JP7462497B2 - Processing Equipment - Google Patents

Description

The present invention relates to a processing device.

There is a processing device that processes a workpiece such as a semiconductor wafer along a planned dividing line in order to divide the workpiece into chip sizes. For example, Patent Document 1 discloses a technology that captures an image of the workpiece and detects the area to be processed.

JP 2005-166991 A

For example, when processing along a planned division line, auto alignment uses an imaging unit to capture an image of the surface of a workpiece placed on a holding table, and automatically detects the position of the planned division line based on the pre-set key pattern of the device to be formed on the surface of the workpiece and the pre-measured distance from the key pattern to the planned division line. However, with conventional technology, there was a problem that if a different pattern was recognized as the key pattern, or if the distance from the set key pattern to the planned division line was incorrect, the processing would be performed in the wrong position.

The present invention has been made in consideration of the above, and its purpose is to provide a processing device that can prevent processing of positions other than the intended dividing line of the workpiece.

In order to solve the above-mentioned problems and achieve the object, the processing apparatus of the present invention is a processing apparatus for processing a workpiece formed on a surface in which a device area is partitioned by a plurality of planned division lines, and the processing apparatus includes a holding table for holding the workpiece, a processing feed unit for processing and feeding the holding table in the X direction, a processing unit for forming a processing groove in the workpiece held on the holding table, an indexing feed unit for indexing and feeding the processing unit in the Y direction intersecting the X direction, an imaging unit for imaging the workpiece held on the holding table, and a control unit, and the control unit is configured to image the surface of the workpiece. The trained program has learned teacher data having an OK image in which the planned division line is located at the center of the captured image and an NG image in which the planned division line is not located at the center of the image, and includes a machine learning unit that processes an input image and determines whether the input image is an OK image or an NG image, an input unit that inputs an image of the planned division line of the workpiece to be processed placed on the holding table to the machine learning unit, and a control unit that controls the processing unit to process the workpiece to be processed when the machine learning unit determines that the image is an OK image, and executes a warning process when the machine learning unit determines that the image is an NG image.

The processing device of the present invention has the effect of preventing processing of positions that are not the planned division lines by using a trained program to judge images captured of the planned division lines.

FIG. 1 is a perspective view that illustrates a configuration example of a processing device according to an embodiment. FIG. 2 is a plan view showing a configuration example of a wafer to be processed by the processing apparatus according to the embodiment. FIG. 3 is a block diagram illustrating an example of a functional configuration of the processing device according to the embodiment. FIG. 4 is a diagram for explaining an example of a trained program used by the processing device according to the embodiment. FIG. 5 is a flowchart illustrating an example of a processing procedure related to registration of image data according to the embodiment. FIG. 6 is a flowchart illustrating an example of a processing procedure related to the determination of the machine learning unit according to the embodiment. FIG. 7 is a flowchart illustrating an example of a processing procedure related to the teaching unit according to the embodiment. FIG. 8 is a diagram showing an example of a relationship between a key pattern and a street according to the embodiment. FIG. 9 is a flowchart illustrating an example of a processing procedure during auto-alignment of the processing apparatus according to the embodiment. FIG. 10 is a perspective view that illustrates a configuration example of a processing device according to a modified example of the embodiment. FIG. 11 is a block diagram showing an example of the configuration of the laser irradiation unit shown in FIG.

The following describes in detail an embodiment of the present invention with reference to the drawings. The present invention is not limited to the contents described in the following embodiment. The components described below include those that a person skilled in the art can easily imagine and those that are substantially the same. Furthermore, the configurations described below can be combined as appropriate. Various omissions, substitutions, or modifications of the configuration can be made without departing from the spirit of the present invention.

In the embodiment described below, an XYZ Cartesian coordinate system is set, and the positional relationship of each part is described with reference to this XYZ Cartesian coordinate system. One direction within a horizontal plane is the X-axis direction, the direction perpendicular to the X-axis direction within the horizontal plane is the Y-axis direction, and the direction perpendicular to both the X-axis direction and the Y-axis direction is the Z-axis direction. The XY plane including the X-axis and Y-axis is parallel to the horizontal plane. The Z-axis direction perpendicular to the XY plane is the vertical direction.

[Embodiment]
The processing device according to the embodiment will be described with reference to the drawings. Fig. 1 is a perspective view showing a typical configuration example of the processing device according to the embodiment. Fig. 2 is a plan view showing a typical configuration example of a wafer processed by the processing device according to the embodiment. Fig. 3 is a block diagram showing a typical example of a functional configuration of the processing device according to the embodiment.

