring the shape of a kerf C. In Fig. 23, a thick line L indicates the contour shape of a cross section (YZ cross section) of the kerf C. Moreover, a shaded area (hatched area) indicates an example of an output range to be set.

[0237] When measuring the kerf C, the positions of the surface Fa1 of the wafer, the inner wall surface Fa2 of the kerf C, and the bottom surface Fa3 of the kerf C are extracted from the image. Therefore, a range including these surfaces is set as the output range of the image. In the example shown in FIG. 23, a range of a certain width centered on the position of the surface to be extracted is set as the output range. This makes it possible to extract and output only the area necessary for shape extraction.

[0238] Setting and control of the output range of the image is performed, for example, by second camera control unit 68E. Second camera control unit 68E sets the output range of the image based on, for example, work information.

[0239] The process of adjusting the output range of the image is performed by digital signal processor 68 D. Second camera controller 68 E controls digital signal processor 68 D so that the image is output according to the settings.

[0240] FIG. 24 is a block diagram of functions that the second camera control unit has regarding control of image output.

[0241] As shown in FIG. 24, the second camera control unit 68E has the functions of a work information acquisition unit 68E1, an image output range setting unit 68E8, and an image output control unit 68E9.

[0242] The function of the workpiece information acquisition unit 68E1 is the same as that in the first embodiment, that is, it acquires workpiece information.

[0243] The image output range setting unit 68E3 sets the image output range based on the workpiece information. The image output range is set for each imaging position in the Z-axis direction.

[0244] As described above, the image output range is set so as to include the surfaces to be extracted (the surface of the wafer W, the inner wall surface of the kerf C, and the bottom surface of the kerf C) (see FIG. 23). For example, a range of a certain width is set as the image output range based on the position of the surface to be extracted. The image output range is set based on information on the machining shape of the kerf C (information on the cross-sectional shape of the kerf C).

[0245] The image output control unit 68E9 controls the digital signal processing unit 68D so that an image is output according to the output range set by the image output range setting unit 68E8. That is, the image output control unit 68E9 controls the digital signal processing unit 68D so that an image is output within the output range determined for each imaging position in the Z-axis direction.

[0246] The digital signal processor 68D outputs the image by setting the pixel values ​​of pixels outside the output range to 0 or NULL. The output image is applied to the image processor 110 and used to generate three-dimensional shape data.

[0247] In this way, according to this embodiment, only the area necessary for shape extraction is extracted from the captured interference image, and the extracted image is output as the captured image. This allows areas unnecessary for measurement to be cut off and output. This makes it possible to eliminate unstable factors and erroneous detection factors when extracting the original shape, and enables stable measurement.

[0248] In the above example, the second imaging unit 60 is configured to perform the process of adjusting the output range of the image, but the same process may be performed in a unit other than the second imaging unit 60. For example, the process may be performed in the system controller 100. Alternatively, the process may be performed in a unit provided separately, or in the image processing unit 110.

[0249] [Seventh embodiment] In this embodiment, the output range of the image is automatically set based on the captured image. Specifically, the position of the interface of the measurement target (or the intersection of a specific surface and another surface that intersects with this surface) is recognized from the captured interference image, and the output range is automatically set. In the case of a kerf C, the positions of the surface of the wafer W, the inner wall surface of the kerf C, and the bottom surface of the kerf C are recognized from the captured interference image, and the output range is set.

[0250] FIG. 25 is a conceptual diagram of setting the output range of an image.

[0251] Fig. 25 shows an example of measuring the shape of a kerf C. In Fig. 25, a thick line L indicates the contour shape of the cross section (YZ cross section) of the kerf C.

[0252] In the case of the kerf C, the surface of the wafer, the inner wall surface of the kerf C, and the bottom surface of the kerf C are the extraction targets. Therefore, a certain range including the surface of the wafer, the inner wall surface of the kerf C, and the bottom surface of the kerf C is set as the output range of the image. In this embodiment, a range of a certain width centered on the position of each surface is set as the output range of the image.

