Wafer determination method, determination program, determination device, wafer manufacturing method, and wafer
By eliminating the misjudgment candidate images and generating teaching data models, the defect determination accuracy of the chip is improved, the problem of low judgment accuracy in the prior art is solved, and the accuracy of product quality judgment is achieved.
Patent Information
- Application Number
- CN202380033370.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-06-15
- Filing Date
- 2023-05-02
- Publication Date
- 2024-11-26
AI Technical Summary
In the prior art, the defect determination accuracy of the wafer is low, resulting in an increase in visual inspection working hours and affecting product quality.
By taking images of the chip, the candidate images for misjudgment are eliminated, the teaching data model is generated, the chip is qualified, and the determination accuracy is improved.
It improves the quality judgment accuracy of chip products, reduces misjudgment, and enhances the accuracy of the judgment of whether the product is qualified or not.
Smart Images

Figure CN119032268A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a wafer determination method, a determination program, a determination device, a wafer manufacturing method, and a wafer. Background Art
[0002] Conventionally, there is known a method for classifying wafer defects using a wafer defect image (see Patent Document 1, etc.).
[0003] Prior art literature Patent Literature Patent document 1: Japanese Patent Application Publication No. 2021-174980. Summary of the invention
[0004] Problems to be solved by the invention In order to eliminate visual inspection or reduce the man-hours of visual inspection, research is being conducted to automatically determine the defects of wafers. In this case, when photographing the wafer to be inspected, an image that is difficult to determine as a defect is sometimes generated due to an abnormality in the photographing device. Such an image sometimes reduces the accuracy of defect determination. It is required to improve the quality of products such as wafers or products using wafers as materials by improving the accuracy of defect determination.
[0005] Therefore, an object of the present disclosure is to propose a determination method, a determination program, a determination device, a wafer manufacturing method, and a wafer that can improve product quality.
[0006] Solutions to Solve Problems One embodiment of the present disclosure that solves the above-mentioned problems is as follows. [1] A determination method, comprising: A step of acquiring an image capturing at least a portion of a wafer as a determination image for determining whether the wafer is acceptable or not; A step of excluding the captured image from the determination image when the captured image corresponds to an erroneously determined candidate image; and A step of determining whether the wafer is acceptable based on the determination image. [2] According to the determination method described in [1] above, the following step is also included: in at least one of the cases where at least a part of the range of the chip reflected in the captured image is defective, or a specified part of the chip is reflected in the captured image, the captured image is determined to be consistent with the misjudgment candidate image. [3] The determination method according to the above-mentioned [1] further includes the following step: generating a model for determining whether the captured image corresponds to the misjudgment candidate image using teaching data including the misjudgment candidate image. [4] A determination program causing a processor to execute the determination method described in any one of [1] to [3] above. [5] A determination device comprising a control unit that executes the determination method described in any one of [1] to [3] above. [6] A method for manufacturing a wafer, comprising the following step: determining whether the wafer is qualified or not by executing any one of the determination methods described in [1] to [3] above. [7] A wafer judged as acceptable by executing the judgment method described in any one of [1] to [3] above.
