Wafer micro-tube defect detection method and device, storage medium and computer equipment

Abstract

The application discloses a wafer micro-tube defect detection method and device, a storage medium and a computer device. The method comprises the following steps: acquiring multiple surface images of a wafer under the same collection angle, which are to be subjected to micro-tube defect detection; determining the maximum gray value of each position pixel point in the multiple surface images respectively, and determining a pretreatment image of the wafer region under the collection angle according to the maximum gray value of each position pixel point; performing preliminary micro-tube defect detection on the pretreatment image to obtain a suspected micro-tube defect region included in the pretreatment image; determining the bright region and the dark region on the surface image corresponding to each preset light source, and verifying the suspected micro-tube defect region according to the bright region and the dark region of the surface image corresponding to each preset light source, so as to determine whether the suspected micro-tube defect region is a real micro-tube defect region. The application can quickly and efficiently detect the micro-tube defect of the wafer in a non-contact manner, and is simple, easy and highly automated.

Description

Technical Field

[0001] The present application relates to the technical field of semiconductor integrated circuit defect detection, and in particular to a method and apparatus for detecting wafer micropipe defects, a storage medium, and a computer device. Background Art

[0002] Wafers are the fundamental raw material for manufacturing semiconductor devices. The front-end wafer manufacturing process includes multiple steps, including silicon ingot growth, cutting, polishing, and grinding. These processes are tedious and complex, and can cause defects such as scratches, edge collapse, and cracks on the wafers. Micropipe defects primarily occur during the silicon ingot growth phase. Micropipe defects can also appear after the ingots are cut. Micropipe defects are defects that extend throughout the entire wafer product, ranging in size from a few microns to tens of microns. The penetrating nature of micropipe defects can directly impact the quality and subsequent yield of semiconductor products, severely impacting back-end processes. Therefore, wafers must be inspected after manufacturing to promptly identify unqualified wafers and prevent them from being passed on to subsequent processes.

[0003] In the prior art, when detecting micropipe defects on wafers, microscopic observation is the main method. Inspectors can use a microscope to observe the wafer surface and measure the dimensions, but this detection method requires manual observation, is time-consuming, and has low efficiency. In addition, a liquid penetration method can be used, which takes advantage of the penetrating characteristics of micropipe defects and covers the upper surface of the wafer with a detection liquid. The location and number of micropipe defects can be determined by detecting whether the detection liquid exists on the lower surface. However, this method requires contact detection, which can easily cause wafer contamination. It is also inefficient and not suitable for large-scale production. Summary of the Invention

[0004] In view of this, the present application provides a method and device for detecting micropipe defects in wafers, a storage medium, and a computer device, which can perform fast, efficient, and non-contact micropipe defect detection on wafers under conventional imaging methods. This method is simple and easy to implement, and has a high degree of automation. At the same time, the detection scheme performs detection in the first process after the silicon ingot is cut, thereby realizing the control of micropipe defects in the front end, thereby avoiding waste of resources in the wafer production stage.

[0005] According to one aspect of the present application, a method for detecting wafer micropipe defects is provided, comprising:

[0006] Acquire multiple surface images corresponding to a wafer to be inspected for micropipe defects at the same acquisition angle, wherein the multiple surface images include surface images acquired after each preset light source is individually activated, and the position of each preset light source relative to the wafer is determined based on the physical characteristics of the micropipe defect;

[0007] Determining the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and determining a pre-processed image of the wafer area under the acquisition viewing angle according to the maximum grayscale value of the pixel at each position;

[0008] performing preliminary microtubule defect detection on the preprocessed image to obtain a suspected microtubule defect region included in the preprocessed image;

[0009] For the surface image corresponding to each preset light source, the bright areas and dark areas on the surface image are determined, and each suspected micropipe defect area is verified based on the bright areas and dark areas of the surface image corresponding to each preset light source to determine whether the suspected micropipe defect area is a real micropipe defect area.

[0010] According to another aspect of the present application, a device for detecting micropipe defects in a wafer is provided, comprising:

[0011] a surface image acquisition module, configured to acquire multiple surface images corresponding to a wafer to be inspected for micropipe defects, taken from the same acquisition angle, wherein the multiple surface images include surface images acquired after each preset light source is individually activated, and the position of each preset light source relative to the wafer is determined based on the physical characteristics of the micropipe defect;

[0012] a pre-processing image generating module, configured to respectively determine the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and determine a pre-processing image of the wafer area under the acquisition viewing angle according to the maximum grayscale value of the pixel at each position;

[0013] a preliminary defect detection module, configured to perform preliminary micropipe defect detection on the preprocessed image to obtain a suspected micropipe defect region included in the preprocessed image;

[0014] The defect verification module is used to determine the bright areas and dark areas on the surface image corresponding to each preset light source, and verify each suspected micropipe defect area based on the bright areas and dark areas of the surface image corresponding to each preset light source to determine whether the suspected micropipe defect area is a real micropipe defect area.

[0015] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the above-mentioned method for detecting wafer micropipe defects is implemented.

[0016] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned method for detecting wafer micropipe defects when executing the program.

[0017] By the technical scheme, the wafer micro-tube defect detection method and device, the storage medium and the computer device provided by the application first acquire multiple surface images of a wafer to be detected for micro-tube defect detection under the same collection angle. Here, when the image collection device collects the surface images, one preset light source is turned on each time, and the image collection device is controlled to collect one or more surface images of the wafer, so that multiple surface images of the wafer collected under each preset light source can be obtained. The setting position of each preset light source relative to the wafer is determined based on the physical characteristics of the micro-tube defect. After the multiple surface images under each preset light source are obtained, the maximum gray value of each same position pixel point in the multiple surface images can be determined, and then the maximum gray value of each position pixel point can be used to determine a pretreatment image of the wafer region under the collection angle of the surface images. After the pretreatment image of the wafer region is obtained, the pretreatment image can be subjected to preliminary micro-tube defect detection, so that a suspected micro-tube defect region can be found in the pretreatment image. For each surface image corresponding to each preset light source, the bright region and the dark region on the surface image are determined. After the bright region and the dark region of the surface image under each preset light source are determined, each suspected micro-tube defect region in the pretreatment image can be verified according to the bright region and the dark region of the surface image under each preset light source, to determine whether the suspected micro-tube defect region is bright or dark in the surface image of the preset light source, and finally, whether each suspected micro-tube defect region is a real micro-tube defect region can be determined according to the corresponding bright-dark judgment result under each preset light source. The embodiment of the application can perform rapid and efficient non-contact micro-tube defect detection on the wafer under the conventional imaging method, and the method is simple and easy to implement and has high automation degree. Meanwhile, the detection scheme detects the wafer in the first process after the silicon ingot is cut, so that the micro-tube defect can be controlled in the front process, thereby avoiding resource waste in the wafer production stage.

[0018] The above description is only a summary of the technical scheme of the application. In order to more clearly understand the technical means of the application, the application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0019] The drawings described herein are used to provide further understanding of the application, and form a part of the application. The schematic embodiments of the application and the description thereof are used to explain the application, and do not constitute an improper limitation on the application. In the drawings:

[0020] Figure 1 A flowchart of a wafer micro-tube defect detection method provided by an embodiment of the application is shown;

[0021] Figure 2A top view showing a configuration method of a preset light source provided in an embodiment of the present application is shown;

[0022] Figure 3 A front view showing a setting method of a preset light source provided in an embodiment of the present application is shown;

[0023] Figure 4 shows the surface image corresponding to the light source 1 provided in the embodiment of the present application;

[0024] Figure 5 shows the surface image corresponding to the light source 2 provided in the embodiment of the present application;

[0025] Figure 6 shows the surface image corresponding to the light source 3 provided in the embodiment of the present application;

[0026] Figure 7 shows the surface image corresponding to the light source 4 provided in the embodiment of the present application;

[0027] Figure 8 An initial preprocessing diagram provided by an embodiment of the present application is shown;

[0028] Figure 9 A schematic structural diagram of a wafer micropipe defect detection device provided in an embodiment of the present application is shown;

[0029] Figure 10 A schematic diagram of the device structure of a computer device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0030] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0031] In this embodiment, a method for detecting micropipe defects in a wafer is provided. Figure 1 As shown, the method includes:

[0032] Step 101, obtaining multiple surface images corresponding to the same acquisition perspective of the wafer to be inspected for micropipe defects, wherein the multiple surface images include surface images acquired after each preset light source is turned on separately, and the setting position of each preset light source relative to the wafer is determined based on the physical characteristics of the micropipe defect.