As shown in FIG. 1, the processing device 10 (an example of a processing device) according to the embodiment is a cutting device that cuts a wafer 100 along a street (planned dividing line) 103. The processing device 10 basically includes a chuck table 11 that holds the wafer 100, an imaging unit 12, a processing unit 13, a driving means 14, a Z-axis moving means 15, a touch panel 17, an LED light 18, and a control unit 40.

As shown in FIG. 2, a wafer (an example of a workpiece) 100 is attached to the surface of an adhesive tape 107 attached to an annular frame 108. The wafer 100 is partitioned into a plurality of regions by a plurality of streets 103 formed in a lattice pattern on the surface 101, and devices 104 made up of semiconductor chips such as ICs and LSIs are formed in these partitioned regions. The wafer 100 thus configured has its back surface 102 attached to the adhesive tape 107 attached to the annular frame 108 with the surface 101 facing up.

Returning to FIG. 1, the chuck table 11 (an example of a holding table) has a holding surface 11a that suction-holds the wafer 100, and is rotatably connected to a motor 19. The chuck table 11 is also movable in the X-axis direction, which is horizontal to the holding surface 11a, by an X-axis moving means 22 of known configuration, such as a ball screw 20, a nut, and a pulse motor 21. This allows the processing device 10 to move the wafer 100 held on the chuck table 11 in the X-axis direction relative to the imaging unit 12 and the processing unit 13.

The imaging unit 12 is an electron microscope equipped with an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor). The imaging unit 12 images the surface 101 of the wafer 100 held on the holding surface 11a of the chuck table 11. The imaging unit 12 can image the surface 101 of the wafer 100 by switching between low magnification (Lo) macro imaging and high magnification (Hi) micro imaging. The area (streets) to be cut are detected based on the image information (key pattern) acquired by the imaging unit 12, and are used to position the processing operation by the processing unit 13. The imaging unit 12 may include, for example, a light source that irradiates illumination light onto the surface of the wafer 100 held on the holding surface 11a of the chuck table 11. This light source can be composed of an epi-illumination light source that illuminates the wafer 100 from directly above and an oblique light source that illuminates the wafer 100 from an oblique direction.

The processing unit 13 cuts the wafer 100 held on the holding surface 11a of the chuck table 11 along the street 103 with a rotating ring-shaped ultra-thin cutting blade 24 to form a cutting groove (kerf). The imaging unit 12 is integrated by being attached to and supported by a part of the housing 25 for the processing unit 13, and is provided so as to be movable in the Z-axis direction by Z-axis moving means 15 including a ball screw, a nut, a pulse motor 26, etc. In addition, the Y-axis moving means 27 for moving the imaging unit 12 and the processing unit 13 in the Y-axis direction relative to the holding surface 11a of the chuck table 11 is composed of a ball screw 28, a nut, a pulse motor 29, etc., and constitutes the driving means 14 together with the X-axis moving means 22.

The machining unit 13 has a spindle (not shown) housed in a housing 25 and rotatably supported by an air bearing. The spindle is driven to rotate by a motor (not shown) housed in the housing 25. The cutting blade 24 is removably attached to the tip of the spindle and rotates when the spindle is driven to rotate.

Under the control of the control unit 40, the touch panel 17 displays the image of the surface 101 of the wafer 100 captured by the imaging unit 12 and various information required for processing, and also receives input operations and the like required for processing from the operator. The touch panel 17 is disposed in an easy-to-see and easy-to-operate location on the housing of the processing device 10. The touch panel 17 is configured with a display device and an input device. The touch panel 17 may be configured with a touch screen display equipped with a display device such as a liquid crystal display or an organic EL display, and a touch screen for specifying an input position or coordinates on the display surface of the display device.

The LED light 18 notifies the operating status of the processing device 10 with multiple display colors. The LED light 18 is configured to be able to light up or blink under the control of the control unit 40.

The processing device 10 configured as described above cuts the wafer 100 to a predetermined cutting depth with the cutting blade 24 rotated at high speed, while moving the chuck table 11 relatively in the X-axis direction with the X-axis moving means 22 with respect to the processing unit 13. This allows the streets 103 on the wafer 100 to be cut to form cut grooves (kerfs). Next, the processing device 10 rotates the wafer 100 by 90° by rotating the chuck table 11, and then repeats the same cutting process with the processing unit 13 for all the new streets 103 arranged in the X-axis direction. This allows the wafer 100 to be divided into individual devices 104.

The storage 30 shown in FIG. 3 stores a control program 201 that realizes functions such as various processes executed by the control unit 40, and data used in the processes by such control programs. In this embodiment, the storage 30 further stores a trained program 202 that has undergone machine learning. The storage 30 can be implemented using a hard disk drive (HDD), a semiconductor memory, or the like. The storage 30 may also be used as a temporary working area when the processor included in the control unit 40 executes instructions written in the control program 201.

