Wafer bubble detection method

Abstract

The invention provides a wafer bubble detection method comprising the following steps: generating a wafer template drawing according to at least two wafer detection pictures, the wafer template drawing comprising wafer circuit textures; and obtaining the wafer bubble defect of a to-be-detected image according to the difference between the to-be-detected image and the wafer template image. The invention also provides a wafer bubble detection method. The method comprises the following steps: generating a wafer template graph comprising wafer circuit textures; intercepting the wafer circuit texture of the wafer template graph as a feature graph; calculating a displacement distance and a rotation angle of a to-be-detected image relative to the feature image; rotating and translating the to-be-detected image to enable the to-be-detected image to be consistent with the wafer template image in direction; and obtaining the wafer bubble defect of the to-be-detected image according to the difference between the to-be-detected image and the wafer template image.

Description

Technical Field

[0001] The present application relates to the field of semiconductor wafer bonding, and more specifically, to a wafer bubble detection method. Background Art

[0002] In the wafer bonding process, due to residual particles on the wafer surface or air not discharged in time, bubble defects will be formed at the bonding interface, which will directly affect the subsequent process and cause the wafer yield to decrease. At present, the wafer bubble detection methods are mainly divided into infrared transmission method, ultrasonic method and X-ray imaging method, among which the infrared transmission method has the advantages of being fast, convenient and cheap. However, when the infrared transmission method is used to detect bubbles, the complex and diverse circuit patterns on the wafer will cause serious interference to the detection of bubbles. Therefore, how to effectively reduce the impact of wafer circuit texture on bubble detection and improve the accuracy of bubble detection is an urgent point to be solved. Summary of the invention

[0003] In view of this, the present application proposes a wafer bubble detection method to solve the above-mentioned problem.

[0004] According to an embodiment of the present application, a wafer bubble detection method is proposed. The wafer bubble detection method comprises generating a wafer template image based on at least two wafer inspection images, wherein the wafer template image comprises a wafer circuit texture; and obtaining a wafer bubble defect of the image to be inspected based on a difference between the image to be inspected and the wafer template image.

[0005] According to an embodiment of the present application, generating the wafer template image based on the at least two wafer inspection images includes: making the directions of the at least two wafer inspection images consistent through rotation and translation operations.

[0006] According to one embodiment of the present application, making the directions of the at least two wafer inspection pictures consistent through rotation and translation operations includes: intercepting the wafer circuit texture of one of the at least two wafer inspection pictures as a feature map; calculating the displacement distance and rotation angle of the remaining wafer inspection pictures in the at least two wafer inspection pictures relative to the feature map; and performing the rotation and translation operations based on the calculated displacement distance and rotation angle.

[0007] According to one embodiment of the present application, generating the wafer template image based on the at least two wafer inspection images also includes: comparing the grayscale value of each corresponding pixel of the at least two wafer inspection images; and selecting pixels with higher grayscale values ​​to generate new pixels at corresponding positions in the wafer template image.

[0008] According to an embodiment of the present application, obtaining the wafer bubble defect of the image to be inspected based on the difference between the image to be inspected and the wafer template image includes: rotating and translating the image to be inspected so that the direction of the image to be inspected is consistent with that of the wafer template image.

[0009] According to one embodiment of the present application, rotating and translating the image to be inspected so that the direction of the image to be inspected is consistent with that of the wafer template image includes: intercepting the wafer circuit texture of the wafer template image as a feature image; calculating the displacement distance and rotation angle of the image to be inspected relative to the wafer template image based on the feature image; and rotating and translating the image to be inspected based on the calculated displacement distance and rotation angle.

[0010] According to an embodiment of the present application, obtaining the wafer bubble defect of the image to be inspected based on the difference between the image to be inspected and the wafer template image also includes: subtracting the background of the image to be inspected based on the wafer template image to obtain the wafer bubble defect.