[0007] Effects of the Invention According to the wafer determination method, determination program, determination device, wafer manufacturing method, and wafer disclosed herein, it is possible to improve product quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Figure 1 This is a block diagram showing a configuration example of a determination system according to one embodiment. Figure 2 It is a plan view showing a structural example of a wafer. Figure 3 This is a diagram showing an example of an image obtained by taking a full circle of images of the end surface of a wafer. Figure 4A yes Figure 3 An enlarged view of portion A surrounded by a frame line. Figure 4B yes Figure 3 An enlarged view of portion B surrounded by a frame line. Figure 4C yes Figure 3 An enlarged view of portion C surrounded by a frame line. Figure 4D yes Figure 3 An enlarged view of portion D surrounded by a frame line. Figure 5A FIG. 1 is a diagram showing an example of a captured image of a wafer end surface including a defective portion. Figure 5B FIG. 1 is a diagram showing an example of a captured image of a wafer surface including a defective portion. Figure 6 FIG. 1 is a diagram showing an example of a captured image showing the outer side of an end of a wafer surface. Figure 7 FIG. 1 is a diagram showing an example of a captured image of a wafer end surface including a notch portion. Figure 8 The diagram shows an example of an image showing the wafer end surface that matches the erroneous determination candidate image and an example that does not match the erroneous determination candidate image. Fig. 9 The diagram shows an example of an image showing the wafer surface that matches an erroneous determination candidate image and an example that does not match an erroneous determination candidate image. Fig.10 This is a flowchart showing an example of the operation procedure of the determination method according to one embodiment. DETAILED DESCRIPTION
[0009] (Configuration Example of Determination System 1) like Figure 1 As shown, the determination system 1 includes a determination device 10 and an imaging device 20. The imaging device 20 captures images of products such as wafers in the process of manufacturing products such as wafers, and is used to determine whether the products are qualified or not. The determination device 10 obtains the images captured by the imaging device 20, and determines whether the products are qualified or not based on the obtained images.
[0010] The determination system 1 of this embodiment can determine whether the appearance of the product meets the factory standard based on the image of the appearance of the product. That is, the determination system 1 can determine whether the appearance of the product is qualified or not. Not limited to the appearance of the product, the determination system 1 can also determine whether the internal state of the product is qualified or not based on an image such as an X-ray image that shows the internal state of the product. The determination system 1 of this embodiment can also obtain an image of the appearance or internal state of a wafer as a product, and determine whether the appearance or internal state of the wafer is qualified or not.
[0011] <Determination Device 10> The determination device 10 includes a control unit 12 , a storage unit 14 , and an interface 16 .
[0012] The control unit 12 determines whether the product is qualified or not based on the image of the product obtained from the camera 20 through the interface 16, and outputs the determination result through the interface 16. The control unit 12 may include at least one processor. The processor can execute programs that implement various functions of the control unit 12. The processor can be implemented as a single integrated circuit. An integrated circuit is also called an IC (Integrated Circuit). The processor can be implemented as a plurality of integrated circuits and discrete circuits that can be communicatively connected. The processor can be implemented based on various other known technologies.
[0013] The storage unit 14 may include an electromagnetic storage medium such as a magnetic disk, or may include a memory such as a semiconductor memory or a magnetic memory. The storage unit 14 may include a non-temporary computer-readable medium. The storage unit 14 stores various information such as images obtained from the camera 20 and programs executed by the control unit 12. The storage unit 14 may function as a working memory of the control unit 12. At least a part of the storage unit 14 may be included in the control unit 12. At least a part of the storage unit 14 may also be configured as a storage device separate from the determination device 10.
[0014] The interface 16 can be configured to include a communication module, which is configured to be able to communicate with the camera 20 so as to obtain images from the camera 20. The communication module can be communicatively connected to the camera 20 in a wired or wireless manner. The communication module can be directly connected to the camera 20 or connected via a communication network. The communication module can have a communication interface such as a LAN (Local Area Network). The communication module can also have a communication interface for non-contact communication such as infrared communication or NFC (Near Field communication). The communication module can implement communication using various communication methods such as 4G (4th Generation) or LTE (Long Term Evolution) or 5G (5th Generation). The communication method performed by the communication module is not limited to the above examples, and may also include various other methods. At least a portion of the communication module may be included in the control unit 12.
[0015] The interface 16 can be configured to include an output device so that the user can be notified of the determination result by the control unit 12. The output device may include a display device that outputs visual information such as images, characters, or graphics. The display device can be configured to include, for example, an LCD (Liquid Crystal Display), an organic EL (Electro-Luminescence) display, an inorganic EL display, or a PDP (Plasma Display Panel). Not limited to these displays, the display device can also be configured to include displays in various other ways. The display device can be configured to include a light-emitting device such as an LED (Light Emitting Diode) or an LD (Laser Diode). The display device can be configured to include various other devices. The output device may include a speaker that outputs sound, etc. Not limited to these examples, the output device may include a device that can output information in various other ways.
