Image enhancement method of silicon wafer image and related equipment

Abstract

The embodiment of the invention discloses an image enhancement method of a silicon wafer image, image processing equipment and a computer readable storage medium, and aims to provide the image enhancement method of the silicon wafer image, so as to improve the quality of the silicon wafer image, reduce the overall noise of the image and improve the definition of a target body. The method provided by the embodiment of the invention comprises the steps of obtaining an original silicon wafer image, inputting the original silicon wafer image into a pre-trained image enhancement model, and determining a target pixel point of which the original gray value is within a preset gray value range in the original silicon wafer image by the image enhancement model, and adjusting an original gray value of a target pixel point of the original silicon wafer image to a target gray value range to obtain an image-enhanced target silicon wafer image output by the image enhancement model, the target pixel point being a pixel point related to the silicon wafer information in the original silicon wafer image, and the target gray value range being included in a preset gray value range.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image enhancement of silicon wafer images, and more specifically, to an image enhancement method, an image processing device, and a computer-readable storage medium for silicon wafer images. Background Art

[0002] In the semiconductor manufacturing process, semiconductor measurement and inspection machines are widely used in the wafer testing stage. Specifically, the detection device can be used to detect the wafer surface, the computer equipment can process the detected image and extract the defects, and the processed image data can be stored in the computer equipment. The wafer defect detection image can be displayed in the computer equipment, and the distribution of defects on the wafer can be determined based on the detection image.

[0003] However, in semiconductor measurement machine imaging, there are often problems such as poor image quality, large overall image noise, and blurred targets. Therefore, in order to better detect wafer defects on the detected images and improve the accuracy of defect location and recognition, there is an urgent need for a method that can improve the quality of silicon wafer images, reduce overall image noise, and improve target clarity.

[0004] The invention provides an image enhancement method for silicon wafer images, which can improve the quality of silicon wafer images, reduce the overall image noise and improve the clarity of target objects. Summary of the invention

[0005] The embodiments of the present application provide an image enhancement method for silicon wafer images, an image processing device, and a computer-readable storage medium, which are used to provide an image enhancement method for silicon wafer images, so as to improve the quality of silicon wafer images, reduce the overall image noise, and improve the clarity of the target object.

[0006] In a first aspect, an embodiment of the present application provides an image enhancement method for a silicon wafer image, comprising:

[0007] Obtaining original silicon wafer images;

[0008] The original silicon wafer image is input into a pre-trained image enhancement model, and the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image, determines the target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range, adjusts the original grayscale value of the target pixel of the original silicon wafer image to within a target grayscale value range, and determines the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0009] In a second aspect, an embodiment of the present application provides an image processing device, including:

[0010] An acquisition unit, used for acquiring an original silicon wafer image;

[0011] An image enhancement processing unit is used to input the original silicon wafer image into a pre-trained image enhancement model, determine the original grayscale value corresponding to each pixel in the original silicon wafer image by the image enhancement model, determine the target pixel in the original silicon wafer image whose original grayscale value is within a preset grayscale value range, adjust the original grayscale value of the target pixel in the original silicon wafer image to within a target grayscale value range, and determine the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0012] In a third aspect, an embodiment of the present application provides an image processing device, including:

[0013] CPU, memory, input and output interfaces, wired or wireless network interfaces, and power supply;

[0014] The memory is a short-term storage memory or a persistent storage memory;

[0015] The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the aforementioned image enhancement method for silicon wafer images.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which includes instructions. When the instructions are executed on a computer, the computer executes the aforementioned image enhancement method for silicon wafer images.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the aforementioned image enhancement method for silicon wafer images.

[0018] It can be seen from the above technical scheme that the embodiments of the present application have the following advantages: the target pixel points whose original grayscale values ​​in the original silicon wafer image are within a preset grayscale value range can be determined by the image enhancement model, and the original grayscale values ​​of the target pixel points of the original silicon wafer image are adjusted to within the target grayscale value range to perform image enhancement on the silicon wafer image, wherein the target pixel points are the pixel points in the original silicon wafer image related to the silicon wafer information, and the target grayscale value range is included in the preset grayscale value range, thereby providing an image enhancement method for silicon wafer images, which can improve the quality of silicon wafer images, reduce the overall image noise and improve the clarity of the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of the architecture of an image enhancement system for silicon wafer images disclosed in an embodiment of the present application;