[0011] According to one embodiment of the present application, a wafer bubble detection method is proposed. The wafer bubble detection method includes: generating a wafer template image including a wafer circuit texture; intercepting the wafer circuit texture of the wafer template image as a first feature image; calculating the displacement distance and rotation angle of the image to be detected relative to the first feature image; rotating and translating the image to be detected so that the direction of the image to be detected is consistent with that of the wafer template image; and obtaining the wafer bubble defect of the image to be detected according to the difference between the image to be detected and the wafer template image.

[0012] According to an embodiment of the present application, generating the wafer template image including the wafer circuit texture includes: generating the wafer template image based on at least two wafer inspection images.

[0013] According to an embodiment of the present application, generating the wafer template image based on the at least two wafer inspection images includes: making the directions of the at least two wafer inspection images consistent through rotation and translation operations.

[0014] According to one embodiment of the present application, making the directions of the at least two wafer inspection pictures consistent through rotation and translation operations includes: intercepting the wafer circuit texture of one of the at least two wafer inspection pictures as a second feature map; calculating the displacement distance and rotation angle of the remaining wafer inspection pictures in the at least two wafer inspection pictures relative to the second feature map; and performing the rotation and translation operations based on the calculated displacement distance and rotation angle.

[0015] According to an embodiment of the present application, obtaining the wafer bubble defect of the image to be inspected based on the difference between the image to be inspected and the wafer template image includes: subtracting the background of the image to be inspected based on the wafer template image to obtain the wafer bubble defect.

[0016] In the wafer bubble detection method proposed in the present application, the wafer circuit texture of the wafer template image is used as a reference, and the background of the image to be detected is subtracted through image processing to obtain a wafer bubble defect image. In this way, the influence of the wafer circuit texture on the image to be detected can be effectively reduced, and the bubble defects on the image to be detected can be accurately obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the present application and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present application but do not constitute a limitation to the present application. In the accompanying drawings:

[0018] Figure 1 A flowchart illustrating a wafer bubble detection method according to an embodiment of the present application.

[0019] Figure 2 A schematic diagram demonstrating generation of a wafer template image from a wafer inspection image according to an embodiment of the present application.

[0020] Figure 3 A schematic diagram illustrating how to obtain a wafer bubble defect map by processing an image to be inspected through a wafer template map according to an embodiment of the present application.

[0021] Figure 4 A flowchart illustrating a wafer bubble detection method according to an embodiment of the present application.

[0022] Figure 5 A flowchart illustrating a wafer bubble detection method according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following disclosure provides a variety of implementations or illustrations that can be used to implement different features of the present disclosure. The specific examples of components and configurations described below are used to simplify the present disclosure. As can be imagined, these descriptions are only illustrative and are not intended to limit the present disclosure. For example, in the description below, forming a first feature on or above a second feature may include certain embodiments in which the first and second features are directly in contact with each other; and may also include certain embodiments in which additional components are formed between the above-mentioned first and second features, so that the first and second features may not be in direct contact. In addition, the present disclosure may reuse component symbols and / or labels in multiple embodiments. Such repetition is based on the purpose of simplicity and clarity, and does not itself represent the relationship between the different embodiments and / or configurations discussed.

[0024] Furthermore, spatially relative terms such as "under," "below," "below," "above," "above," and the like may be used herein to facilitate description of the relationship between one component or feature shown in the figure relative to another or more components or features. These spatially relative terms are intended to encompass a variety of different orientations of the device in use or operation in addition to the orientation shown in the figure. The device may be placed in other orientations (e.g., rotated 90 degrees or in other orientations), and these spatially relative descriptive terms should be interpreted accordingly.