[0016] The interface 16 may be configured to include an input device that receives, for example, an operation input such as starting or stopping the determination of the product by the determination device 10, or other various instruction inputs to the determination device 10. The interface 16 outputs information input from the user to the control unit 12. The input device may be configured to include a touch panel or a touch sensor, or an indicating device such as a mouse. The input device may be configured to include physical keys. The input device may be configured to include a sound input device such as a microphone.
[0017] <Camera 20> The imaging device 20 may be configured to include various cameras such as a visible light camera, an infrared camera, or an X-ray camera, etc. The imaging device 20 may also be configured to include a light source such as a visible light source or an X-ray source for irradiating when imaging a product such as a wafer.
[0018] The imaging device 20 is configured to capture at least a portion of a product such as a wafer. Figure 2 When the wafer 30 is shown, for example, the end face 32 of the wafer 30 can be photographed all around, and as shown in FIG. Figure 3 As shown, an image of the appearance of the end surface 32 of the wafer 30 is generated as a long image. Figure 3 The image of the end surface 32 of the wafer 30 shown includes repeated intervals of photographing the same place to include the entire circumference of the end surface 32. In order to photograph the end surface 32 of the wafer 30 all around, the photographing device 20 can be configured so that the wafer 30 rotates relative to a fixed camera or the camera rotates around the periphery of the wafer 30.
[0019] The wafer 30 has a defect 36 on the end surface 32. The defect 36 may include a flaw or chipping generated on the end surface 32 or the surface 31 of the wafer 30. The defect 36 may include foreign matter such as dust attached to the end surface 32 or the surface 31 of the wafer 30. Figure 3 Defects 36 shown in the image include defect 36A and defect 36B indicating flaws.
[0020] The wafer 30 has a notch 34 as a mark showing the direction of the crystal axis of the wafer 30. The notch 34 is formed as a cut inward from the end surface 32 of the wafer 30. The notch 34 looks like a cut when viewed from the surface 31 of the wafer 30. The notch 34 looks like a depression when viewed from the end surface 32 of the wafer 30.
[0021] like Figure 4A , 4B 4C and 4D, the photographing device 20 can generate an image in which a portion of the end surface 32 is cut out. Figure 4A Corresponds to Figure 3 The enlarged image of the area surrounded by the dotted line indicated by A in FIG. Figure 4AThe enlarged image of includes neither the defect 36 nor the notch 34 . Figure 4B Corresponds to Figure 3 The enlarged image of the area surrounded by the dotted line indicated by B in the figure. Figure 4B The enlarged image of includes defect 36A. Figure 4C Corresponds to Figure 3 The enlarged image of the area surrounded by the dotted line indicated by C in FIG. Figure 4C The enlarged image of includes defect 36B. Figure 4D Corresponds to Figure 3 The enlarged image of the area surrounded by the dotted line indicated by D in FIG. Figure 4D The enlarged image of includes the notch 34 .
[0022] The imaging device 20 may also generate an image of a portion of the end surface 32 of the wafer 30 cut out from a long image generated by imaging the end surface 32 of the wafer 30 all around. The imaging device 20 may image the end surface 32 of the wafer 30 and generate an image of a portion where the defect 36 may appear.
[0023] When photographing the surface 31 of the wafer 30, the imaging device 20 may generate an image in which a portion of the surface 31 of the wafer 30 is cut out, based on the entire surface image generated by scanning the surface 31 of the wafer 30. The imaging device 20 may scan the surface 31 of the wafer 30 to generate an image of a portion where the defect 36 may be reflected.
[0024] As described above, the imaging device 20 is configured to image at least a portion of the wafer 30. The imaging device 20 outputs an image of at least a portion of the wafer 30 to the determination device 10.