[0033] An embodiment of the present application provides a method for detecting micropipe defects on a wafer, which can automatically identify whether a wafer has micropipe defects without contact with the wafer. Before performing micropipe defect detection on the wafer, an image acquisition device is set on one side of the wafer, and multiple preset light sources are set on the other side; or the image acquisition device and the multiple preset light sources are set on the same side of the wafer. Among them, the setting position of the preset light source relative to the wafer can be determined according to the physical characteristics of the micropipe defect, so that by setting the preset light source at different positions, and the image acquisition device only turns on one preset light source at a time when collecting surface images of the wafer, it can be determined whether the through defect is a micropipe defect based on the surface image collected later. It should be noted that one side and the other side of the wafer refer to the upper and lower surfaces of the wafer. When the preset light source and image acquisition device are positioned on opposite sides of the wafer, if the preset light source is positioned on the upper surface of the wafer, the image acquisition device is positioned on the lower surface of the wafer; if the preset light source is positioned on the lower surface of the wafer, the image acquisition device is positioned on the upper surface of the wafer. When the preset light source and image acquisition device are positioned on the same side of the wafer, the distance between the preset light source and the image acquisition device and the same surface of the wafer can be determined based on actual circumstances and is not limited herein. Specifically, multiple surface images of the wafer can be captured using the same image acquisition device, with the position of the image acquisition device relative to the wafer remaining unchanged. This allows the capture angle of each captured surface image to be the same. When the image acquisition device captures surface images, it activates each preset light source individually and controls the image acquisition device to capture one or more surface images of the wafer. This allows the acquisition of multiple surface images of the wafer captured under each preset light source. For example, if there are four preset light sources, the multiple surface images captured by the image acquisition device include at least four. In one embodiment, the image acquisition device can be a line array camera image sensor or an area array camera image sensor. When using a line scan camera or area scan camera image sensor, the surface image captured can be either a grayscale image or a color image. Since subsequent image processing is performed on the surface image, if the surface image is a color image, the color image can be converted to a grayscale image before performing subsequent operations.

[0034] Step 102 : determining the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and determining a pre-processed image of the wafer area under the acquisition viewing angle according to the maximum grayscale value of each pixel at each position.

[0035] In this embodiment, after obtaining multiple surface images under various preset light sources, the maximum grayscale value of each pixel at the same position in the multiple surface images can be determined respectively. Then, based on the maximum grayscale value of each pixel at each position, the pre-processed image of the wafer area under the acquisition viewing angle of the surface image can be determined. For example, for a pixel (xi ,y i ), since the acquisition angles of multiple surface images are the same, the pixel point (x i ,y i ), and determine the pixel point (x i ,y i ) in different surface images, and the maximum gray value among these gray values ​​is used as the pixel point (x i ,y i ) in the pre-processed image. Due to the circular shape of the wafer, when the image acquisition device captures the wafer surface image, the surface image under certain acquisition angles may partially represent the wafer area, while the remaining portion represents the wafer placement surface area. However, since the analysis is for micropipe defects on the wafer, only the pre-processed image of the wafer area needs to be determined.

[0036] Step 103 : performing preliminary micropipe defect detection on the preprocessed image to obtain suspected micropipe defect regions included in the preprocessed image.

[0037] In this embodiment, after obtaining a pre-processed image of the wafer region, a preliminary micropipe defect detection can be performed on the pre-processed image, thereby identifying suspected micropipe defect regions within the pre-processed image. Here, the suspected micropipe defect regions can be bright spots in the pre-processed image, representing a through-hole defect. Because micropipe defects are themselves through-hole defects, suspected micropipe defect regions are first screened based on the bright spots in the pre-processed image. Subsequently, the suspected micropipe defect regions are screened based on a determination of which are actual micropipe defect regions.

[0038] In step 104, for each surface image corresponding to a preset light source, the bright area and the dark area on the surface image are determined, and each suspected micropipe defect area is verified based on the bright area and the dark area of ​​the surface image corresponding to each preset light source to determine whether the suspected micropipe defect area is a real micropipe defect area.

[0039] In this embodiment, for each surface image corresponding to a preset light source, the bright and dark areas on the surface image are determined. When there are multiple surface images corresponding to a preset light source, these surface images can be merged to obtain a merged surface image, and then the bright and dark areas are determined based on the merged surface image. When there is only one surface image corresponding to a preset light source, the bright and dark areas of the surface image can be directly determined. After determining the bright and dark areas of the surface image under each preset light source, each suspected micropipe defect area in the preprocessing image can be verified based on the bright and dark areas of the surface image under each preset light source to determine whether the suspected micropipe defect area is bright or dark in the surface image under the preset light source. Finally, based on the brightness and darkness judgment results corresponding to each preset light source, it is determined whether each suspected micropipe defect area is a true micropipe defect area.

[0040] By applying the technical solution of this embodiment, multiple surface images corresponding to the same acquisition angle are first acquired for the wafer to be inspected for micropipe defects. Here, when the image acquisition device acquires surface images, each preset light source is activated individually, and the image acquisition device is controlled to acquire one or more surface images of the wafer. In this way, multiple surface images of the wafer acquired under each preset light source are obtained. The position of each preset light source relative to the wafer is determined based on the physical characteristics of micropipe defects. After obtaining multiple surface images under each preset light source, the maximum grayscale value of each pixel at the same position in the multiple surface images can be determined. Then, based on the maximum grayscale value of each pixel at each position, a pre-processed image of the wafer region under the acquisition angle from which the surface images were acquired can be determined. After obtaining the pre-processed image of the wafer region, preliminary micropipe defect detection can be performed on the pre-processed image, thereby identifying suspected micropipe defect areas within the pre-processed image. For each surface image acquired corresponding to the preset light source, bright and dark areas within the surface image are determined. After determining the bright and dark areas of the surface image under each preset light source, each suspected micropipe defect area in the preprocessing image can be verified based on the bright and dark areas of the surface image under each preset light source to determine whether the suspected micropipe defect area is bright or dark in the surface image under the preset light source. Finally, based on the corresponding light and dark judgment results under each preset light source, it is determined whether each suspected micropipe defect area is a real micropipe defect area. The embodiment of the present application can perform fast, efficient, and non-contact micropipe defect detection on wafers under conventional imaging methods. This method is simple and easy to implement, with a high degree of automation. At the same time, the detection scheme performs detection in the first process after the silicon ingot is cut, realizing the control of micropipe defects in the front-end, thereby avoiding resource waste in the wafer production stage.

[0041] In an embodiment of the present application, optionally, the image acquisition device setting surface and the preset light source setting surface are distributed on both sides or the same side of the wafer, the preset light source is set on the preset light source setting surface, and the distance between each preset light source and the projection point of the center of the wafer on the preset light source setting surface is the same, the angle between each two adjacent preset light sources and the projection point is 90°, the irradiation direction of each preset light source is toward the wafer, and the angle formed between the irradiation direction of each preset light source and the wafer surface is equal.