[0025] Although the numerical ranges and parameters used to define the broader scope of the present application are approximate values, the relevant numerical values ​​in the specific embodiments have been presented as accurately as possible. However, any numerical value inherently inevitably contains standard deviations due to individual test methods. Here, "about" generally refers to the actual value within plus or minus 10%, 5%, 1% or 0.5% of a specific value or range. Alternatively, the term "about" means that the actual value falls within the acceptable standard error of the mean value, depending on the consideration of a person of ordinary skill in the art to which the present application belongs. It should be understood that, except for the experimental examples, or unless otherwise explicitly stated, all ranges, quantities, values ​​and percentages used herein (for example, to describe the amount of material used, the length of time, temperature, operating conditions, quantitative ratios and the like) are modified by "about". Therefore, unless otherwise stated to the contrary, the numerical parameters disclosed in this specification and the attached claims are all approximate values ​​and can be changed as needed. At least these numerical parameters should be understood as the number of significant digits indicated and the values ​​obtained by applying the general rounding method. Herein, numerical ranges are expressed from one endpoint to another endpoint or between two endpoints; unless otherwise stated, the numerical ranges described herein include the endpoints.

[0026] Figure 1 A flowchart illustrating a wafer bubble detection method 1 according to an embodiment of the present application. In some embodiments, the wafer bubble detection method 1 is configured to detect bubble defects formed at the bonding interface during the wafer bonding process. In some embodiments, the wafer bubble detection method 1 can be implemented by a device using an infrared transmission method. If similar results can be obtained, the present application is not limited to the method completely in accordance with Figure 1 Wafer bubble detection method 1 can be roughly summarized as follows:

[0027] Step 101: Generate a wafer template image;

[0028] Step 102: obtaining a to-be-detected image; and

[0029] Step 103: Obtain wafer bubble defects based on the difference between the image to be inspected and the wafer template image.

[0030] For details about step 101, please refer to Figure 2 . Figure 2 The following is a schematic diagram showing how to generate a wafer template image X1 using wafer inspection images Y1 and Y2 according to an embodiment of the present application. The wafer inspection images Y1 and Y2 can be wafer images obtained by photographing different wafers containing the same circuit pattern using, for example, an infrared transmission method, wherein the wafer inspection images Y1 and Y2 include a few bubble defects (e.g., Figure 2 The bubble defects Z1, Z2 and Z3 shown in the figure represent the bubble defects existing on the bonding interface of the photographed wafer. The wafer template image X1 generated based on the wafer inspection image Y1 and the wafer inspection image Y2 only includes the wafer circuit texture and does not have bubble defects.

[0031] In some embodiments, when the wafer inspection image Y1 and the wafer inspection image Y2 are obtained, the directions of the wafer inspection image Y1 and the wafer inspection image Y2 can be made consistent by rotation and translation. In some embodiments, the wafer circuit texture of the wafer inspection image Y1 can be first intercepted as a feature map. Specifically, the wafer portion in the wafer inspection image Y1 can be quasi-circularized, and then the interception range (for example, the center and radius of the circle obtained by fitting) can be determined. Figure 2 The wafer circuit texture in the wafer inspection picture Y1 is intercepted as a feature map by using the portion surrounded by the dotted frame H1 in Y1). In some embodiments, the interception range can be obtained by expanding the radius by a set value on the basis of the fitting circle. Then, the displacement distance and rotation angle of the wafer inspection picture Y2 relative to the feature map (that is, the displacement distance and rotation angle relative to the wafer inspection picture Y1) can be calculated according to the feature map. Finally, based on the calculated displacement distance and rotation angle, the wafer inspection picture Y2 is rotated and translated so that the direction of the wafer inspection picture Y2 is consistent with that of the wafer inspection picture Y1. Compared with matching using the entire picture, the use of feature maps can better ensure positioning accuracy.