The storage 30 includes an image storage unit 31. The image storage unit 31 stores image data 300 in which an image of the surface 101 of the wafer 100 is labeled as an OK image and an NG image. The image data 300 is used as teacher data for the learned program 202. The learned program 202 is a program that has learned a plurality of image data 300 as teacher data, including an OK image in which the street 103 is located at the center of an image of the surface 101 of the wafer 100, and an NG image in which the street 103 is not located at the center of the image. The learned program 202 includes a program that has been machine-learned for at least one of the types of wafer 100 and the types of device 104. The image data 300 includes an image of the surface 101 of the wafer 100 and information that labels the image as an OK image and an NG image. The image data 300 includes, for example, information indicating an image and a label. The image data 300 includes, for example, an image captured by the processing device 10 and an image acquired from outside the processing device 10.

In this embodiment, the processing device 10 stores the learned program 202 in the storage 30, but is not limited to this. The processing device 10 may be configured to use the learned program 202 stored in, for example, an external server, storage device, another processing device 10, etc. that can be accessed by the processing device 10. The processing device 10 may, for example, cause the learned program 202 to be executed by a server, etc.

Figure 4 is a diagram for explaining an example of a learned program 202 used by the processing device 10 according to the embodiment. In the example shown in Figure 4, detailed description of the device 104 of the wafer 100 is omitted in the image data 300.

As shown in FIG. 4, the learned program 202 is a program (machine learning data) that has been learned using image data 300 having OK images and image data 300 having NG images as teacher data for each type of wafer 100. The learned program 202 has identification data that can identify the type of wafer 100 and a program that has been learned corresponding to the identification data. In other words, the learned program 202 is trained using the image data 300 of OK images and NG images linked to the identification data as teacher data for machine learning. Machine learning may be performed using only teacher data of wafers 100 of the same type as the wafer 100 to be processed, and the created trained program 202 may be used during processing. The processing device 10 can make a highly accurate judgment by judging using the trained program 202 for the same wafer 100.

In addition, the processing apparatus 10 may perform machine learning using training data from multiple types of wafers 100, regardless of the type of device 104, and use the created learned program 202 during processing. In the latter case, linking with identification data is not required. In this case, the processing apparatus 10 can use one learned program 202 for various types of wafers 100 by creating one program, which is highly versatile and efficient because there is no need to create a learned program 202 for each type of wafer 100.

In the example shown in FIG. 4, when the identification data is "0001" to "0004", the trained program 202 performs machine learning on a plurality of image data 300 capturing an image of the surface 101 of the wafer 100. In detail, the trained program 202 performs machine learning on a model using a plurality of image data 300 showing the characteristics of a good image and a plurality of image data 300 showing the characteristics of a bad image as a data set, thereby realizing the determination of whether an input image is a good image or a bad image.

In this embodiment, the trained program 202 is a program that trains a model using a plurality of image data 300 as a data set, and is capable of determining whether an input image to be judged is an OK image or an NG image based on the trained model.

Returning to FIG. 3, the control unit 40 includes an arithmetic processing device such as a CPU (Central Processing Unit), a storage device such as a ROM (Read Only Memory) or a RAM (Random Access Memory), and an input/output interface device. The control unit 40 is a computer that can use these devices to execute control programs and the like for controlling each of the above-mentioned components in accordance with a series of processing steps performed by the processing device 10.

The control unit 40 controls the overall operation of the processing device 10 according to the processing conditions set by the operator via the touch panel 17. As shown in FIG. 3, the control unit 40 includes a machine learning unit 41, an input unit 42, a control unit 43, a labeling unit 44, a teach unit 45, and an alignment unit 46. The control unit 40 executes a control program 201 to realize the functions, actions, etc. of each unit.

The machine learning unit 41 processes an input image with the trained program 202 and provides a function of determining whether the input image is a OK image or a NG image. The machine learning unit 41 executes the trained program 202 that has learned teacher data having OK images and NG images, thereby determining the input image with the trained program 202, and provides information indicating the result of the determination to the control unit 43.

The machine learning unit 41 provides a function to generate and update a learned program 202 that has executed machine learning using the image data 300 in the image storage unit 31. The machine learning unit 41 provides a function to execute machine learning for each identification data using the image data 300 in the image storage unit 31 that corresponds to the identification data of the wafer 100.

The input unit 42 inputs an image of the street 103 of the wafer 100 to be processed placed on the chuck table 11 to the machine learning unit 41. The input unit 42 acquires an image of the wafer 100 to be processed captured by the imaging unit 12, and inputs the image to the machine learning unit 41.