[0042] In this embodiment, the wafer includes upper and lower surfaces. Specifically, an image acquisition device placement surface can be defined on the outer side of one surface of the wafer, and a preset light source placement surface can be defined on the outer side of the other surface of the wafer. That is, the image acquisition device placement surface and the preset light source placement surface are located on either side of the upper and lower surfaces of the wafer. Furthermore, the image acquisition device placement surface and the preset light source placement surface can also be located on the same side of the wafer surface. The image acquisition device can be located on the image acquisition device placement surface, and the preset light source can be located on the preset light source placement surface. The center of the wafer can be projected onto the preset light source placement surface, with each preset light source being equidistant from the projection point of the wafer center on the preset light source placement surface, and the angle formed by the line connecting each two adjacent preset light sources and the projection point being 90 degrees. That is, four preset light sources are located on the preset light source placement surface. Furthermore, the illumination direction of each preset light source is toward the wafer, and the angles formed between the illumination directions of different preset light sources and the wafer surface are equal.

[0043] like Figure 2 As shown, a setting method of a preset light source is shown. Figure 2 It is a top view, and the four preset light sources are set on the preset light source setting surface that is different from the upper and lower surfaces of the wafer. The setting angles are 0°, 90°, 180° and 270° respectively. It should be noted that the above angles are only relative angles, and the 0° direction can be set arbitrarily. The distance between each preset light source and the projection point of the wafer center on the preset light source setting surface is equal, and the angle formed by the line between each two adjacent preset light sources and the above projection points is 90 degrees. Figure 3 As shown, Figure 2 The main view of the setup mode, where Figure 3Only two preset light sources are shown, namely light source 1 and light source 3. The angle formed between each preset light source and the wafer surface is α, and the light source irradiation angle is toward the wafer. In the embodiment of the present application, a preset light source is set at each of the four angles. Each time a preset light source is turned on separately, an image is captured by an image acquisition device. This is because micropipe defects are through-defects. In order to capture micropipe defects, images are captured from four angles. Different angles will have different brightness and darkness changes. This feature can be used to distinguish micropipe defects from other defects. At the same time, when only one surface image is captured under each preset light source, because the four surface images are only the changes of the preset light source, the image acquisition device does not move, so a group of surface images are obtained. The position of each surface image is consistent and there is no offset, which is convenient for subsequent further processing.

[0044] In an embodiment of the present application, optionally, step 102 includes: respectively determining the maximum grayscale value corresponding to each pixel point at the same position in the multiple surface images, and obtaining an initial preprocessing image based on the maximum grayscale value of the pixel point at each position; selecting a target surface image from the multiple surface images, and for each pixel point in the target surface image, determining the relationship between the grayscale value corresponding to the pixel point in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold; if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, then taking the pixel point as the first target pixel point, and determining the wafer area corresponding to the target surface image based on the first target pixel point, wherein the first preset grayscale threshold is less than the second preset grayscale threshold; aligning the initial preprocessing image with the target surface image, and intercepting the initial preprocessing image based on the wafer area in the target surface image to obtain a preprocessing image of the wafer area under the acquisition viewing angle.

[0045] In this embodiment, a pre-processed image of the wafer area under the acquisition viewing angle can be obtained in the following manner. First, for each pixel at the same position in multiple surface images, the grayscale value of the pixel in each surface image is determined, and then the maximum grayscale value is determined from these grayscale values. Then, an initial pre-processed image is generated based on the maximum grayscale value of the pixel at each position. This image reflects the maximum brightness or grayscale performance of the pixel at each position in all surface images. Figures 4 to 7 As shown, Figure 2 In the preset light source setting mode, the surface image under each preset light source captured by the image acquisition device, Figure 8 The initial pre-processed images are obtained based on these four surface images. The gray areas in these images are the wafer areas.

[0046] Next, one of the multiple surface images is selected as the target surface image. The target surface image can be any one of the multiple surface images, or a surface image selected based on factors such as image quality and clarity. Since the multiple surface images correspond to the same acquisition perspective and only differ in the preset light sources, the wafer area captured in these surface images is identical. For each pixel in the target surface image, its grayscale value is checked to see if it falls between two preset grayscale thresholds. The two preset grayscale thresholds include a first preset grayscale threshold and a second preset grayscale threshold, with the first preset grayscale threshold being less than the second preset grayscale threshold. If the grayscale value of a pixel is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, the pixel is marked as a first target pixel. The selected first target pixels are then used to construct the wafer area in the target surface image. The initial pre-processed image generated above is then registered with the target surface image. Registration refers to the spatial alignment of two or more images. After registration, the initial pre-processed image can be intercepted according to the wafer area determined in the target surface image, and only the part of the initial pre-processed image corresponding to the wafer area is retained, thereby obtaining a pre-processed image of the wafer area under the acquisition viewing angle. The embodiment of the present application synthesizes the initial pre-processed image by the maximum grayscale value of the pixel points at the same position in multiple surface images, so that the initial pre-processed image can include "bright spots" under the illumination angles of each preset light source, so that the initial pre-processed image effectively includes all through-defect areas, that is, all potential areas where micropipe defects may exist. The initial pre-processed image is then intercepted using the wafer area of ​​the target surface image to obtain a pre-processed image of the wafer area. Subsequently, micropipe defects are identified from the pre-processed image of the wafer area, which can effectively reduce the range of micropipe defect recognition, thereby improving the accuracy and efficiency of micropipe defect recognition.

[0047] In an embodiment of the present application, optionally, step 102 includes: selecting a target surface image from multiple surface images, and for each pixel point in the target surface image, determining the relationship between the grayscale value corresponding to the pixel point in the target surface image and the first preset grayscale threshold and the second preset grayscale threshold; if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, then taking the pixel point as the first target pixel point, and determining the wafer area corresponding to the target surface image based on the first target pixel point, wherein the first preset grayscale threshold is less than the second preset grayscale threshold; according to the wafer area corresponding to the target surface image, intercepting each surface image to obtain a surface image containing only the wafer area, and according to the maximum grayscale value corresponding to each pixel point at the same position in the multiple surface images containing only the wafer area, obtaining a pre-processed image of the wafer area under the acquisition perspective.

[0048] In this embodiment, a pre-processed image of the wafer area under the acquisition perspective can also be obtained in the following manner. First, one of the multiple surface images is selected as the target surface image. Here, the target surface image can be any one of the multiple surface images, or a surface image selected based on factors such as image quality and clarity. Since the multiple surface images correspond to the same acquisition perspective and only differ in the preset light sources, the wafer area captured in the surface images is the same. For each pixel in the target surface image, its grayscale value is checked to see if it falls between two preset grayscale thresholds. Here, the two preset grayscale thresholds include a first preset grayscale threshold and a second preset grayscale threshold, with the first preset grayscale threshold being less than the second preset grayscale threshold. If the grayscale value of a pixel is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, the pixel is marked as a first target pixel. Subsequently, the selected first target pixels are used to form the wafer area in the target surface image. Then, the wafer area in the target surface image is used as a template or reference to perform interception processing on each surface image. Specifically, each surface image is registered with the target surface image, and then the portion corresponding to the wafer area in the target surface image is extracted from each surface image. In this way, a plurality of surface images containing only the wafer area can be obtained. For these surface images containing only the wafer area, for each pixel at the same position, the grayscale value of the pixel in the above-mentioned images is determined, and the maximum grayscale value is further determined from these grayscale values. According to the maximum grayscale value of the pixel at each position, a pre-processed image of the wafer area can be generated. This pre-processed image reflects the maximum brightness or grayscale performance of the pixel at each position in the wafer area under the acquisition viewing angle. The embodiment of the present application synthesizes the pre-processed image of the wafer area by using the maximum grayscale value of the pixel at the same position in a plurality of surface images that only include the wafer area, so that the pre-processed image of the wafer area includes "bright spots" under the illumination angles of each preset light source, so that the pre-processed image effectively includes all through-defect areas, that is, includes all potential areas where micropipe defects may exist. The embodiment of the present application identifies micropipe defects from a pre-processed image of a wafer area, which can effectively reduce the micropipe defect identification range, thereby improving the accuracy and efficiency of micropipe defect identification.