[0025] (Operation Example of Determination System 1) In the determination system 1, the imaging device 20 generates an image of at least a portion of the wafer 30, and outputs the image to the determination device 10. The image of at least a portion of the wafer 30 captured by the imaging device 20 is also referred to as a captured image. The determination device 10 obtains the captured image generated by the imaging device 20 as an image to be determined through the interface 16. The image to be determined by the determination device 10 is also referred to as a determination image. The control unit 12 of the determination device 10 determines whether the wafer 30 is acceptable or not based on the determination image.
[0026] The captured image may include an image including the defect 36. On the other hand, the captured image may include an image not including the defect 36. The captured image may include an image that may be erroneously determined as the defect 36 by the control unit 12 although it does not include the defect 36. When the control unit 12 uses an image that may be erroneously determined as the defect 36 although it is not the defect 36 in the determination, the control unit 12 is likely to erroneously determine whether the wafer 30 is acceptable or not.
[0027] Therefore, the control unit 12 excludes images that are not defects 36 but may be erroneously determined as defects 36 from the images to be determined. Images that are not defects 36 but may be erroneously determined as defects 36 are also referred to as erroneous determination candidate images.
[0028] The control unit 12 determines whether the captured image obtained from the imaging device 20 corresponds to the erroneous determination candidate image. The control unit 12 uses the captured image determined not to correspond to the erroneous determination candidate image as the determination image to determine whether the wafer 30 is acceptable or not. The control unit 12 does not use the captured image determined to correspond to the erroneous determination candidate image as the determination image. That is, the control unit 12 excludes the captured image determined to correspond to the erroneous determination candidate image from the determination image.
[0029] The control unit 12 may generate a model for determining whether a captured image matches a false judgment candidate image. The model for determining whether a captured image matches a false judgment candidate image is also referred to as a determination model for a false judgment candidate image. The determination model for a false judgment candidate image is configured to output a determination result of whether the input captured image matches a false judgment candidate image. The control unit 12 may generate a model for determining whether the chip 30 is acceptable or not based on a determination image after excluding captured images that match false judgment candidate images. The model for determining whether the chip 30 is acceptable or not based on a determination image after excluding captured images that match false judgment candidate images is also referred to as a determination model for a defect 36. The determination model for a defect 36 is configured to output a determination result of whether the chip 30 shown in the determination image is acceptable or not based on the input determination image.
[0030] The control unit 12 can use the judgment model of the wrong judgment candidate image and the judgment model of the defect 36 in conjunction. The control unit 12 can input the captured image to the judgment model of the wrong judgment candidate image, and exclude the wrong judgment candidate image from the judgment image based on the judgment result output from the judgment model of the wrong judgment candidate image. The control unit 12 can input the judgment image after excluding the wrong judgment candidate image to the judgment model of the defect 36, and judge whether the wafer 30 shown in the captured image initially input to the judgment model of the wrong judgment candidate image is acceptable or not based on the judgment result output from the judgment model of the defect 36.
[0031] The control unit 12 may also implement the determination of whether the captured image matches the false judgment candidate image and the pass / fail determination of the wafer 30 based on the determination image after excluding the captured image matching the false judgment candidate image through one model. A model that implements the determination of the false judgment candidate image and the pass / fail determination is also called a composite determination model. The control unit 12 may input the captured image into the composite determination model and determine the pass / fail of the wafer 30 shown in the captured image input into the composite determination model based on the determination result output from the composite determination model.
[0032] <Determination of Misjudged Candidate Images> The determination model of the misjudged candidate image can be generated by learning using teaching data, wherein the teaching data includes at least one of an image that matches the misjudged candidate image or an image that does not match the misjudged candidate image. The determination model of the misjudged candidate image can be generated as a pattern matching model, wherein the pattern matching model determines at least one of a pattern that matches the misjudged candidate image or a pattern that does not match the misjudged candidate image. Without being limited to these, the determination model of the misjudged candidate image can be generated so as to be able to determine the misjudged candidate image using various algorithms. The control unit 12 can generate the determination model of the misjudged candidate image by the control unit 12 itself, or can obtain it from an external device.