[0020] Figure 2 A schematic diagram of a flow chart of an image enhancement method for a silicon wafer image disclosed in an embodiment of the present application;

[0021] Figure 3 A schematic flow chart of another method for enhancing silicon wafer images disclosed in an embodiment of the present application;

[0022] Figure 4 A schematic diagram of the structure of an image processing device disclosed in an embodiment of the present application;

[0023] Figure 5 A schematic diagram of the structure of another image processing device disclosed in an embodiment of the present application;

[0024] Figure 6 This is a schematic diagram of the structure of another image processing device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application provide an image enhancement method for silicon wafer images, an image processing device, and a computer-readable storage medium, which are used to provide an image enhancement method for silicon wafer images, so as to improve the quality of silicon wafer images, reduce the overall image noise, and improve the clarity of the target object.

[0026] See also Figure 1 The architecture of the image enhancement system for silicon wafer images in the embodiment of the present application includes:

[0027] Image processing device 101 and semiconductor measurement and inspection machine 102. When performing image enhancement of a silicon wafer image, the image processing device 101 can be connected to the semiconductor measurement and inspection machine 102. The semiconductor measurement and inspection machine 102 can send the original silicon wafer image to the target image processing device 102, and the image processing device 101 can input the original silicon wafer image into a pre-trained image enhancement model to obtain the target silicon wafer image after image enhancement output by the image enhancement model.

[0028] based on Figure 1 The image enhancement system for the silicon wafer image shown is shown in Figure 2 , Figure 2 The following is a flow chart of an image enhancement method for a silicon wafer image disclosed in an embodiment of the present application, wherein the method comprises:

[0029] 201. Obtain an original silicon wafer image.

[0030] In this embodiment, when the image enhancement of the silicon wafer image is performed, the original silicon wafer image can be obtained.

[0031] 202. Input the original silicon wafer image into a pre-trained image enhancement model, and use the image enhancement model to determine the original grayscale value corresponding to each pixel in the original silicon wafer image, determine the target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range, adjust the original grayscale value of the target pixel in the original silicon wafer image to within the target grayscale value range, and determine the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to the silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0032] After obtaining the original silicon wafer image, the original silicon wafer image can be input into a pre-trained image enhancement model, and the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image, determines the target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range, adjusts the original grayscale value of the target pixel in the original silicon wafer image to within the target grayscale value range, and determines the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to the silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0033] Among them, the image enhancement model can be trained before the original silicon wafer image is input into the pre-trained image enhancement model. Specifically, the method for training the image enhancement model may be to first obtain an original silicon wafer image sample, wherein the original silicon wafer image sample is annotated with an image-enhanced silicon wafer image, and the image clarity of the annotated silicon wafer image reaches a preset clarity standard, and then input the original silicon wafer image sample into the image enhancement model, and the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image sample, determines the target pixel whose original grayscale value in the original silicon wafer image sample is within a preset grayscale value range, adjusts the original grayscale value of the target pixel of the original silicon wafer image sample to within a target grayscale value range, and determines the grayscale value of each pixel of the target silicon wafer image after image enhancement of the original silicon wafer image sample based on the adjusted grayscale value of the target pixel of the original silicon wafer image sample, so as to obtain the image-enhanced predicted silicon wafer image output by the image enhancement model, wherein the target pixel of the original silicon wafer image sample is a pixel related to the silicon wafer information in the original silicon wafer image sample, and the target grayscale value range is included in the preset grayscale value range, and finally calculates the loss between the predicted silicon wafer image and the annotated silicon wafer image according to the regression loss function, and when the loss meets the convergence condition, obtains the trained image enhancement model.

[0034] The method for adjusting the original grayscale value of the target pixel of the original silicon wafer image sample to within the target grayscale value range may be, for each target pixel, multiplying the original grayscale value of the target pixel of the original silicon wafer image sample by the preset cumulative percentage of the grayscale level corresponding to the target pixel of the original silicon wafer image sample, so as to adjust the original grayscale value of the target pixel of the original silicon wafer image sample to within the target grayscale value range. It is understandable that in addition to the above-described method for adjusting the original grayscale value of the target pixel of the original silicon wafer image sample to within the target grayscale value range, other reasonable methods may also be used, which are not specifically limited here.