[0253] In FIG. 25, the range indicated by rectangular frames F1 and F5 is the image output range set for the surface of the wafer. The range indicated by rectangular frames F2 and F4 is the image output range set for the inner wall surface of the kerf C. The range indicated by rectangular frame F3 is the image output range set for the bottom surface of the kerf C. Each of the frames F1 to F5 has a configuration in which adjacent frames are connected at their ends. For example, frame F2 connects one end of frame F1 to one end of frame F3, and frame F4 connects one end of frame F5 to one end of frame F3.

[0254] In this embodiment, frames F1 to F5 are set on each surface recognized from the image (the surface of the wafer, the inner wall surface of the kerf C, and the bottom surface of the kerf C) to dynamically set the output range of the image.

[0255] FIG. 26 is a conceptual diagram of frame setting.

[0256] 26A shows the setting state of the frames F1 to F5 at the start of imaging. At the start of imaging, the frames F1 to F5 are aligned horizontally in a row.

[0257] Each of the frames F1 to F5 moves in conjunction with the movement of the imaging position.

[0258] Fig. 26(B) shows the setting state of frames F1 to F5 when the surface of the wafer is recognized. When the position of the wafer surface is recognized from the image, the movement of frames F1 and F5 is stopped. Also, the edge positions are adjusted. This determines the output range of the image to be set for the wafer surface.

[0259] FIG. 26(C) shows the setting state of the frames F1 to F5 while the inner wall surface of the kerf C is being imaged. The frames F2 and F4 are configured as expandable frames (so-called rubber bands). As described above, when the position of the surface of the wafer is recognized, the movement of the frames F1 and F5 is stopped. On the other hand, the frames F2 to F4 continue to move in conjunction with the movement of the imaging position. At this time, the frames F2 and F4 are extended between the frames F1 and F5 and the frame F3. At this time, the inclination of the frames F2 and F4 is adjusted and extended based on the position of the inner wall surface of the kerf C recognized from the image. This determines the output range of the image to be set for the inner wall surface of the kerf C.

[0260] 26(D) shows the setting state of the frames F1 to F5 when the bottom surface of the kerf C is recognized. When the position of the bottom surface of the kerf C is recognized from the image, the movement of the frame F3 is stopped. This determines the output range of the image to be set for the bottom surface of the kerf C.

[0261] In this manner, in this embodiment, the respective surfaces to be extracted (the surface of the wafer, the inner wall surface of the kerf C, and the bottom surface of the kerf C) are recognized from the image, frames F1 to F5 are set, and the output range of the image is dynamically set. In other words, the position of the interface is tracked, frames F1 to F5 are set, and the output range of the image is set.

[0262] FIG. 27 is a block diagram of the control system for image output.

[0263] The control of image output is performed by the second camera control unit 68E and the digital signal processing unit 68D in cooperation with each other. With regard to the control of image output, the second camera control unit 68E has the functions of an image output range setting unit 68E8 and an image output control unit 68E9. In addition, the digital signal processing unit 68D has the functions of a buffer processing unit 68D1, an image recognition unit 68D2, and an image output processing unit 68D3.

[0264] The buffer processing unit 68D1 buffers the image data to be processed. That is, it performs a process of temporarily storing the image data in a memory. The image data is stored in the memory in chronological order.

[0265] The image recognition unit 68D2 analyzes the captured image and recognizes the surface to be extracted. In this embodiment, since the kerf C is the measurement target, the surface of the wafer W, the inner wall surface of the kerf C, and the bottom surface of the kerf C are recognized. A known method can be adopted for image recognition. For example, for the surface of the wafer W, a method can be adopted in which the points with strong interference intensity inside the frames F1 and F5 are recognized as the surface when they can be recognized as the surface (using features such as the number of points, area, and Hough approximation lines). Similarly, for the bottom surface of the kerf C, a method can be adopted in which the points with strong interference intensity inside F3 are recognized as the bottom when they can be recognized as the bottom. In addition, for the inner wall surface of the kerf C, the angle of the region is analyzed from the arrangement of candidate points (for example, the inclination of the Hough line, the inclination of the least squares line, and the inclination of the major axis of the point sequence) to obtain the inclination angle.