[0033] The images that correspond to the misjudgment candidate images are the following exemplified images. Figure 5A and Figure 5B As shown, the captured image 40 including the defective portion 44 corresponds to the erroneous determination candidate image.
[0034] exist Figure 5A In the captured image 40 shown, the end face 32 of the wafer 30 and the background portion 42 without the wafer 30 are shown. The control unit 12 is configured to ignore the background portion 42 in the captured image 40 and determine whether the wafer 30 is acceptable based on the image of the portion from the left end to the right end of the captured image 40 where the end face 32 is shown. Figure 5A In the captured image 40 shown in FIG. 1 , a portion of the image on the right side is a black image due to an abnormality of the imaging device 20 or the like. That is, a portion of the image of the end surface 32 that should be shown at the right end is missing. The missing portion of the image of the end surface 32 is indicated as a missing portion 44. The control unit 12 Figure 5A When the captured image 40 shown is used as a determination image, there is a possibility that the missing portion 44 is erroneously determined to be the defect 36 .
[0035] In addition, Figure 5BIn the captured image 40 shown, the surface 31 of the wafer 30 and the defective portion 44 which is a black image are reflected. That is, the image of the surface 31 of the wafer 30 is missing. The control unit 12 is configured to determine whether the wafer 30 is acceptable or not based on the image of the surface 31 of the wafer 30 reflected on the entire surface of the captured image 40. However, Figure 5B The image of the interior and the end of the captured image 40 shown in the figure is a black image due to an abnormality of the imaging device 20 or the like. That is, a part of the image of the surface 31 is missing. The missing part of the image of the surface 31 is indicated as a missing part 44. The control unit 12 Figure 5B When the captured image 40 shown is used as a determination image, there is a possibility that the missing portion 44 is erroneously determined to be the defect 36 .
[0036] In addition, for example, Figure 6 In the captured image 40 shown in FIG. 1 , the portion of the wafer 30 outside the surface 31 where the wafer 30 is not present is shown as a background portion 42. When the control unit 12 determines whether the wafer 30 is acceptable or not based on the premise that the surface 31 is shown on the entire surface of the captured image 40, it is possible that the control unit 12 Figure 6 The background portion 42 reflected in the captured image 40 shown is erroneously determined to be the defect 36 .
[0037] In addition, for example, Figure 7 In the captured image 40 shown, a notch 34 is shown in the end surface 32 of the wafer 30. Figure 7 When the captured image 40 shown is used as a determination image, there is a possibility that the notch 34 is erroneously determined to be a defect 36 .
[0038] In addition, Figure 8 In the figure, the captured image 40 of the end surface 32 of the wafer 30 is classified into examples that meet the misjudgment candidate image and examples that do not meet the misjudgment candidate image. As an example of a misjudgment candidate image, an image showing a notch 34 is shown. As an example of a misjudgment candidate image, an image showing a chipping defect 36 is shown. In addition, in Fig. 9 In the figure, the captured image 40 of the surface 31 of the wafer 30 is classified into examples that meet the misjudgment candidate image and examples that do not meet the misjudgment candidate image. As an example of a misjudgment candidate image, an image in which a point 38 generated by a laser marker is reflected on the surface 31 of the wafer 30 is shown. As an example of an image that does not meet the misjudgment candidate image, an image in which a pinhole defect 36 is reflected is shown.
[0039] As described above, the captured image 40 can be classified according to whether it matches the erroneous determination candidate image. The control unit 12 can determine that the captured image 40 matches the erroneous determination candidate image when at least a portion of the range in which the wafer 30 is shown in the captured image 40 is defective. Alternatively, the control unit 12 can determine that the captured image 40 matches the erroneous determination candidate image when a predetermined portion such as the notch 34 of the wafer 30 or the spot 38 of the laser marker is shown in the captured image 40.