[0035] In an embodiment of the present application, an image enhancement model can be used to determine target pixel points in the original silicon wafer image whose original grayscale values ​​are within a preset grayscale value range, and the original grayscale values ​​of the target pixel points in the original silicon wafer image are adjusted to within the target grayscale value range to perform image enhancement on the silicon wafer image, wherein the target pixel points are pixel points in the original silicon wafer image that are related to silicon wafer information, and the target grayscale value range is included in the preset grayscale value range. An image enhancement method for silicon wafer images is provided, which can improve the quality of silicon wafer images, reduce overall image noise, and improve target object clarity.

[0036] In the embodiment of the present application, before the original silicon wafer image is input into the pre-trained image enhancement model, the image enhancement model can be trained. Specifically, there are multiple methods for training the image enhancement model, based on Figure 2The image enhancement method of the silicon wafer image is shown, and one of the methods is described below.

[0037] In this embodiment, when the image enhancement of the silicon wafer image is performed, the original silicon wafer image can be obtained.

[0038] After obtaining the original silicon wafer image, the original silicon wafer image can be input into a pre-trained image enhancement model, and the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image, determines the target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range, adjusts the original grayscale value of the target pixel in the original silicon wafer image to within the target grayscale value range, and determines the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to the silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0039] Specifically, for example, the grayscale value range can be preset to be, for example, 0 to 100, and the target grayscale value range is 50 to 100. It should be understood that the grayscale value range is 0 to 255, where 0 represents black and 255 represents white. The grayscale value represents the brightness or grayscale intensity of the pixel. A smaller grayscale value represents a darker color, while a larger grayscale value represents a brighter color. Grayscale values ​​of 0 and 255 are extreme values, corresponding to the darkest and brightest colors, respectively.

[0040] It is worth mentioning that determining the target pixel points related to the silicon wafer information in the original silicon wafer image, where the silicon wafer information includes but is not limited to the edge information and the internal information of the silicon wafer, can improve the efficiency of image enhancement of the original silicon wafer image.

[0041] It is also worth mentioning that the target pixel points can be not only the pixel points related to the silicon wafer information in the original silicon wafer image, but also all the pixel points in the original silicon wafer image. The specific pixel points can be determined according to actual needs and are not limited here.

[0042] Among them, the image enhancement model can be trained before the original silicon wafer image is input into the pre-trained image enhancement model. Specifically, the method for training the image enhancement model may be to first obtain an original silicon wafer image sample, wherein the original silicon wafer image sample is annotated with an image-enhanced silicon wafer image, and the image clarity of the annotated silicon wafer image reaches a preset clarity standard, and then input the original silicon wafer image sample into the image enhancement model, and the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image sample, determines the target pixel whose original grayscale value in the original silicon wafer image sample is within a preset grayscale value range, adjusts the original grayscale value of the target pixel of the original silicon wafer image sample to within a target grayscale value range, and determines the grayscale value of each pixel of the target silicon wafer image after image enhancement of the original silicon wafer image sample based on the adjusted grayscale value of the target pixel of the original silicon wafer image sample, so as to obtain the image-enhanced predicted silicon wafer image output by the image enhancement model, wherein the target pixel of the original silicon wafer image sample is a pixel related to the silicon wafer information in the original silicon wafer image sample, and the target grayscale value range is included in the preset grayscale value range, and finally, calculates the loss between the predicted silicon wafer image and the annotated silicon wafer image according to the regression loss function, and when the loss meets the convergence condition, obtains the trained image enhancement model.

[0043] Among them, the method for adjusting the original grayscale value of the target pixel point of the original silicon wafer image sample to within the target grayscale value range can be, for each target pixel point, multiplying the original grayscale value of the target pixel point of the original silicon wafer image sample by a preset cumulative percentage of the grayscale level corresponding to the target pixel point of the original silicon wafer image sample to adjust the original grayscale value of the target pixel point of the original silicon wafer image sample to within the target grayscale value range.

[0044] It is worth mentioning that the original grayscale value of the target pixel of the original silicon wafer image sample is multiplied by the preset cumulative percentage of the grayscale level corresponding to the target pixel of the original silicon wafer image sample. First, the original grayscale value can be mapped to the target grayscale value range, so as to adjust the overall grayscale distribution of the image. For example, the pixels that are too dark or too bright can be adjusted to the appropriate grayscale range, making the image clearer and easier to analyze; second, the contrast of the image can be enhanced, and the lower grayscale value will be pushed to a darker range, and the higher grayscale value will be pushed to a brighter range, thereby enhancing the ability to distinguish different parts of the image; third, the original grayscale value can be adjusted to better highlight the details in the image. For example, in a color image, the outline and texture of the object can be highlighted by graying and adjusting the grayscale value, which helps to better analyze and identify the target (silicon wafer) to highlight the target (silicon wafer) or area of ​​interest (silicon wafer area) in the image, making it more eye-catching and obvious.