[0032] When the directions of wafer inspection image Y1 and wafer inspection image Y2 are consistent, the grayscale values ​​of each corresponding pixel of wafer inspection image Y1 and wafer inspection image Y2 can be compared and the pixel with higher grayscale value can be selected to generate a new pixel at the corresponding position in wafer template image X1 (for example, the grayscale value of the new pixel is the grayscale value of the selected pixel). Generally speaking, the grayscale value of the bubble part in the picture will be lower than the grayscale value of the wafer circuit part in the picture. In this way, when the bubble defects (such as wafer inspection image Y1 and wafer inspection image Y2) are Figure 2When the bubble defects (Z1, Z2 and Z3) shown in the figure are located at different pixels, by comparing the grayscale values ​​of each corresponding pixel of the wafer inspection image Y1 and the wafer inspection image Y2 and selecting the pixel with the higher grayscale value to generate a new pixel at the corresponding position in the wafer template image X1, each pixel of the wafer template image X1 finally obtained will represent the wafer circuit texture part instead of the bubble defect. In other words, the wafer template image X1 will only have the wafer circuit texture without the bubble defect.

[0033] Those skilled in the art should understand that in other embodiments, more than two wafer inspection images can be used to generate the wafer template image X1, which can reduce the possibility of overlapping bubble positions on different images and further reduce noise. For example, in the case of three wafer inspection images, one wafer inspection image can be first selected to generate a feature image, and the feature image can be used to make the directions of all wafer inspection images consistent, and then two wafer inspection images are used to generate an intermediate wafer template image according to the above method, and then the third wafer inspection image and the intermediate wafer template image are used to generate the final wafer template image X1 according to the above method.

[0034] In addition, if a wafer image without bubble defects can be directly obtained, it can be directly used as the wafer template image X1, and there is no need to synthesize the wafer template image X1 by comparing grayscale values.

[0035] For details about step 103, please refer to Figure 3 . Figure 3 A schematic diagram is provided to demonstrate the process of processing the image D1 to be inspected by the wafer template image X1 to obtain the wafer bubble defect image E1 according to an embodiment of the present application. The image D1 to be inspected can be a wafer image obtained by photographing the wafer to be inspected by, for example, an infrared transmission method. There are many bubble defects on the image, which represent the bubble defects existing on the bonding interface of the wafer to be inspected. The wafer to be inspected and the wafer photographed in the wafer inspection image contain the same circuit pattern. In order to obtain the bubble defects of the wafer to be inspected, after processing the wafer template image X1 by, for example, an infrared transmission method, the wafer template image D1 can be processed by the wafer template image X1 to obtain the bubble defect image E1. Figure 2 After the wafer template image X1 is obtained by the method described in the embodiment, the directions of the image to be detected D1 and the wafer template image X1 can be made consistent by rotation and translation. In some embodiments, the wafer circuit texture of the wafer template image X1 can be first intercepted as a feature image. Specifically, the wafer portion in the wafer template image X1 can be quasi-circularized, and then the interception range (for example, the center and radius of the circle obtained by fitting) can be determined. Figure 3The portion surrounded by the dotted box H2 in X1 is used to intercept the wafer circuit texture in the wafer template image X1 as a feature map. In some embodiments, the interception range can be obtained by expanding the radius by a set value on the basis of the fitting circle. Then, the displacement distance and rotation angle of the image to be detected D1 relative to the feature map (i.e., the displacement distance and rotation angle relative to the wafer template image X1) can be calculated according to the feature map. Finally, based on the calculated displacement distance and rotation angle, the image to be detected D1 is rotated and translated so that the direction of the image to be detected D1 is consistent with that of the wafer template image X1.

[0036] When the directions of the image to be inspected D1 and the wafer template image X1 are consistent, it can be considered that the backgrounds of the two are consistent. Therefore, the wafer circuit texture of the wafer template image X1 can be used as a reference, and the background of the image to be inspected D1 can be subtracted through image processing (for example, subtracting each pixel value of the image to be inspected D1 from the pixel value at the corresponding position of the wafer template image X1), thereby obtaining a wafer bubble defect image E1, wherein the wafer bubble defect image E1 presents the wafer bubble defect on the image to be inspected D1. In some embodiments, after obtaining the wafer bubble defect image E1, the wafer bubble defect image E1 can be binarized to obtain a binarized image E2 with a clearer bubble image.