[0049] In the above embodiment, the first preset grayscale threshold and the second preset grayscale threshold can be determined empirically, specifically by counting the grayscale values ​​of pixels corresponding to the wafer region in the wafer surface image. That is, the first preset grayscale threshold and the second preset grayscale threshold are used to distinguish the background from the foreground in the surface image, that is, to extract the wafer region in the surface image.

[0050] In an embodiment of the present application, optionally, step 103 includes: performing mean filtering on the preprocessed image to obtain a mean filtered image; subtracting the grayscale value of the corresponding pixel point in the preprocessed image from the grayscale value of each pixel point in the mean filtered image to obtain the grayscale difference corresponding to each pixel point, and obtaining a difference image based on the grayscale difference corresponding to each pixel point; for each pixel point in the difference image, determining the relationship between the grayscale value corresponding to the pixel point in the difference image and a third preset grayscale threshold, and taking the pixel point whose grayscale value is greater than the third preset grayscale threshold as the second target pixel point, and taking the area composed of the second target pixel points as the suspected microtubule defect area.

[0051] In this embodiment, after obtaining the pre-processed image corresponding to the wafer area, the pre-processed image can be mean filtered to obtain a mean filtered image. Mean filtering is a technique for smoothing images. It reduces noise in the image by replacing the value of each pixel in the image with the average value of the surrounding pixels. This can smooth the image and reduce the impact of random noise on subsequent steps. Next, the grayscale value of each pixel in the mean filtered image is subtracted from the grayscale value of the corresponding pixel in the pre-processed image to obtain the grayscale difference of each pixel. This operation can calculate the difference between the smoothed image and the original pre-processed image. After obtaining the grayscale difference corresponding to each pixel, a difference map can be generated based on these grayscale differences. The grayscale value of each pixel in the difference map represents the grayscale difference between the pre-processed image and the mean filtered image at that pixel. Next, for each pixel in the difference map, the grayscale value is compared with the third preset grayscale threshold. If the grayscale value of a certain pixel is greater than the third preset grayscale threshold, it is considered that this pixel may be caused by a micropipe defect, because micropipe defects usually cause large grayscale changes in local areas of the image. Afterwards, this pixel can be marked as a second target pixel. The area composed of all second target pixels is regarded as a suspected micropipe defect area. It should be noted that the above-mentioned second target pixels may constitute a plurality of unconnected areas. In this case, each area can be regarded as a suspected micropipe defect area, so that all suspected micropipe defect areas included in the preprocessing image can be obtained. In addition, the third preset grayscale threshold can also be determined based on experience, or after a large number of surface images are processed according to the above steps to obtain a difference map, it is determined by statistically analyzing the grayscale values ​​of the pixels that pass through the area in the difference map, and used to determine the suspected micropipe defect detection area from the difference map. The embodiment of the present application identifies possible micropipe defect areas by judging the size relationship between the grayscale value of the pixel point in the difference map and the third preset grayscale threshold, which can greatly improve the detection accuracy of the suspected micropipe defect area.

[0052] In an embodiment of the present application, optionally, when the surface image corresponding to any preset light source includes only one, the "for each surface image corresponding to the preset light source, determining the bright area and the dark area on the surface image" in step 104 includes: for each pixel point in the surface image corresponding to any preset light source, determining the relationship between the grayscale value of the pixel point in the surface image corresponding to any preset light source and the fourth preset grayscale threshold and the fifth preset grayscale threshold; if the grayscale value is greater than the fourth preset grayscale threshold and less than the fifth preset grayscale threshold, the pixel point is used as the third target pixel point, and the bright area of ​​the surface image corresponding to any preset light source is determined according to the third target pixel point, and the remaining area is used as the dark area, wherein the fourth preset grayscale threshold is less than the fifth preset grayscale threshold.

[0053] In this embodiment, the surface image corresponding to a preset light source or certain preset light sources may be one. For a preset light source that includes only one surface image, for each pixel in the surface image, the relationship between its grayscale value and two preset grayscale thresholds is determined. Here, the two preset grayscale thresholds may include a fourth preset grayscale threshold and a fifth preset grayscale threshold, with the fourth preset grayscale threshold being less than the fifth preset grayscale threshold. For each pixel in the surface image, if the grayscale value of a pixel in the surface image is greater than the fourth preset grayscale threshold and less than the fifth preset grayscale threshold, the pixel is considered a third target pixel. It should be noted that the fourth and fifth preset grayscale thresholds can be determined empirically or based on historical surface images. The fourth and fifth preset grayscale thresholds are set to extract bright and dark areas from the surface image. By setting the fourth and fifth preset grayscale thresholds, the grayscale values ​​of pixels in the micropipe defect area can be limited, so that the extracted bright area is closer to the "bright spot" corresponding to the micropipe defect. Subsequently, the bright area in the surface image is determined based on all third target pixels. The remaining area (i.e., the area consisting of pixels whose grayscale values ​​are not between the fourth preset grayscale threshold and the fifth preset grayscale threshold) is considered a dark area. This embodiment of the present application compares the relationship between the grayscale value of each pixel in the surface image and the fourth preset grayscale threshold and the fifth preset grayscale threshold, classifies the pixels into a third target pixel and other pixels, determines the bright area of ​​the surface image based on the third target pixel, and treats the remaining area as a dark area. This effectively extracts "bright spots" in the surface image caused by micropipe defects.

[0054] In the embodiments of the present application, optionally, when the surface image corresponding to any preset light source includes multiple images, before the step of determining, for each pixel point in the surface image corresponding to the any preset light source, the relationship between the gray value of the pixel point in the surface image corresponding to the any preset light source and the fourth preset gray threshold and the fifth preset gray threshold, the method further includes: performing merging processing on the multiple surface images corresponding to the any preset light source according to the maximum gray value of the pixel points at the same position in the multiple surface images corresponding to the any preset light source, to replace the surface images.

[0055] In the embodiments, the surface image corresponding to some preset light source can also include multiple images. For this case, the multiple surface images can be merged first, and then the bright area and the dark area are extracted from the merged surface image. Specifically, if the surface image corresponding to a preset light source includes multiple images, for each pixel point at the same position in these surface images, the gray value of the pixel point in the multiple surface images corresponding to the preset light source is determined, and the maximum value among these gray values is found. The maximum gray value found above is used to perform merging processing on the multiple surface images, so as to obtain a merged surface image, and then the subsequent extraction of the bright area and the dark area is performed by using the merged surface image. Specifically, for the merged surface image, the relationship between the gray value of each pixel point in the surface image and two preset gray thresholds is determined, where the two preset gray thresholds can include the fourth preset gray threshold and the fifth preset gray threshold, and the fourth preset gray threshold is less than the fifth preset gray threshold. For each pixel point in the merged surface image, if the gray value of a certain pixel point in the surface image is greater than the fourth preset gray threshold and less than the fifth preset gray threshold, the pixel point is regarded as a third target pixel point. Subsequently, the bright area in the merged surface image is determined according to all the third target pixel points, and the remaining area in the merged surface image is regarded as the dark area. In the case where the preset light source corresponds to multiple surface images, the embodiments of the present application perform merging processing on the multiple surface images according to the maximum gray value of the pixel points at the same position, which can improve the accuracy of the extraction of the bright area of the surface image under the preset light source, and further improve the detection accuracy of the microtubule defects.