The control unit 43 provides a function to control the processing unit 13 to process the wafer 100 to be processed when the machine learning unit 41 judges that the image is OK. That is, when the control unit 43 receives information from the machine learning unit 41 indicating that the learned program 202 judges that the image is OK, the control unit 43 executes a process to control the processing. The control unit 43 controls the processing unit 13 to process, for example, the street 103 of the wafer 100 to be processed. The control unit 43 also provides a function to execute a warning process when the machine learning unit 41 judges that the image is NG. The warning process includes, for example, a process of notifying that the position of the street 103 is incorrect and a process of stopping or suspending the processing of the processing unit 13, thereby preventing the wafer 100 from being erroneously processed. The notification process includes, for example, a means of displaying information indicating a warning on the touch panel 17, issuing information indicating a warning by an error sound or voice, and lighting or blinking the LED light 18 exposed from the exterior cover covering the processing device 10 in a color corresponding to an error.

The labeling unit 44 labels the image of the surface 101 of the wafer 100 into an OK image and an NG image, and stores the image data 300 in the image storage unit 31. The labeling unit 44 acquires, for example, an image captured by the processing device 10, an image captured by another processing device 10, etc., and associates a label indicating an OK image or an NG image with the image. The labeling unit 44 may be configured to display, for example, an image and a screen on the touch panel 17 that allows the user to select whether the image is an OK image or an NG image, and label the image based on the operation result on the screen. The labeling unit 44 may be configured to automatically associate a label based on the judgment conditions of the street 103 in the image, such as a line that can be recognized as a straight line in the middle area when the screen is vertically divided into thirds.

The teach unit 45 provides a function for registering a key pattern formed on the surface 101 of the wafer 100 and the distance from the key pattern to the street 103. The teach unit 45 displays on the touch panel 17 a screen for registering the distance from a key pattern, which is part of a device 104 formed on the surface 101 of the wafer 100, to the street 103, and corresponds the key pattern registered via the touch panel 17 and the distance to the street 103 to the type of wafer 100, etc.

The alignment unit 46 provides a function of capturing an image of the wafer 100 to be processed placed on the chuck table 11, detecting the key pattern by pattern matching, detecting the streets 103 on the wafer 100, and positioning the detected streets 103 at the center of the imaging unit 12 and capturing an image. The alignment unit 46 provides the image captured by the imaging unit 12 to the input unit 42.

The above describes an example of the configuration of the processing device 10 according to this embodiment. Note that the above configuration described using Figures 1 to 4 is merely an example, and the configuration of the processing device 10 according to this embodiment is not limited to this example. The functional configuration of the processing device 10 according to this embodiment can be flexibly modified according to the specifications and operation.

(Example of image data registration)
Next, an example of registration of image data 300 executed by the processing device 10 according to the embodiment will be described. Fig. 5 is a flowchart showing an example of a processing procedure related to registration of image data 300 according to the embodiment. The processing procedure shown in Fig. 5 is realized by the control unit 40 executing the control program 201. The processing procedure shown in Fig. 5 is executed, for example, at the timing of registering an image as teacher data, at the timing of occurrence of an image to be registered in the learned program 202, etc.

As shown in FIG. 5, the control unit 40 acquires an image of the surface 101 of the wafer 100 (step 1001). The control unit 40 acquires information on an image of, for example, the device 104 on the surface 101 of the wafer 100. For example, when precision is required, the control unit 40 acquires images for each type of device 104. For example, when versatility is required, the control unit 40 acquires images of not only one type of device 104 but an unspecified number of devices 104. For example, the control unit 40 may be configured to acquire images not only from the imaging unit 12 but also from a server, storage device, etc. external to the processing device 10. When the process of step 1001 is completed, the control unit 40 advances the process to step 1002.

The control unit 40 stores image data 300 in which the acquired images have been labeled as OK images and NG images in the image storage section 31 (step 1002). The control unit 40 can be configured to label images as OK images and NG images based on the layout and layout conditions of the street 103 of the image, for example. The control unit 40 can be configured to display a screen on the touch panel 17 that allows the user to determine whether the image is OK or NG, and to label the images based on the result of the determination. When the process of step 1002 ends, the control unit 40 advances the process to step 1003.

The control unit 40 performs machine learning using the labeled image data 300 to generate a trained program 202 capable of distinguishing between OK and NG images (step 1003). The control unit 40 generates the trained program 202 that has learned the rules (model) for judging unknown images based on the characteristics and patterns of the OK and NG images in the image data 300, which is the teacher data, for each wafer 100 or device 104, for example. In other words, generating the trained program 202 means, for example, that the rules for judging unknown images have been learned. When the process of step 1003 is completed, the control unit 40 advances the process to step 1004.

The control unit 40 stores the generated learned program 202 in the storage 30 (step 1004). For example, if the learned program 202 is already stored in the storage 30, the control unit 40 overwrites the generated learned program 202 or stores it as a new program to be added. When the control unit 40 ends the processing of step 1004, it ends the processing procedure shown in FIG. 5.