[0040] The control unit 12 may first exclude images showing a predetermined portion such as the notch 34 or the dot 38 from the captured image 40. The control unit 12 may further exclude images including the defective portion 44 or the unnecessary background portion 42 from the images not showing the predetermined portion in the captured image 40. That is, the determination model of the erroneously determined candidate image may be divided into a determination model of an image showing the predetermined portion and a determination model of an image showing the defective portion 44 or the like.
[0041] <Determination of Misjudged Candidate Images> The control unit 12 determines whether the wafer 30 is acceptable or not based on the determination image showing the wafer 30. The control unit 12 may determine that the wafer 30 is acceptable when the defect 36 is not shown in the determination image showing the wafer 30. The control unit 12 may determine that the wafer 30 is unacceptable when the defect 36 is shown in the determination image showing the wafer 30.
[0042] The determination model of the defect 36 can be generated by learning using teaching data, the teaching data including at least one of an image including the defect 36 or an image not including the defect 36. The determination model of the defect 36 can be generated as a pattern matching model that determines at least one of a pattern that matches the defect 36 or a pattern that does not match the defect 36. Without being limited to these, the determination model of the defect 36 can be generated so as to be able to determine the presence or absence of the defect 36 using various algorithms. The control unit 12 can generate the determination model of the defect 36 by the control unit 12 itself, or can obtain it from an external device.
[0043] Defect 36 may include chipping in end surface 32 of wafer 30, such as Figure 8 Defect 36 may include a pinhole in surface 31 of wafer 30, such as Fig. 9 The defect 36 may be in various forms such as dust or dirt attached to the surface 31 or the end surface 32 of the wafer 30, without being limited to these examples.
[0044] As described above, the control unit 12 determines whether the wafer 30 has the defect 36 based on the determination image, and determines that the wafer 30 is defective if the wafer 30 has the defect 36. That is, the control unit 12 determines whether the wafer 30 is good or bad based on the determination image.
[0045] When the control unit 12 uses a composite judgment model formed by combining the judgment model of the false judgment candidate image and the judgment model of the defect 36 to judge whether the wafer 30 is acceptable or not, the composite judgment model can be generated by the control unit 12 itself or obtained from an external device. The composite judgment model can be generated by learning an image that matches the false judgment candidate image and an image that does not match the false judgment candidate image and includes the defect 36 or an image that does not include the defect 36 as teaching data. The composite judgment model can be generated as a pattern matching model that determines at least one of a pattern that matches the false judgment candidate image or a pattern that does not match the false judgment candidate image and includes the defect 36 or a pattern that does not include the defect 36.
[0046] <Example of Flowchart of Determination Method> The control unit 12 of the determination device 10 can execute the following steps: Fig.10 The determination method of the operation sequence of the flowchart shown is used to determine whether the wafer 30 is good or bad. The determination method can also be implemented as a determination program that the control unit 12 executes.
[0047] The control unit 12 obtains a determination model of an erroneously determined candidate image (step S1). The control unit 12 obtains a captured image 40 from the imaging device 20 as a determination image (step S2). The control unit 12 determines whether the captured image 40 corresponds to an erroneously determined candidate image (step S3). If the captured image 40 does not correspond to an erroneously determined candidate image (step S3: No), the control unit 12 proceeds to the operation sequence of step S5. If the captured image 40 corresponds to an erroneously determined candidate image (step S3: Yes), the control unit 12 excludes the captured image 40 from the determination image (step S4).
[0048] The control unit 12 obtains the determination model of the defect 36 (step S5). The control unit 12 applies the determination image obtained in the operation sequence of steps S1 to S4 to the determination model of the defect 36 to determine whether the determination image is qualified or not (step S6). If the determination image is qualified, the control unit 12 determines that the chip 30 shown in the determination image is qualified, and if the determination image is unqualified, the chip 30 shown in the determination image is unqualified. After the execution of the operation sequence of step S6, the control unit 12 ends. Fig.10 The execution of the jobs in the flowchart sequence.