[0266] The image output range setting unit 68E8 sets the image output range based on the recognition result of the surface to be extracted by the image recognition unit 68D2. In this embodiment, frames F1 to F5 are set, and the image output range on each surface (the surface of the wafer, the inner wall surface of the kerf C, and the bottom surface of the kerf C) is set.

[0267] The image output control unit 68E9 causes the image output processing unit 68D3 to perform image output processing based on the output range set by the image output range setting unit 68E8.

[0268] The image output processing unit 68D3 processes the image data stored in the memory based on the instruction of the image output control unit 68E9, and generates image data for output. That is, it extracts the image of the set output range and generates image data for output. The image data for output is generated by setting the range other than the range set as the output range to 0 or NULL.

[0269] In this manner, in this embodiment, the captured image is recognized and the output range of the image is set, whereby the output range can be set appropriately.

[0270] In the above example, the frames F2 and F4 that set the output range of the inner wall surface of the kerf C are configured as expandable frames, but they can also be configured as so-called multi-joint frames.

[0271] FIG. 28 is a diagram showing an example of a case where an output range of an image is set using a multi-joint frame.

[0272] 28 shows an example of measuring the shape of a step-cut kerf. In step cutting, after half-cutting, the inside of the half-cut is fully cut. Therefore, the cross section of the kerf C has a step.

[0273] The multi-jointed frames F2A, F4A are constructed by continuously generating frames F2a, F4a with a fixed length. Each frame F2a, F4a is connected so that adjacent frames can be bent. This allows the frames F2A, F4A to be set in a way that better matches the cross-sectional shape. This is particularly effective when the inner wall surface is not linear (such as when it has a step or is curved), as in the case of the groove C in this example.

[0274] FIG. 29 is a diagram showing another example of the frame for setting the output range of the image.

[0275] The example shown in Fig. 29 shows a case where the frame is divided and set on a pole point basis. The frames F2B and F4B, which set the output range of the inner wall surface of the kerf C, are configured to be expandable, but the extension is stopped at each pole point and a new frame is generated. In the example shown in Fig. 29, the extension of the frames F2b and F4b is stopped at the end points of the step portion, which is the pole point, and the end points of the curved surface, and new frames F2b and F4b are generated. This makes it possible to set frames F2B and F4B that match the shape of the cross section.

[0276] In the above example, the width of the frame is constant, but it may be set according to the inclination of the surface to be measured, etc.

[0277] In addition, when workpiece information can be acquired, it is preferable to set each frame F1 to F5 using the workpiece information. For example, it is preferable to set the width and length of each of the initial frames F1 to F5 using the workpiece information. It is also preferable to recognize each surface using the workpiece information.

[0278] Furthermore, it is preferable that the image used for image recognition is an image captured by performing the above-mentioned imaging control, which makes image recognition easier.

[0279] In the above example, the second imaging unit 60 is configured to perform the process of adjusting the output range of the image, but a similar process may be performed in a unit other than the second imaging unit 60.

[0280] In addition, while the above is based on the image output range, it is also possible to output the least squares line, Hough line, and principal axis as the interface = cross-sectional shape region.

[0281] [Other embodiments] The above-described embodiments can be implemented in appropriate combination. That is, adjustment of the amount of illumination light, gain adjustment, gradation correction, exposure time adjustment, and adjustment of the output range of pixel values ​​can be implemented in appropriate combination. Adjustment of the output range of an image can also be implemented in appropriate combination.