[0056] In an embodiment of the present application, optionally, the step 104 of "verifying each suspected micropipe defect region according to the bright area and the dark area of ​​the surface image corresponding to each preset light source" includes: for each suspected micropipe defect region included in the preprocessing image, determining a target light source from each preset light source, and registering the surface image corresponding to the target light source with the preprocessing image, finding the intersection between the bright area of ​​the registered surface image and the suspected micropipe defect region to obtain a first intersection area, and finding the intersection between the dark area of ​​the registered surface image and the suspected micropipe defect region to obtain a second intersection area; calculating the ratio between the first intersection area and the suspected micropipe defect region to obtain a first ratio, and When the first ratio is greater than a preset threshold, the number of bright times corresponding to the target light source is counted up by one, and the ratio between the second intersection area and the suspected micropipe defect area is calculated to obtain a second ratio. When the second ratio is greater than the preset threshold, the number of dark times corresponding to the target light source is counted up by one; the target light source is eliminated from the preset light sources, and the process returns to the step of determining the target light source from the preset light sources until all the preset light sources are eliminated. The bright time count results corresponding to the preset light sources are summed, and the dark time count results are summed. When the sum of the bright times is greater than 0 and the sum of the dark times is greater than 1, it is determined that the suspected micropipe defect area is a real micropipe defect area.

[0057] In this embodiment, the bright and dark areas of the surface images under each preset light source can be used to verify each suspected micropipe defect area in the preprocessed image to determine whether the suspected micropipe defect area is a true micropipe defect area. First, any one of the preset light sources can be selected as a target light source. Then, the surface image corresponding to the target light source is registered with the preprocessed image so that the surface image corresponding to the target light source and the wafer area in the preprocessed image spatially overlap. Here, when there is only one surface image of the target light source, this surface image can be directly used for registration with the preprocessed image; when there are multiple surface images of the target light source, a surface image obtained by merging these surface images can be used for registration with the preprocessed image.

[0058] After registration, for each suspected micropipe defect region in the preprocessed image, the intersection between the suspected micropipe defect region and the bright region of the surface image can be calculated to obtain a first intersection region. Simultaneously, the intersection between the suspected micropipe defect region and the dark region of the surface image can be calculated to obtain a second intersection region. Next, the ratio between the first intersection region and the suspected micropipe defect region is calculated to obtain a first ratio. If the first ratio is greater than a preset threshold, it indicates that the overlap between the suspected micropipe defect region and the bright region in the surface image corresponding to the target light source meets the requirement, meaning that the suspected micropipe defect region is bright in the surface image corresponding to the target light source. In this case, the number of bright times corresponding to the target light source can be incremented by 1, thereby increasing the number of bright times from 0 to 1. Similarly, the ratio between the second intersection area and the suspected micropipe defect area is calculated to obtain a second ratio. If the second ratio is greater than the preset threshold, it indicates that the suspected micropipe defect area overlaps with the dark area in the surface image corresponding to the target light source, meaning that the suspected micropipe defect area is dark in the surface image corresponding to the target light source. The dark count corresponding to the target light source is then incremented by 1, changing the dark count from 0 to 1. At this point, verification of the target light source is complete, and the target light source can be removed from the preset light sources. Next, another target light source is selected from the remaining preset light sources as the target light source, and the above steps are repeated until all preset light sources have been processed, obtaining the bright count and dark count corresponding to each preset light source. Next, the bright counts for all preset light sources are summed to obtain a bright count sum; simultaneously, the dark counts for all preset light sources are summed to obtain a dark count sum. If the bright count sum is greater than 0 and the dark count sum is greater than 1, the suspected micropipe defect area is determined to be a true micropipe defect area.

[0059] For example, for suspected micropipe defect region 1 in the preprocessed image, a first intersection area between suspected micropipe defect region 1 and the bright area of ​​the target light source can be calculated. A first ratio between the first intersection area and suspected micropipe defect region 1 is then calculated. If the first ratio is found to be greater than a preset threshold (assuming 0.7), the bright count corresponding to the target light source is incremented by 1. Simultaneously, a second intersection area between suspected micropipe defect region 1 and the dark area of ​​the target light source is calculated. A second ratio between the second intersection area and suspected micropipe defect region 1 is then calculated. If the second ratio is found to be less than the preset threshold (0.7), the dark count corresponding to the target light source remains zero. Then use the same method to calculate the bright and dark counts of the suspected microtubule defect area 1 under the remaining preset light sources: assuming there are 4 preset light sources, the dark count under the remaining two preset light sources is 1, and the bright count is 0; the bright and dark counts under the remaining one preset light source are both 0, then the total bright count of the suspected microtubule defect area under the four preset light sources is 1, and the total dark count is 2, which means that the suspected microtubule defect area 1 is a real microtubule defect area. The above method can be used to verify each suspected microtubule defect area. The embodiment of the present application is as follows Figure 2 When setting the preset light sources in a certain manner, the physical property of micropipe defects penetrating the wafer is utilized. When imaging under four preset light sources, the micropipe defects will inevitably appear bright under at least one preset light source and dark under two preset light sources. Based on this feature, the real micropipe defect area can be quickly identified from multiple suspected micropipe defect areas.

[0060] In an embodiment of the present application, optionally, after the "verification of each of the suspected microtubule defect regions" in step 104, the method further includes: when the suspected microtubule defect region is a real microtubule defect region, marking the suspected microtubule defect region with a first marker in the pre-processing image; otherwise, marking the suspected microtubule defect region with a second marker.

[0061] In this embodiment, once the suspected micropipe defect area is confirmed to be a real micropipe defect area, it can be marked in the pre-processing image using a first mark so that subsequent processing or analysis can quickly identify these real micropipe defect areas. Specifically, the first mark can be a color mark, a border mark, etc. For those suspected micropipe defect areas that are not real micropipe defect areas, a second mark can be used for marking processing to prompt relevant staff to perform subsequent evaluation and analysis of the real defects corresponding to these suspected micropipe defect areas. The embodiment of the present application uses different marks to mark the real micropipe defect areas and the remaining suspected micropipe defect areas in the pre-processing image, which not only helps to quickly identify the real micropipe defect areas and the remaining suspected micropipe defect areas, but also helps to accurately and efficiently perform subsequent processing steps (such as defect analysis, repair suggestions, report generation, etc.).

[0062] In an embodiment of the present application, optionally, before step 101, the method further includes: for each of the preset light sources, controlling the preset light source to turn on, and controlling the image acquisition device to acquire a surface image of the wafer, inputting the surface image of the wafer into a control parameter calculation model, obtaining image acquisition control parameters of the image acquisition device, and recording the image acquisition control parameters corresponding to the preset light source; accordingly, the method further includes: in response to a detection instruction for wafer micropipe defects, turning on one preset light source at a time in a preset order, adjusting the preset parameters of the image acquisition device based on the image acquisition control parameters corresponding to the turned-on preset light source, and controlling the image acquisition device to acquire the surface image corresponding to the wafer after the adjustment.

[0063] In this embodiment, for each preset light source, after a certain preset light source is turned on separately, the image acquisition device can be controlled to acquire the surface image of the wafer. The acquired wafer surface image is then input into a control parameter calculation model. The model can be obtained based on machine learning or deep learning training, and the optimal image acquisition control parameters are calculated by analyzing the information in the surface image (such as brightness, contrast, clarity, etc.). These image acquisition control parameters may include the exposure time, gain, white balance, etc. of the camera, which together determine the quality and characteristics of the acquired image. After calculating the optimal image acquisition control parameters, the image acquisition control parameters under the preset light source can be recorded. In this way, in the subsequent wafer micropipe defect detection process, the optimal image acquisition control parameters can be quickly selected according to the light source number, thereby improving the detection efficiency.