[0049] exist Fig.10In the example of the flowchart, the control unit 12 separately uses the determination model of the misjudgment candidate image and the determination model of the defect 36, but a composite determination model may be used.
[0050] <Example> In the determination system 1 of the present embodiment, an example is described in which an effect of determining whether or not a wafer 30 is qualified by determining whether or not it matches a false determination candidate image is verified. In the example, the result of determining whether or not the wafer 30 is qualified by a person viewing the captured image 40 is compared with the result of determining whether or not the wafer 30 is qualified by the control unit 12 of the determination device 10 based on the captured image. The combination of the result of the qualified determination by a person and the result of the qualified determination by the control unit 12 is classified into four types.
[0051] The combination in which the determination by the person and the determination by the control unit 12 are consistent is classified into the following two types. When the determination of failure by the person and the determination of failure by the control unit 12 for a certain wafer 30 are consistent, the determination of failure by the control unit 12 for the wafer 30 is classified as a true positive (TP). On the other hand, when the determination of pass by the person and the determination of pass by the control unit 12 for a certain wafer 30 are consistent, the determination of pass by the control unit 12 for the wafer 30 is classified as a true negative (TN).
[0052] Combinations in which the determination by a person and the determination by the control unit 12 are inconsistent are classified into the following two types. When a certain wafer 30 is determined to be unqualified by a person and to be qualified by the control unit 12, the determination of the pass of the wafer 30 by the control unit 12 is classified as a false negative (FN). On the other hand, when a certain wafer 30 is determined to be qualified by a person and to be unqualified by the control unit 12, the determination of the unqualified wafer 30 by the control unit 12 is classified as a false positive (FP).
[0053] The accuracy of the judgment is calculated based on the frequency of being classified into each type. The value (TP+TN) / (TP+TN+FP+FN) obtained by dividing the sum of the number of samples classified as true positive and the number of samples classified as true negative by the total number of samples is calculated as the accuracy rate. The accuracy rate indicates the proportion of correct judgments. It can be said that the higher the accuracy rate, the higher the judgment accuracy.
[0054] The value obtained by dividing the number of samples classified as true positives by the number of samples classified as true positives and false positives (TP / (TP+FP)) is calculated as the compliance rate. The compliance rate indicates the proportion of samples that are actually "unqualified" among the samples that are judged as "unqualified". The higher the compliance rate, the lower the possibility that the product will be wasted and discarded.
[0055] The value obtained by dividing the number of samples classified as true positives by the number of samples classified as true positives and false negatives (TP / (TP+FN)) is calculated as the reproducibility. The reproducibility indicates the proportion of samples judged as "failed" among samples that are actually "failed". The higher the reproducibility, the lower the possibility of "failed" products flowing out.
[0056] Table 1 shows the results of classifying the combination of the determination result when the control unit 12 of the determination device 10 excludes the erroneously determined candidate image from the determination image and the determination result by a human into the above four types. [Table 1]
[0057] The accuracy calculated from the results shown in Table 1 is (42437+27) / (42437+1093+27)=97.5%. The coincidence rate is 27 / (1093+27)=2.4%. The reproducibility rate is 27 / 27=100%.
[0058] On the other hand, as a comparative example, Table 2 shows the combination of the judgment results when the captured image 40 is used directly as the judgment image to judge whether the chip 30 is qualified without judging whether the captured image 40 corresponds to the misjudgment candidate image, and the judgment results by humans, which are classified into the above four types of results. [Table 2]
[0059] The accuracy calculated from the results of the comparative example shown in Table 2 is (42437+27) / (42437+2872+27)=93.7%. The coincidence rate is 27 / (2872+27)=0.9%. The reproducibility rate is 27 / 27=100%.