By executing the processing procedure shown in FIG. 5, the control unit 40 functions as a machine learning unit 41 and a labeling unit 44.

(Machine learning decision)
Next, an example of the determination by the machine learning unit 41 executed by the processing device 10 according to the embodiment will be described. Fig. 6 is a flowchart showing an example of a processing procedure related to the determination by the machine learning unit 41 according to the embodiment. The processing procedure shown in Fig. 6 is realized by the control unit 40 executing the control program 201. The processing procedure shown in Fig. 6 is executed, for example, when an image is input from the input unit 42.

As shown in FIG. 6, the control unit 40 processes the input image with the learned program 202 and determines whether the input image is an OK image or an NG image (step 1101). For example, the control unit 40 executes the learned program 202 to cause the learned program 202 to determine whether the input image is an OK image or an NG image. When the control unit 40 obtains the determination result from the learned program 202, it proceeds to step 1102.

The control unit 40 outputs the judgment result of the input image to the control unit 43 (step 1102). When the control unit 40 outputs the judgment result of the input image obtained by executing the learned program 202 to the control unit 43, it ends the processing procedure shown in FIG. 6.

In this embodiment, the control unit 40 functions as a machine learning unit 41 by executing the processing procedure shown in FIG. 6.

(Teach section processing)
Next, an example of processing of the teach unit 45 executed by the processing device 10 according to the embodiment will be described. Fig. 7 is a flowchart showing an example of a processing procedure related to the teach unit 45 according to the embodiment. The processing procedure shown in Fig. 7 is realized by the control unit 40 executing the control program 201. The processing procedure shown in Fig. 7 is executed, for example, when a teach operation is performed by the processing device 10.

As shown in FIG. 7, the control unit 40 acquires an image of the surface 101 of the wafer 100 (step 2001). The control unit 40 registers the key pattern formed on the surface 101 of the wafer 100 and the distance from the key pattern to the street 103 (step 2002). In auto-alignment, which images the wafer 100 before processing and detects the position of the planned division line, the position of the street 103 is automatically determined using the key pattern as a clue.

Figure 8 is a diagram showing an example of the relationship between a key pattern and a street according to an embodiment. As shown in Figure 8, the control unit 40 recognizes the positional relationship between a target key pattern 120 and a street 103 from an image, detects a distance 130 between the key pattern 120 and the street 103, and links and registers the key pattern 120 and the distance 130. The control unit 40 may detect and register the distances 130 between multiple key patterns 120 and the streets 103.

Returning to FIG. 7, the control unit 40 controls the driving means 14 so that the street 103 is located at the center of the imaging unit 12 (step 2003). When the street 103 is located at the center of the imaging unit 12, the control unit 40 advances the process to step 2004. Note that the center when positioning the location assumed to be the planned division line at the center of the imaging unit 12 is the center in the Y direction, which is the indexing feed direction that intersects with the X direction, which is the processing feed direction in which the planned division line extends.

The control unit 40 displays a screen for registering the street 103 on the touch panel 17 (step 2004). The screen has, for example, an image for confirming the position of the street 103 and a registration button. The control unit 40 then determines whether or not to register based on information input via the touch panel 17 (step 2005). For example, the control unit 40 determines to register when the registration button is selected. When the control unit 40 determines not to register (No in step 2005), it ends the processing procedure shown in FIG. 7 without registering the street 103.

If the control unit 40 determines to register (Yes in step 2005), the process proceeds to step 2006. The control unit 40 registers the street 103 of the wafer 100 (step 2006). For example, the control unit 40 registers a position positioned at the center of the imaging unit 412 as the street 103. When step 2006 ends, the control unit 40 ends the process procedure shown in FIG. 7.

In this embodiment, the control unit 40 functions as the teach section 45 by executing the processing procedure shown in FIG. 7.

(Auto alignment process overview)
Next, an example of a process for preventing erroneous processing during auto-alignment executed by the processing apparatus 10 according to the embodiment will be described. Fig. 9 is a flowchart showing an example of a process procedure during auto-alignment of the processing apparatus 10 according to the embodiment. The process procedure shown in Fig. 9 is realized by the control unit 40 executing the control program 201. The process procedure shown in Fig. 9 is executed in a state where the chuck table 11 on which the wafer 100 is placed is disposed at the imaging position of the imaging unit 12, etc.

As shown in FIG. 9, the control unit 40 images the wafer 100 to be processed with the imaging unit 12 (step 3001). The control unit 40, for example, causes the imaging unit 12 to image the surface 101 of the wafer 100 and acquires the captured image. When the process of step 3001 ends, the control unit 40 advances the process to step 3002.