[0064] After recording the optimal image acquisition control parameters corresponding to each preset light source, when a detection instruction for wafer micropipe defects is received, one preset light source can be turned on each time in a preset order. For the currently turned-on preset light source, the preset parameters of the image acquisition device can be adjusted based on the previously recorded image acquisition control parameters, thereby ensuring that the image quality captured under the lighting conditions of the current preset light source is optimal. After adjusting the preset parameters of the image acquisition device, the image acquisition device is controlled to capture the surface image of the wafer. The embodiment of the present application automatically adjusts the image acquisition control parameters of the image acquisition device according to different preset light source activation environments, thereby ensuring that the captured image quality is optimal, thereby improving the subsequent detection accuracy and efficiency of wafer micropipe defects.

[0065] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides a device for detecting micro-pipe defects in wafers, such as Figure 9 As shown, the device includes:

[0066] a surface image acquisition module, configured to acquire multiple surface images corresponding to a wafer to be inspected for micropipe defects, taken from the same acquisition angle, wherein the multiple surface images include surface images acquired after each preset light source is individually activated, and the position of each preset light source relative to the wafer is determined based on the physical characteristics of the micropipe defect;

[0067] a pre-processing image generating module, configured to respectively determine the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and determine a pre-processing image of the wafer area under the acquisition viewing angle according to the maximum grayscale value of the pixel at each position;

[0068] a preliminary defect detection module, configured to perform preliminary micropipe defect detection on the preprocessed image to obtain a suspected micropipe defect region included in the preprocessed image;

[0069] The defect verification module is used to determine the bright areas and dark areas on the surface image corresponding to each preset light source, and verify each suspected micropipe defect area based on the bright areas and dark areas of the surface image corresponding to each preset light source to determine whether the suspected micropipe defect area is a real micropipe defect area.

[0070] Optionally, the image acquisition device setting surface and the preset light source setting surface are distributed on both sides or the same side of the wafer, the preset light source is set on the preset light source setting surface, the distance between each preset light source and the projection point of the center of the wafer on the preset light source setting surface is the same, the angle between each two adjacent preset light sources and the projection point is 90°, the illumination direction of each preset light source is toward the wafer, and the angle formed between the illumination direction of each preset light source and the wafer surface is equal.

[0071] Optionally, the pre-processing graph generation module is used to:

[0072] Determining the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and obtaining an initial preprocessing image according to the maximum grayscale value of the pixel at each position;

[0073] Selecting a target surface image from the plurality of surface images, determining, for each pixel in the target surface image, a relationship between a grayscale value corresponding to the pixel in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold, and if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, taking the pixel as a first target pixel, and determining a wafer area corresponding to the target surface image based on the first target pixel, wherein the first preset grayscale threshold is less than the second preset grayscale threshold;

[0074] The initial pre-processed image is registered with the target surface image, and the initial pre-processed image is intercepted according to the wafer area in the target surface image to obtain a pre-processed image of the wafer area under the acquisition viewing angle.

[0075] Optionally, the pre-processing graph generation module is used to:

[0076] Selecting a target surface image from the plurality of surface images, determining, for each pixel in the target surface image, a relationship between a grayscale value corresponding to the pixel in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold, and if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, taking the pixel as a first target pixel, and determining a wafer area corresponding to the target surface image based on the first target pixel, wherein the first preset grayscale threshold is less than the second preset grayscale threshold;

[0077] According to the wafer area corresponding to the target surface image, each surface image is clipped to obtain a surface image containing only the wafer area, and based on the maximum grayscale value corresponding to the pixel points at the same position in multiple surface images containing only the wafer area, a preprocessing image of the wafer area under the acquisition perspective is obtained.

[0078] Optionally, the preliminary defect detection module is configured to:

[0079] Performing mean filtering on the preprocessed image to obtain a mean filtered image;

[0080] Subtracting the grayscale value of each pixel in the mean filtered image from the grayscale value of the corresponding pixel in the preprocessed image to obtain a grayscale difference corresponding to each pixel, and obtaining a difference image based on the grayscale difference corresponding to each pixel;

[0081] For each pixel point in the difference map, determine the relationship between the grayscale value corresponding to the pixel point in the difference map and the third preset grayscale threshold, and take the pixel point with a grayscale value greater than the third preset grayscale threshold as the second target pixel point, and take the area formed by the second target pixel point as the suspected micropipe defect area.

[0082] Optionally, when there is only one surface image corresponding to any preset light source, the defect verification module is configured to:

[0083] For each pixel point in the surface image corresponding to any one of the preset light sources, determine the relationship between the grayscale value of the pixel point in the surface image corresponding to any one of the preset light sources and the fourth preset grayscale threshold and the fifth preset grayscale threshold; if the grayscale value is greater than the fourth preset grayscale threshold and less than the fifth preset grayscale threshold, take the pixel point as the third target pixel point, determine the bright area of ​​the surface image corresponding to any one of the preset light sources based on the third target pixel point, and take the remaining area as the dark area, wherein the fourth preset grayscale threshold is less than the fifth preset grayscale threshold.

[0084] Optionally, when any preset light source corresponds to a plurality of surface images, the device further includes:

[0085] An image merging module is used to, for each pixel point in the surface image corresponding to any preset light source, determine the relationship between the grayscale value of the pixel point in the surface image corresponding to any preset light source and the fourth preset grayscale threshold and the fifth preset grayscale threshold, and then merge the multiple surface images corresponding to any preset light source according to the maximum grayscale value of the pixel points at the same position in the multiple surface images corresponding to any preset light source to replace the surface image.

[0086] Optionally, the defect verification module is used to:

[0087] For each suspected micropipe defect region included in the preprocessed image, determining a target light source from among the preset light sources, and registering the surface image corresponding to the target light source with the preprocessed image, intersecting a bright region of the registered surface image with the suspected micropipe defect region to obtain a first intersection region, and intersecting a dark region of the registered surface image with the suspected micropipe defect region to obtain a second intersection region;

[0088] Calculating a ratio between the first intersection area and the suspected micropipe defect area to obtain a first ratio, and when the first ratio is greater than a preset threshold, increasing the number of bright times corresponding to the target light source by one; and calculating a ratio between the second intersection area and the suspected micropipe defect area to obtain a second ratio, and when the second ratio is greater than the preset threshold, increasing the number of dark times corresponding to the target light source by one;

[0089] The target light source is removed from the preset light sources, and the process returns to the step of determining the target light source from the preset light sources until all the preset light sources are removed. The bright number counting results corresponding to each preset light source are summed, and the dark number counting results are summed. When the sum of the bright number counts is greater than 0 and the sum of the dark number counts is greater than 1, it is determined that the suspected micropipe defect area is a real micropipe defect area.

[0090] Optionally, the device further comprises:

[0091] a marking module for marking each suspected microtubule defect region with a first marker in the preprocessing image after verifying the suspected microtubule defect region and, if the suspected microtubule defect region is a true microtubule defect region, marking the suspected microtubule defect region with a second marker otherwise.

[0092] Optionally, the device further comprises:

[0093] a control parameter calculation module configured to, before acquiring a plurality of surface images corresponding to the same acquisition angle of view of a wafer to be inspected for micropipe defects, control each of the preset light sources to be turned on, control an image acquisition device to acquire a surface image of the wafer, input the surface image of the wafer into a control parameter calculation model, obtain image acquisition control parameters of the image acquisition device, and record the image acquisition control parameters corresponding to the preset light sources;

[0094] Accordingly, the device further comprises:

[0095] The control parameter adjustment module is used to respond to the detection instruction of wafer micropipe defects, turn on one preset light source at a time in a preset order, adjust the preset parameters of the image acquisition device based on the image acquisition control parameters corresponding to the turned-on preset light sources, and control the image acquisition device to capture the surface image corresponding to the wafer after the adjustment.