[0037] In the wafer bubble detection method 1 proposed in the present application, the wafer circuit texture of the wafer template image X1 is used as a reference, and the background of the image to be inspected D1 is subtracted to obtain the wafer bubble defect image E1, thereby effectively reducing the influence of the wafer circuit texture on the image to be inspected D1 and accurately obtaining the bubble defect on the image to be inspected D1.

[0038] Figure 4 A flowchart illustrating a wafer bubble detection method 4 according to an embodiment of the present application. In some embodiments, the wafer bubble detection method 4 is configured to detect bubble defects formed at the bonding interface during the wafer bonding process. In some embodiments, the wafer bubble detection method 4 can be implemented by a device using an infrared transmission method. If similar results can be obtained, the present application is not limited to the method completely in accordance with Figure 4 The wafer bubble detection method 4 can be roughly summarized as follows:

[0039] Step 401: Generate a wafer template image according to at least two wafer inspection images; and

[0040] Step 402: Obtain wafer bubble defects of the image to be inspected based on the difference between the image to be inspected and the wafer template image.

[0041] In step 401, at least two wafer inspection images can be referenced Figure 2 The wafer inspection picture Y1 and the wafer inspection picture Y2 in the embodiment and the wafer template picture can be referred to Figure 2The wafer template image X1 in the embodiment, the specific details of generating the wafer template image can be referred to Figure 2 In step 402, the specific details of the wafer bubble defect of the image to be inspected are obtained according to the difference between the image to be inspected and the wafer template image, which can be referred to but not limited to Figure 3 The method shown in the embodiment.

[0042] Figure 5 A flowchart illustrating a wafer bubble detection method 5 according to an embodiment of the present application. In some embodiments, the wafer bubble detection method 5 is configured to detect bubble defects formed at the bonding interface during the wafer bonding process. In some embodiments, the wafer bubble detection method 5 can be implemented by a device using an infrared transmission method. If similar results can be obtained, the present application is not limited to the method completely in accordance with Figure 5 The wafer bubble detection method 5 can be roughly summarized as follows:

[0043] Step 501: Generate a wafer template image including a wafer circuit texture;

[0044] Step 502: intercepting the wafer circuit texture of the wafer template image as a feature image;

[0045] Step 503: Calculate the displacement distance and rotation angle of the image to be detected relative to the feature image;

[0046] Step 504: rotating and translating the image to be inspected so that the direction of the image to be inspected is consistent with the direction of the wafer template image; and

[0047] Step 505: Obtain wafer bubble defects of the image to be inspected according to the difference between the image to be inspected and the wafer template image.

[0048] In step 501, the specific details of generating the wafer template map can refer to but are not limited to Figure 2 The details of step 502 to step 504 can be referred to Figure 3 In step 505, the specific details of the wafer bubble defect of the image to be inspected are obtained according to the difference between the image to be inspected and the wafer template image, which can be referred to but not limited to Figure 3 The method shown in the embodiment.

[0049] Although the embodiments of the present application are described in conjunction with wafer images obtained by the infrared transmission method, those skilled in the art should understand that the method proposed in the present application can also be applied to images with similar features obtained by other methods.

[0050] As used herein, the terms "approximately", "substantially", "substantially" and "about" are used to describe and explain small variations. When used in conjunction with an event or circumstance, the terms may refer to instances where the event or circumstance occurred exactly as well as instances where the event or circumstance occurred very approximately.

[0051] As used herein, the singular terms "a", "an" and "the" may include plural referents unless the context clearly dictates otherwise. In the description of some embodiments, a component provided "on" or "over" another component may encompass the case where the former component is directly on (e.g., physically in contact with) the latter component, as well as the case where one or more intermediate components are located between the former component and the latter component.

[0052] The foregoing summarizes several embodiments and detailed features of the present disclosure. The embodiments described in the present disclosure can be easily used as a basis for designing or modifying other processes and structures for performing the same or similar purposes and / or obtaining the same or similar advantages of the embodiments introduced herein. These equivalent constructions do not depart from the spirit and scope of the present disclosure and various changes, substitutions and modifications may be made without departing from the spirit and scope of the present disclosure.