The control unit 40 detects the key pattern 120 from the image by pattern matching (step 3002). The control unit 40 detects the key pattern 120 that matches or is similar to a pre-registered key pattern, for example, by analyzing the acquired image. When the control unit 40 finishes the process of step 3002, it advances the process to step 3003.

The control unit 40 detects the street 103 of the wafer 100 based on the key pattern 120 (step 3003). For example, based on the key pattern 120 and the distance 130 associated with the key pattern 120, the control unit 40 detects a position that is a distance 130 away from the key pattern 120 as the street 103. When the process of step 3003 ends, the control unit 40 advances the process to step 3004.

The control unit 40 positions the detected street 103 at the center of the imaging unit 12 and images it (step 3004). The control unit 40, for example, drives the driving means 14 so that the detected street 103 is located at the center of the imaging unit 12, and controls the imaging unit 12 to image the street 103 portion of the wafer 100. When the processing of step 3004 is completed, the control unit 40 advances the processing to step 3005.

The control unit 40 inputs the captured image to the machine learning unit 41 (step 3005). For example, the control unit 40 inputs the captured image to the machine learning unit 41, which executes the learned program 202 and determines whether the image is an OK image or an NG image. When the control unit 40 acquires the determination result output by the machine learning unit 41, it proceeds to step 3006.

The control unit 40 determines whether the judgment result is an OK image (step 3006). For example, when the judgment result output by the machine learning unit 41 indicates an OK image, the control unit 40 determines that the judgment result is OK. When the control unit 40 determines that the judgment result is an OK image (Yes in step 3006), the control unit 40 advances the process to step 3007.

The control unit 40 controls the processing of the wafer 100 (step 3007). The control unit 40 forms a cutting groove in the wafer 100 while controlling the cutting depth, for example by controlling the cutting blade 24 to move up and down according to the top surface height of the street 103. More specifically, the control unit 40 moves the cutting blade 24 along the Y-axis direction by the driving means 14, and positions the cutting blade 24 along the Z-axis direction by the Z-axis moving means 15 at a position according to the top surface height, thereby forming a cutting groove in the street 103. When the process of step 3007 is completed, the control unit 40 ends the process procedure shown in FIG. 9.

Furthermore, if the control unit 40 determines that the image is not an OK image (No in step 3006), the control unit 40 advances the process to step 3008. The control unit 40 executes the above-mentioned warning process (step 3008). For example, by executing the warning process, the control unit 40 warns that the position of the street 103 is incorrect, and stops or suspends processing by the processing unit 13. When the process of step 3008 ends, the control unit 40 ends the processing procedure shown in FIG. 9.

In the processing procedure shown in FIG. 9, the control unit 40 functions as the alignment section 46 by executing a series of processes from step 3001 to step 3004. The control unit 40 functions as the input section 42 by executing the process of step 3005. The control unit 40 functions as the control section 43 by executing the processes from step 3006 to step 3008.

As described above, the processing device 10 uses the learned program 202 to determine whether the image of the street 103 of the wafer 100 is an OK image or an NG image, and executes a warning process if it is determined to be an NG image. If the learned program 202 determines that the image is OK, the processing device 10 processes the wafer 100. As a result, the processing device 10 can improve the accuracy of image determination by using the determination result of the learned program 202 that has learned teacher data having OK images and NG images actually captured. As a result, the processing device 10 can prevent processing an incorrect position of the wafer 100 that is not the street (planned division line) 103 when the planned division line is not captured in the center of the image. Furthermore, the processing device 10 can make a suitable determination for each wafer 100 or device 104 by using the learned program 202 that has learned actual images, thereby improving convenience.

The processing device 10 can also have the learned program 202 learn image data 300, which is obtained by labeling an image of the surface 101 of the wafer 100 as an OK image or an NG image, as teacher data. This allows the processing device 10 to improve the accuracy of the learned program 202 in determining the image input to the learned program 202. As a result, the processing device 10 can more reliably prevent processing the wrong position on the wafer 100 that is not the planned dividing line.

The processing device 10 also captures an image of the wafer 100 to be processed placed on the chuck table 11, detects the key pattern 120 by pattern matching to detect the street 103 of the wafer 100, and positions the detected street 103 at the center of the imaging unit 12 and captures the image. This allows the processing device 10 to improve the accuracy of the image itself input to the learned program 202. As a result, the processing device 10 can more reliably prevent processing an incorrect position on the wafer 100 that is not a planned dividing line. For example, by applying this to auto-alignment, the processing device 10 can prevent processing an incorrect position on the wafer 100 even if the position of the street 103 is automatically determined using the key pattern as a clue.

[Modification]
A processing device 10 according to a modified example of the above embodiment will be described below. Fig. 10 is a perspective view showing a schematic configuration example of the processing device 10 according to the modified example of the embodiment. Fig. 11 is a block diagram showing a configuration example of the laser irradiation unit shown in Fig. 10. In the following embodiments, the same parts are designated by the same reference numerals, and duplicated descriptions will be omitted.