[0282] Furthermore, the imaging conditions set for each part when scanning and imaging in the Z-axis direction are not limited to those in the above embodiment, and various conditions can be set.

[0283] In addition, in the above embodiment, the process of generating three-dimensional shape data is performed by an image processing unit 110 separate from the second imaging unit 60, but the second imaging unit 60 may also have the functions of the image processing unit 110.

[0284] In the above embodiment, the second imaging unit 60 is configured as a white light interferometer, but the present invention can be applied to cases where the second imaging unit 60 is configured as another shape measuring device using an optical interference method. As for the interference optical system, in addition to the Mirau type, a Michelson type, a Fizeau type, or other interference optical system can also be used.

[0285] Furthermore, the present invention can also be applied to cases where the second imaging unit 60 is configured as a laser microscope (an optical microscope using a laser as an illumination light source and a confocal optical system) and cases where imaging is performed using a focus variation method.

[0286] In the above embodiment, the present invention is applied to a so-called twin spindle dicer, but the application of the present invention is not limited to this. The number of spindles mounted on the device may be one, or may be three or more. In the above embodiment, the present invention is applied to a so-called twin spindle dicer, but the application of the present invention is not limited to this. The number of spindles mounted on the device may be one, or may be three or more. In the above embodiment, the present invention is applied to a so-called twin spindle dicer, but the application of the present invention is not limited to this.

[0287] In the above embodiment, the drive unit of the second imaging unit 60 is configured to be shared with the drive unit of the second processing unit 40B, but each may be provided with an independent drive unit. Alternatively, a separate measurement unit may be provided in the device, and measurements may be performed by the measurement unit.

[0288] In the above embodiment, the second imaging unit 60 is moved when scanning and imaging in the Z-axis direction, but the table 10 may be moved instead. Alternatively, both may be moved. That is, it is sufficient that the second imaging unit 60 can move relatively along the Z-axis with respect to the work on the table 10. [Explanation of symbols]

[0289] 1... dicing device, 2... base, 3... column, 10... table, 10A... holding surface, 11... table drive unit, 30X... X-axis feed mechanism, 30YA... first Y-axis feed mechanism, 30YB... second Y-axis feed mechanism, 30ZA... first Z-axis feed mechanism, 30ZB... second Z-axis feed mechanism, 31X... X-axis guide rail, 31YA... first Y-axis guide rail, 31YB... second Y-axis guide rail, 31ZA... first Z-axis guide rail, 31ZB... second Z-axis guide rail, 32X... X-axis table, 32YA... first Y-axis table, 32YB... second Y-axis table, 32ZA... first Z-axis table, 32ZB...second Z-axis table, 33X...X-axis actuator, 33YA...first Y-axis actuator, 33YB...second Y-axis actuator, 33ZA...first Z-axis actuator, 33ZB...second Z-axis actuator, 40A...first processing unit, 40B...second processing unit, 41...blade, 42A...first spindle, 42B...second spindle, 43A...first spindle motor, 43B...second spindle motor, 44A...bracket, 44B...bracket, 50...first imaging unit, 51...first microscope section, 52...illumination section, 52A...illumination light source, 52B...for illumination Lens, 53...beam splitter, 54...objective lens, 55...imaging lens, 56...first camera unit, 56A...image sensor, 56B...analog signal processing unit, 56D...digital signal processing unit, 56E...first camera control unit, 60...second imaging unit, 60B...second imaging unit, 61...second microscope unit, 62...illumination unit, 62A...illumination light source, 62B...illumination lens, 63...first beam splitter, 64...objective lens, 65...glass plate, 65A...reference mirror, 66...second beam splitter, 67...imaging lens, 68...second camera unit, 68A...image sensor, 68B...analog signal processing unit, 68D... digital signal processing unit, 68D1... buffer processing unit, 68D2... image recognition unit, 68D3... image output processing unit, 68E... second camera control unit, 68E1... work information acquisition unit, 68E2... imaging condition setting unit, 68E3... illumination control unit, 68E4... gain control unit, 68E5... exposure control unit, 68E6... gradation control unit, 68E7... pixel output control unit, 68E8... image output range setting unit, 68E9... image output control unit, 100... system controller, 110... image processing unit, 110A... alignment unit, 110B... kerf check unit, 110C... kerf shape measurement unit,120...operation unit, 130...display unit, 140...communication unit, C...kerf, DF...dicing frame, DT...dicing tape, F1...frame, F2...frame, F2A...frame, F2B...frame, F2a...frame, F2b...frame, F3...frame, F4...frame, F5...frame, Fa1...surface of wafer, Fa2...inner wall surface of kerf, Fa3...bottom surface of kerf, L...contour shape of cross section of kerf, W...wafer, Z0...imaging start position, Z1...position of wafer surface, Z2...position of bottom surface of kerf, ZE...imaging end position,