[0096] It should be noted that for other corresponding descriptions of the functional units involved in the device for detecting micropipe defects in wafers provided in the embodiment of the present application, reference can be made to Figures 1 to 8 The corresponding description in the method will not be repeated here.

[0097] The present application also provides a computer device, which can be a personal computer, a server, a network device, etc. Figure 10 As shown, the computer device includes a bus, a processor, a memory, and a communication interface, and may also include an input / output interface and a display device. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps of each method embodiment are implemented.

[0098] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0099] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program thereon. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0100] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0101] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0102] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0103] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for detecting micropipe defects in wafers, characterized in that: include: Acquire multiple surface images corresponding to a wafer to be inspected for micropipe defects at the same acquisition angle, wherein the multiple surface images include surface images acquired after each preset light source is individually activated, and the position of each preset light source relative to the wafer is determined based on the physical characteristics of the micropipe defect; Determining the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and determining a pre-processed image of the wafer area under the acquisition viewing angle according to the maximum grayscale value of the pixel at each position; performing preliminary microtubule defect detection on the preprocessed image to obtain a suspected microtubule defect region included in the preprocessed image; For each surface image corresponding to the preset light source, determining a bright area and a dark area on the surface image, and verifying each suspected micropipe defect area based on the bright area and the dark area of ​​the surface image corresponding to each preset light source to determine whether the suspected micropipe defect area is a real micropipe defect area; The verifying each suspected micropipe defect area according to the bright area and the dark area of ​​the surface image corresponding to each preset light source includes: For each suspected micropipe defect region included in the preprocessed image, determining a target light source from among the preset light sources, and registering the surface image corresponding to the target light source with the preprocessed image, intersecting a bright region of the registered surface image with the suspected micropipe defect region to obtain a first intersection region, and intersecting a dark region of the registered surface image with the suspected micropipe defect region to obtain a second intersection region; Calculating a ratio between the first intersection area and the suspected micropipe defect area to obtain a first ratio, and when the first ratio is greater than a preset threshold, increasing the number of bright times corresponding to the target light source by one; and calculating a ratio between the second intersection area and the suspected micropipe defect area to obtain a second ratio, and when the second ratio is greater than the preset threshold, increasing the number of dark times corresponding to the target light source by one; The target light source is removed from the preset light sources, and the process returns to the step of determining the target light source from the preset light sources until all the preset light sources are removed. The bright number counting results corresponding to each preset light source are summed, and the dark number counting results are summed. When the sum of the bright number counts is greater than 0 and the sum of the dark number counts is greater than 1, it is determined that the suspected micropipe defect area is a real micropipe defect area.

2. The method according to claim 1, characterized in that The image acquisition device setting surface and the preset light source setting surface are distributed on both sides or the same side of the wafer. The preset light sources are set on the preset light source setting surface. The distances between each preset light source and the projection point of the center of the wafer on the preset light source setting surface are the same. The angle between the line connecting each two adjacent preset light sources and the projection point is 90°. The illumination direction of each preset light source is toward the wafer, and the angle formed between the illumination direction of each preset light source and the wafer surface is equal.

3. The method according to claim 2, characterized in that The step of respectively determining the maximum grayscale values ​​corresponding to the pixels at the same position in the plurality of surface images, and determining the pre-processed image of the wafer area under the acquisition viewing angle according to the maximum grayscale values ​​of the pixels at each position, includes: Determining the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and obtaining an initial preprocessing image according to the maximum grayscale value of the pixel at each position; Selecting a target surface image from the plurality of surface images, determining, for each pixel in the target surface image, a relationship between a grayscale value corresponding to the pixel in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold, and if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, taking the pixel as a first target pixel, and determining a wafer area corresponding to the target surface image based on the first target pixel, wherein the first preset grayscale threshold is less than the second preset grayscale threshold; The initial pre-processed image is registered with the target surface image, and the initial pre-processed image is intercepted according to the wafer area in the target surface image to obtain a pre-processed image of the wafer area under the acquisition viewing angle.

4. The method according to claim 2, characterized in that The step of respectively determining the maximum grayscale values ​​corresponding to the pixels at the same position in the plurality of surface images, and determining the pre-processed image of the wafer area under the acquisition viewing angle according to the maximum grayscale values ​​of the pixels at each position, includes: Selecting a target surface image from the plurality of surface images, determining, for each pixel in the target surface image, a relationship between a grayscale value corresponding to the pixel in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold, and if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, taking the pixel as a first target pixel, and determining a wafer area corresponding to the target surface image based on the first target pixel, wherein the first preset grayscale threshold is less than the second preset grayscale threshold; According to the wafer area corresponding to the target surface image, each surface image is clipped to obtain a surface image containing only the wafer area, and based on the maximum grayscale value corresponding to the pixel points at the same position in multiple surface images containing only the wafer area, a preprocessing image of the wafer area under the acquisition perspective is obtained.

5. The method according to claim 3 or 4, characterized in that The performing preliminary micropipe defect detection on the preprocessed image to obtain a suspected micropipe defect region in the preprocessed image includes: Performing mean filtering on the preprocessed image to obtain a mean filtered image; Subtracting the grayscale value of each pixel in the mean filtered image from the grayscale value of the corresponding pixel in the preprocessed image to obtain a grayscale difference corresponding to each pixel, and obtaining a difference image based on the grayscale difference corresponding to each pixel; For each pixel point in the difference map, determine the relationship between the grayscale value corresponding to the pixel point in the difference map and the third preset grayscale threshold, and take the pixel point with a grayscale value greater than the third preset grayscale threshold as the second target pixel point, and take the area formed by the second target pixel point as the suspected micropipe defect area.

6. The method according to claim 2, characterized in that When there is only one surface image corresponding to any preset light source, determining the bright area and the dark area on the surface image for each surface image corresponding to the preset light source includes: For each pixel point in the surface image corresponding to any one of the preset light sources, determine the relationship between the grayscale value of the pixel point in the surface image corresponding to any one of the preset light sources and the fourth preset grayscale threshold and the fifth preset grayscale threshold; if the grayscale value is greater than the fourth preset grayscale threshold and less than the fifth preset grayscale threshold, take the pixel point as the third target pixel point, determine the bright area of ​​the surface image corresponding to any one of the preset light sources based on the third target pixel point, and take the remaining area as the dark area, wherein the fourth preset grayscale threshold is less than the fifth preset grayscale threshold.

7. The method according to claim 6, characterized in that When the surface images corresponding to any one of the preset light sources include multiple images, before determining, for each pixel point in the surface image corresponding to the any one of the preset light sources, a relationship between the grayscale value of the pixel point in the surface image corresponding to the any one of the preset light sources and the fourth preset grayscale threshold and the fifth preset grayscale threshold, the method further includes: The multiple surface images corresponding to any one of the preset light sources are merged according to the maximum grayscale value of the pixel points at the same position in the multiple surface images corresponding to any one of the preset light sources to replace the surface image.

8. The method according to claim 1, characterized in that After verifying each suspected microtubule defect region, the method further includes: When the suspected microtubule defect region is a real microtubule defect region, the suspected microtubule defect region is marked using a first marker in the pre-processing image; otherwise, the suspected microtubule defect region is marked using a second marker.