The processing device 10 shown in FIG. 10 is a laser processing device that irradiates a laser beam onto a wafer 100. The processing device 10 basically comprises a chuck table 11 that holds the wafer 100, an imaging unit 12, a processing unit 13, a driving means 14, a Z-axis moving means 15, a touch panel 17, and a control unit 40. The processing unit 13 has a laser irradiation unit 5 instead of the cutting blade 24 described above.

As shown in FIG. 11, the laser irradiation unit 5 includes an oscillator 51, an intensity adjustment unit (attenuator) 52, a polarization direction setting unit 53, a mirror element 54, and a condenser lens 55. The laser irradiation unit 5 is controlled by the control unit 40.

The oscillator 51 is for oscillating a laser beam of a predetermined wavelength, and is composed of, for example, a laser beam oscillator such as a YAG laser oscillator or a YVO4 laser oscillator, or an oscillator that oscillates a laser beam in the infrared wavelength range such as 1064 nm. The intensity adjustment unit 52 includes a half-wave plate 521, a motor 522, a polarizing beam splitter 523, and an absorber 524.

The half-wave plate 521 is disposed after the oscillator 51 in a state in which it can be rotated by the motor 522. The half-wave plate 521 changes the linear polarization direction of the laser beam LB oscillated by the oscillator 51 according to the rotation angle. The polarizing beam splitter 523 transmits the laser beam having a predetermined linear polarization direction among the laser beams that have passed through the half-wave plate 521 toward the polarization direction setting unit 53, and branches the laser beam having a linear polarization direction other than the predetermined linear polarization direction to the absorber 524 side. The absorber 524 is for absorbing the laser beams branched by the polarizing beam splitter 523, and may be made of, for example, a matte black metal to absorb the laser beam well.

The polarization direction setting unit 53 includes a half-wave plate 531 and a motor 532. The half-wave plate 531 is arranged after the polarizing beam splitter 523 in a state in which it can be rotated by the motor 532. The half-wave plate 531 changes the linear polarization direction of the laser beam transmitted through the polarizing beam splitter 523 according to the rotation angle. The polarization direction setting unit 53 functions as a polarization direction setting means according to the present invention. The mirror element 54 is for reflecting the laser beam that has passed through the half-wave plate 531 toward the condenser lens 55. The condenser lens 55 is arranged to face the wafer 100, and focuses the laser beam reflected by the mirror element 54 by positioning the focal point inside the wafer 100.

In the example shown in FIG. 11, the focusing lens 55 functions, for example, as a focusing device. The intensity adjustment unit 52 and the polarization direction setting unit 53 arranged between the oscillator 51 and the focusing lens 55 function, for example, as an optical system.

A configuration example of the processing device 10 according to a modified example of this embodiment has been described above. Note that the above configuration described using FIG. 8 is merely an example, and the configuration of the processing device 10 according to this embodiment is not limited to this example. The functional configuration of the processing device 10 according to this embodiment can be flexibly modified according to the specifications and operation.

The processing device 10 according to the modified embodiment processes the wafer 100 divided by the streets 103, which are the planned dividing lines, along the streets 103. The processing device 10, for example, irradiates the wafer 100 with a laser beam having a wavelength that is transparent to the wafer 100, and continuously forms a modified layer (degraded region) along the streets 103 inside the wafer 100. Furthermore, the processing device 10 may have a function of dividing the wafer 100 by applying an external force along the streets 103 whose strength has been reduced by the formation of the modified layer. In addition, the processing device 10 irradiates the wafer 100 with a laser beam having a wavelength that is absorbent to the wafer 100, and forms a cutting groove along the streets 103 on the surface of the wafer 100. The cutting groove may have a depth that half-cuts the wafer 100, or may have a depth that fully cuts the wafer 100.

The processing device 10 can use the processing procedure shown in FIG. 9 described above. For example, the processing device 10 can replace the processing of step 3007 with processing using the laser irradiation unit 5.

For example, when the control unit 40 of the processing device 10 determines in step 3005 that the image is an OK image (Yes in step 3005), it controls the processing of the wafer 100 (step 3007). For example, the control unit 40 positions one end of the street 103 vertically below the condenser lens 55, and while the laser beam LB is being emitted from the oscillator 51, causes the chuck table 11 to be processed and fed in the X-axis direction relative to the processing unit 13 by the X-axis moving means 22. When the processing of step 3007 is completed, the control unit 40 ends the processing procedure shown in FIG. 9.