[0060] In the comparison between Table 1 and Table 2, in the determination system 1 of the present embodiment, the accuracy and the coincidence rate are increased by excluding the erroneously determined candidate images from the captured images 40 in order to determine whether the wafer 30 is acceptable or not. Therefore, by excluding the erroneously determined candidate images from the captured images 40, the accuracy of determining whether the wafer 30 is acceptable or not is improved.
[0061] (Summarize) As described above, the determination system 1, the determination device 10, and the determination method of the present embodiment exclude erroneously determined candidate images from the captured image 40 capturing at least a portion of the product in order to determine whether the product such as the wafer 30 is acceptable or not. As a result, it is difficult to erroneously determine the defect 44 or the background 42 which is not the defect 36 as the defect 36. As a result, the accuracy of determining whether the product such as the wafer 30 is acceptable or not can be improved.
[0062] A model such as a determination model or a composite determination model of the defect 36 may be configured to output a result of determining whether a product such as the wafer 30 meets the factory standard. In the case where the product is the wafer 30, the model may be configured to output a result of determining whether the wafer 30 meets the factory standard. The classification in this case is the simplest classification of qualified or unqualified. The model may, for example, classify the product into a plurality of quality levels. In the case where the product is the wafer 30, the model may be configured to output a result of determining whether the wafer 30 is a grade used for equipment use or a grade used for monitor use, for example, based on the quality of the wafer 30. The quality level may be determined based on the number or size of the defects 36, or the type of the defects 36.
[0063] (Other embodiments) A method for manufacturing the wafer 30 including the step of executing the method for determining whether the wafer 30 is good or not according to the present embodiment can be realized. Also, a wafer 30 determined to be good by executing the determination method can be realized.
[0064] Although the embodiments of the present disclosure have been described based on the various drawings and embodiments, it should be noted that as long as one is skilled in the art, various deformations or changes can be made based on the present disclosure. Therefore, it should be noted that these deformations or changes are included in the scope of the present disclosure. For example, the functions included in each component or each step can be reconfigured in a logically non-contradictory manner, and multiple components or steps can be combined into one or divided. Although the embodiments of the present disclosure are described with the device as the center, the embodiments of the present disclosure can also be implemented as a method including the steps performed by each component of the device. The embodiments of the present disclosure can also be implemented as a method executed by a processor possessed by the device, a program, or a storage medium having a program recorded thereon. It should be understood that the scope of the present disclosure also includes these.
[0065] Industrial Applicability According to the embodiments of the present disclosure, the quality of products can be improved.
[0066] Description of Reference Numerals 1: Judgment system 10: determination device (12: control unit, 14: storage unit, 16: interface) 20: Shooting device 30: Wafer (31: Surface, 32: End face, 34: Notch, 36, 36A, 36B: Defect, 38: Laser marked point) 40: captured image (42: background part, 44: defective part).
Claims
1. A determination method, in, include: A step of acquiring an image capturing at least a portion of a wafer as a determination image for determining whether the wafer is acceptable; A step of excluding the captured image from the determination image when the captured image matches the erroneously determined candidate image; as well as A step of determining whether the wafer is acceptable based on the determination image.
2. The determination method according to claim 1, in, The method further includes the following steps: determining that the captured image matches the misjudgment candidate image in at least one of the following cases: a case where at least a portion of the range of the chip reflected in the captured image is defective, or a case where a specified portion of the chip is reflected in the captured image.
3. The determination method according to claim 1, in, The method further includes the step of generating a model for determining whether the captured image corresponds to the erroneously determined candidate image using teaching data including the erroneously determined candidate image.
4. A determination procedure, in, A processor is caused to execute the determination method according to any one of claims 1 to 3.
5. A determination device, in, The method comprises a control unit configured to execute the determination method according to any one of claims 1 to 3.
6. A method for manufacturing a wafer, in, The method comprises the following steps: determining whether a wafer is qualified or not by executing the determination method according to any one of claims 1 to 3.
7. A chip, in, The determination as being acceptable is made by executing the determination method according to any one of claims 1 to 3.