Claims

1. A wafer bubble detection method, characterized in that: include: Generating a wafer template image according to at least two wafer inspection images, wherein the wafer template image includes a wafer circuit texture; as well as The wafer bubble defect of the image to be inspected is obtained according to the difference between the image to be inspected and the wafer template image.

2. The wafer bubble detection method according to claim 1, characterized in that: Generating the wafer template image according to the at least two wafer inspection images includes: The directions of the at least two wafer inspection pictures are made consistent through rotation and translation operations.

3. The wafer bubble detection method according to claim 2, characterized in that: Making the directions of the at least two wafer inspection images consistent by rotation and translation operations includes: intercepting a wafer circuit texture of one wafer inspection picture among the at least two wafer inspection pictures as a feature map; Calculating the displacement distance and rotation angle of the remaining wafer inspection pictures in the at least two wafer inspection pictures relative to the feature map; and The rotation and translation operations are performed based on the calculated displacement distance and rotation angle.

4. The wafer bubble detection method according to claim 2, characterized in that: Generating the wafer template image according to the at least two wafer inspection images also includes: comparing the grayscale value of each corresponding pixel of the at least two wafer inspection images; and Pixels with higher grayscale values ​​are selected to generate new pixels at corresponding positions in the wafer template image.

5. The wafer bubble detection method according to claim 1, characterized in that: The wafer bubble defect of the image to be inspected obtained according to the difference between the image to be inspected and the wafer template image includes: The image to be inspected is rotated and translated so that the direction of the image to be inspected is consistent with that of the wafer template image.

6. The wafer bubble detection method according to claim 5, characterized in that: Rotating and translating the image to be inspected so that the direction of the image to be inspected is consistent with the direction of the wafer template image includes: intercepting the wafer circuit texture of the wafer template image as a feature image; Calculating the displacement distance and rotation angle of the image to be inspected relative to the wafer template image according to the feature image; and The image to be detected is rotated or translated based on the calculated displacement distance and rotation angle.

7. The wafer bubble detection method according to claim 6, characterized in that: Obtaining the wafer bubble defect of the image to be inspected according to the difference between the image to be inspected and the wafer template image also includes: The wafer bubble defect is obtained by subtracting the background of the image to be inspected based on the wafer template image.

8. A wafer bubble detection method, characterized in that: include: generating a wafer template image including a wafer circuit texture; intercepting the wafer circuit texture of the wafer template image as a first feature image; Calculating the displacement distance and rotation angle of the image to be detected relative to the first feature image; Rotating and translating the image to be inspected so that the direction of the image to be inspected is consistent with that of the wafer template image; as well as The wafer bubble defect of the image to be inspected is obtained according to the difference between the image to be inspected and the wafer template image.

9. The wafer bubble detection method according to claim 8, characterized in that: Generating the wafer template image including the wafer circuit texture includes: The wafer template image is generated according to at least two wafer inspection images.

10. The wafer bubble detection method according to claim 9, characterized in that: Generating the wafer template image according to the at least two wafer inspection images includes: The directions of the at least two wafer inspection pictures are made consistent through rotation and translation operations.

11. The wafer bubble detection method according to claim 10, characterized in that: Making the directions of the at least two wafer inspection images consistent by rotation and translation operations includes: Intercepting a wafer circuit texture of one wafer inspection picture among the at least two wafer inspection pictures as a second feature picture; Calculating the displacement distance and rotation angle of the remaining wafer inspection pictures in the at least two wafer inspection pictures relative to the second feature map; and The rotation and translation operations are performed based on the calculated displacement distance and rotation angle.

12. The wafer bubble detection method according to claim 8, characterized in that: The wafer bubble defect of the image to be inspected obtained according to the difference between the image to be inspected and the wafer template image includes: The wafer bubble defect is obtained by subtracting the background of the image to be inspected based on the wafer template image.

Images