Claims

1. a table for holding a workpiece on a holding surface perpendicular to the Z-axis; a processing unit that processes the workpiece on the table; An imaging unit that images the surface of the workpiece using an optical interference method; a drive unit that moves the imaging unit relative to the table along the Z-axis direction; an imaging control unit that controls the drive unit and the imaging unit to scan and image a surface of the workpiece on the table in the Z-axis direction; Equipped with The imaging control unit, when scanning and imaging the surface of the workpiece in the Z-axis direction, images the surface according to imaging conditions determined for each portion to be imaged. Processing equipment.

2. The imaging control unit irradiates illumination light with a light amount determined for each portion and images the surface of the workpiece. The processing device according to claim 1 .

3. The imaging control unit adjusts gain with a setting determined for each portion and images the surface of the workpiece. The processing device according to claim 1 .

4. The imaging control unit controls an exposure time according to a setting determined for each part, and images the surface of the workpiece. The processing device according to claim 1 .

5. The imaging control unit performs gradation correction with settings determined for each of the portions and images the surface of the workpiece. The processing device according to claim 1 .

6. The imaging control unit outputs a pixel value from each pixel within an output range determined for each portion, and images the surface of the workpiece. The processing device according to claim 1 .

7. When imaging a groove machined on the surface of the workpiece, the workpiece is divided into a first portion including the surface of the workpiece, a third portion including a bottom surface of the groove, and a second portion between the first portion and the third portion, and the imaging conditions are determined for each divided portion. The processing device according to claim 1 .

8. An image processing unit further includes: an image processing unit that processes an image captured by scanning the surface of the workpiece in the Z-axis direction to measure a shape of the surface of the workpiece; The processing device according to claim 1 .

9. A work information acquisition unit that acquires work information including information on a cross-sectional shape of the work; an imaging condition setting unit that sets the imaging conditions based on the work information; Further comprising: The processing device according to any one of claims 1 to 8.

10. An output range setting unit that sets an output range of the image based on information on the cross-sectional shape of the workpiece, The imaging control unit extracts an image of the range set by the output range setting unit from the captured image and outputs the image. The processing apparatus according to claim 9.

11. The image capturing apparatus further includes an output range setting unit that analyzes the captured image and sets an output range, The imaging control unit extracts an image of the range set by the output range setting unit from the captured image and outputs the image. The processing device according to any one of claims 1 to 8.

12. The output range setting unit analyzes the captured image to recognize the interface of the workpiece, and sets a predetermined range including the recognized interface to the output range. The processing device according to claim 11.

13. When imaging a groove machined on the surface of the workpiece, the output range setting unit recognizes the surface of the workpiece, the inner wall surface of the groove, and the bottom surface of the groove, and sets a predetermined range including each recognized surface as the output range. The processing device according to claim 12.

14. The imaging unit images the surface of the workpiece using a white light interference method. The processing device according to any one of claims 1 to 8.

Images