Furthermore, if the control unit 40 determines that the image is not an OK image (No in step 3006), the control unit 40 advances the process to step 3008. The control unit 40 executes the above-mentioned warning process (step 3008). For example, by executing the warning process, the control unit 40 warns that the position of the street 103 is incorrect, and stops or suspends processing by the processing unit 13. When the process of step 3008 ends, the control unit 40 ends the processing procedure shown in FIG. 9.

The processing device 10 according to the modified embodiment uses the learned program 202 to determine whether an image of the street 103 of the wafer 100 is an OK image or an NG image, as in the embodiment, and executes a warning process if it is determined to be an NG image. If the learned program 202 determines that the image is an OK image, the processing device 10 processes the wafer 100. As a result, the processing device 10 can improve the accuracy of image determination by using the determination result of the learned program 202 that has learned teacher data having an OK image and an NG image actually captured. As a result, the processing device 10 can prevent processing an incorrect position of the wafer 100 that is not a planned dividing line when the planned dividing line is not captured in the center of the image. Furthermore, the processing device 10 can make a suitable determination for each wafer 100 or device 104 by using the learned program 202 that has learned actual images, thereby improving convenience.

The present invention is not limited to the above embodiment. In other words, the present invention can be implemented in various modifications without departing from the gist of the invention.

For example, in the above embodiment and modified example, the processing device 10 has been described as having a control unit 40 including a machine learning unit 41, an input unit 42, a control unit 43, a labeling unit 44, a teach unit 45, and an alignment unit 46, but is not limited to this. For example, the processing device 10 may be configured not to include the labeling unit 44, the teach unit 45, and the alignment unit 46, or may include any of the labeling unit 44, the teach unit 45, and the alignment unit 46.

In the above embodiment and modified examples, the learned program 202 is created by performing machine learning for each type of wafer 100, but this is not limited to the above. The learned program 202 may be created by performing machine learning on multiple types of wafers 100. In this way, the learned program 202 can be made a general-purpose program that has been machine-learned for various types of wafers 100, eliminating the need to create the program for each type of wafer 100 and increasing versatility.

REFERENCE SIGNS LIST 10 Processing device 11 Chuck table 12 Imaging unit 13 Processing unit 17 Touch panel 18 LED light 30 Storage 31 Image storage unit 40 Control unit 41 Machine learning unit 42 Input unit 43 Control unit 44 Labeling unit 45 Teach unit 46 Alignment unit 100 Wafer (workpiece)
103 Street (Planned division line)
104 Device 201 Control program 202 Learned program 300 Image data (teaching data)

Claims (5)

A processing apparatus for processing a workpiece having a device region formed on a surface thereof partitioned by a plurality of planned division lines, comprising:
The processing device is
A holding table for holding the workpiece;
a processing feed unit that feeds the holding table in the X direction;
a machining unit for forming a machining groove in the workpiece held on the holding table;
an indexing feed unit that indexes and feeds the processing unit in a Y direction intersecting with the X direction;
an imaging unit for imaging the workpiece held on the holding table;
A control unit;
Equipped with
The control unit
a machine learning unit that processes an input image and determines whether the input image is an OK image or an NG image with a trained program that has trained with teacher data having an OK image in which the planned division line is located at the center of an image obtained by capturing an image of the surface of a workpiece and an NG image in which the planned division line is not located at the center of the image;
an input unit that inputs an image of the planned dividing line of the workpiece to be processed placed on the holding table into the machine learning unit;
a control unit that controls the processing unit to process the workpiece to be processed when the machine learning unit determines that the image is an OK image, and executes a warning process when the machine learning unit determines that the image is an NG image;
A processing apparatus comprising: The control unit
A labeling unit is further provided for labeling the captured image of the surface of the workpiece into a good image and a bad image and storing the label as image data in the image recording unit.
The processing device according to claim 1 , wherein the machine learning unit determines whether an input image is an OK image or an NG image using a trained program that has learned the image data labeled by the labeling unit of the image recording unit as part of teacher data. The control unit
A teach unit that registers a key pattern formed on a surface of a workpiece and a distance from the key pattern to the planned division line;
an alignment unit that captures an image of a workpiece to be processed placed on the holding table, detects the key pattern by pattern matching to detect the division line of the workpiece, and positions the detected division line at the center of the imaging unit to capture an image;
Further equipped with
The processing apparatus according to claim 1 or 2, wherein the input unit inputs an image captured by the alignment unit to the machine learning unit. The processing unit includes:
4. The processing device according to claim 1, further comprising a cutting blade held at the tip of a rotatable spindle. The processing unit includes:
An oscillator that emits a laser beam;
a condenser that condenses the laser beam oscillated by the oscillator and irradiates the workpiece held on the holding table;
an optical system disposed between the oscillator and the condenser for directing a laser beam to the oscillator;
The processing apparatus according to claim 1 or 2, comprising:

Images