9. The method according to claim 1, characterized in that Before acquiring a plurality of surface images corresponding to the wafer to be inspected for micropipe defects under the same acquisition viewing angle, the method further includes: For each of the preset light sources, controlling the preset light source to turn on, and controlling an image acquisition device to acquire a surface image of the wafer, inputting the surface image of the wafer into a control parameter calculation model to obtain image acquisition control parameters of the image acquisition device, and recording the image acquisition control parameters corresponding to the preset light source; Accordingly, the method further comprises: In response to a detection instruction for wafer micropipe defects, one preset light source is turned on each time in a preset order, and the preset parameters of the image acquisition device are adjusted based on the image acquisition control parameters corresponding to the turned-on preset light sources. After the adjustment, the image acquisition device is controlled to acquire the surface image corresponding to the wafer.

10. A device for detecting micropipe defects in wafers, characterized in that: include: a surface image acquisition module, configured to acquire multiple surface images corresponding to a wafer to be inspected for micropipe defects, taken from the same acquisition angle, wherein the multiple surface images include surface images acquired after each preset light source is individually activated, and the position of each preset light source relative to the wafer is determined based on the physical characteristics of the micropipe defect; a pre-processing image generating module, configured to respectively determine the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and determine a pre-processing image of the wafer area under the acquisition viewing angle according to the maximum grayscale value of the pixel at each position; a preliminary defect detection module, configured to perform preliminary micropipe defect detection on the preprocessed image to obtain a suspected micropipe defect region included in the preprocessed image; a defect verification module, configured to determine, for each surface image corresponding to a preset light source, a bright area and a dark area on the surface image, and verify each suspected micropipe defect area based on the bright area and dark area of ​​the surface image corresponding to each preset light source to determine whether the suspected micropipe defect area is a real micropipe defect area; The defect verification module is used to: For each suspected micropipe defect region included in the preprocessed image, determining a target light source from among the preset light sources, and registering the surface image corresponding to the target light source with the preprocessed image, intersecting a bright region of the registered surface image with the suspected micropipe defect region to obtain a first intersection region, and intersecting a dark region of the registered surface image with the suspected micropipe defect region to obtain a second intersection region; Calculating a ratio between the first intersection area and the suspected micropipe defect area to obtain a first ratio, and when the first ratio is greater than a preset threshold, increasing the number of bright times corresponding to the target light source by one; and calculating a ratio between the second intersection area and the suspected micropipe defect area to obtain a second ratio, and when the second ratio is greater than the preset threshold, increasing the number of dark times corresponding to the target light source by one; The target light source is removed from the preset light sources, and the process returns to the step of determining the target light source from the preset light sources until all the preset light sources are removed. The bright number counting results corresponding to each preset light source are summed, and the dark number counting results are summed. When the sum of the bright number counts is greater than 0 and the sum of the dark number counts is greater than 1, it is determined that the suspected micropipe defect area is a real micropipe defect area.

11. The device according to claim 10, characterized in that The image acquisition device setting surface and the preset light source setting surface are distributed on both sides or the same side of the wafer. The preset light sources are set on the preset light source setting surface. The distances between each preset light source and the projection point of the center of the wafer on the preset light source setting surface are the same. The angle between the line connecting each two adjacent preset light sources and the projection point is 90°. The illumination direction of each preset light source is toward the wafer, and the angle formed between the illumination direction of each preset light source and the wafer surface is equal.

12. The device according to claim 11, characterized in that The pre-processing graph generation module is used to: Determining the maximum grayscale value corresponding to each pixel at the same position in the plurality of surface images, and obtaining an initial preprocessing image according to the maximum grayscale value of the pixel at each position; Selecting a target surface image from the plurality of surface images, determining, for each pixel in the target surface image, a relationship between a grayscale value corresponding to the pixel in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold, and if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, taking the pixel as a first target pixel, and determining a wafer area corresponding to the target surface image based on the first target pixel, wherein the first preset grayscale threshold is less than the second preset grayscale threshold; The initial pre-processed image is registered with the target surface image, and the initial pre-processed image is intercepted according to the wafer area in the target surface image to obtain a pre-processed image of the wafer area under the acquisition viewing angle.

13. The device according to claim 11, characterized in that The pre-processing graph generation module is used to: Selecting a target surface image from the plurality of surface images, determining, for each pixel in the target surface image, a relationship between a grayscale value corresponding to the pixel in the target surface image and a first preset grayscale threshold and a second preset grayscale threshold, and if the grayscale value is greater than the first preset grayscale threshold and less than the second preset grayscale threshold, taking the pixel as a first target pixel, and determining a wafer area corresponding to the target surface image based on the first target pixel, wherein the first preset grayscale threshold is less than the second preset grayscale threshold; According to the wafer area corresponding to the target surface image, each surface image is clipped to obtain a surface image containing only the wafer area, and based on the maximum grayscale value corresponding to the pixel points at the same position in multiple surface images containing only the wafer area, a preprocessing image of the wafer area under the acquisition perspective is obtained.

14. The device according to claim 12 or 13, characterized in that The preliminary defect detection module is used to: Performing mean filtering on the preprocessed image to obtain a mean filtered image; Subtracting the grayscale value of each pixel in the mean filtered image from the grayscale value of the corresponding pixel in the preprocessed image to obtain a grayscale difference corresponding to each pixel, and obtaining a difference image based on the grayscale difference corresponding to each pixel; For each pixel point in the difference map, determine the relationship between the grayscale value corresponding to the pixel point in the difference map and the third preset grayscale threshold, and take the pixel point with a grayscale value greater than the third preset grayscale threshold as the second target pixel point, and take the area formed by the second target pixel point as the suspected micropipe defect area.

15. The device according to claim 11, characterized in that When there is only one surface image corresponding to any preset light source, the defect verification module is configured to: For each pixel point in the surface image corresponding to any one of the preset light sources, determine the relationship between the grayscale value of the pixel point in the surface image corresponding to any one of the preset light sources and the fourth preset grayscale threshold and the fifth preset grayscale threshold; if the grayscale value is greater than the fourth preset grayscale threshold and less than the fifth preset grayscale threshold, take the pixel point as the third target pixel point, determine the bright area of ​​the surface image corresponding to any one of the preset light sources based on the third target pixel point, and take the remaining area as the dark area, wherein the fourth preset grayscale threshold is less than the fifth preset grayscale threshold.

16. The device according to claim 15, characterized in that When any preset light source corresponds to a plurality of surface images, the device further includes: An image merging module is used to, for each pixel point in the surface image corresponding to any preset light source, determine the relationship between the grayscale value of the pixel point in the surface image corresponding to any preset light source and the fourth preset grayscale threshold and the fifth preset grayscale threshold, and then merge the multiple surface images corresponding to any preset light source according to the maximum grayscale value of the pixel points at the same position in the multiple surface images corresponding to any preset light source to replace the surface image.

17. The device according to claim 10, characterized in that The device further comprises: a marking module for marking each suspected microtubule defect region with a first marker in the preprocessing image after verifying the suspected microtubule defect region and, if the suspected microtubule defect region is a true microtubule defect region, marking the suspected microtubule defect region with a second marker otherwise.

18. The device according to claim 10, characterized in that The device further comprises: a control parameter calculation module configured to, before acquiring a plurality of surface images corresponding to the same acquisition angle of view of a wafer to be inspected for micropipe defects, control each of the preset light sources to be turned on, control an image acquisition device to acquire a surface image of the wafer, input the surface image of the wafer into a control parameter calculation model, obtain image acquisition control parameters of the image acquisition device, and record the image acquisition control parameters corresponding to the preset light sources; Accordingly, the device further comprises: The control parameter adjustment module is used to respond to the detection instruction of wafer micropipe defects, turn on one preset light source at a time in a preset order, adjust the preset parameters of the image acquisition device based on the image acquisition control parameters corresponding to the turned-on preset light sources, and control the image acquisition device to capture the surface image corresponding to the wafer after the adjustment.

19. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

20. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 9 is implemented.

Images