[0045] Among them, for each target pixel point, the original grayscale value of the target pixel point of the original silicon wafer image sample is multiplied by the preset cumulative percentage of the grayscale level corresponding to the target pixel point of the original silicon wafer image sample, so that the preset cumulative percentage corresponding to each grayscale level can be determined before the original grayscale value of the target pixel point of the original silicon wafer image sample is adjusted to within the target grayscale value range.

[0046] Specifically, the method for determining the preset cumulative percentage corresponding to each gray level may be to first determine the number of pixels corresponding to each gray level in the original silicon wafer image sample, and the total number of pixels corresponding to all gray levels in the original silicon wafer image sample, and then, for each gray level, take the ratio of the number of pixels corresponding to the gray level to the total number of pixels corresponding to all gray levels as the preset cumulative percentage corresponding to the gray level.

[0047] It is worth mentioning that the preset cumulative percentage can be used to control the brightness distribution of the image. The cumulative percentage can be determined based on the original silicon wafer image sample and can be adjusted according to actual needs. By adjusting the value of the cumulative percentage, the display effect of different gray levels of the image can be adjusted. Therefore, according to the specific needs of the image, the target area of ​​interest (silicon wafer area) can be enhanced while retaining the image details.

[0048] After obtaining the original silicon wafer image, the grayscale value of the pixel in the original silicon wafer image can be converted into a binary domain to obtain the grayscale value of the pixel in the original silicon wafer image after the converted representation. And the method of inputting the original silicon wafer image into the pre-trained image enhancement model can be to input the grayscale value of the pixel in the original silicon wafer image after the converted representation into the pre-trained image enhancement model.

[0049] It is worth mentioning that converting the grayscale value representation of the pixel points in the original silicon wafer image into a binary domain and performing image enhancement processing based on the grayscale value of the binary domain of the pixel points in the original silicon wafer image can provide higher storage efficiency, transmission efficiency and processing efficiency. This conversion can make image data easier to store, transmit and process, and provides more convenience and flexibility for subsequent operations. Therefore, compared with the spatial domain and frequency domain, the efficiency of image enhancement is improved.

[0050] After the grayscale value of the pixel in the original silicon wafer image is converted into a binary domain, the grayscale value of the pixel in the original silicon wafer image after the conversion can be saved to the memory, and the grayscale value of the pixel in the original silicon wafer image after the conversion can be locked by a byte lock. After the original grayscale value of the target pixel in the original silicon wafer image is adjusted to within the target grayscale value range, the byte lock can be closed, and the grayscale value of each pixel of the target silicon wafer image after image enhancement is determined based on the adjusted grayscale value of the target pixel in the original silicon wafer image to obtain the target silicon wafer image after image enhancement output by the image enhancement model.

[0051] It is worth mentioning that the use of byte locks can avoid memory leaks and illegal operations, improving the security and accuracy of image enhancement.

[0052] Among them, the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image, determines the target pixel whose original grayscale value is within the preset grayscale value range in the original silicon wafer image, adjusts the original grayscale value of the target pixel of the original silicon wafer image to within the target grayscale value range, and determines the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image to obtain the image enhanced target silicon wafer image output by the image enhancement model. The method can be, performing a first task through a first thread, wherein the first task is a task corresponding to the original grayscale value corresponding to each pixel in the original silicon wafer image determined by the image enhancement model, and performing a second task through a second thread, wherein In the embodiment, the second task is a task corresponding to a target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range determined by the image enhancement model, and the third task is executed by the third thread, wherein the third task is a task corresponding to adjusting the original grayscale value of the target pixel of the original silicon wafer image to within the target grayscale value range by the image enhancement model, and the fourth task is executed by the fourth thread, wherein the fourth task is a task corresponding to determining the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image by the image enhancement model, so as to obtain the task corresponding to the target silicon wafer image after image enhancement output by the image enhancement model, and the first thread, the second thread, the third thread and the fourth thread are executed in parallel.

[0053] It is worth mentioning that image enhancement processing can be performed in a multi-threaded parallel manner, which improves the efficiency of image enhancement.

[0054] Specifically, to facilitate understanding of the embodiments of the present application, a specific example is given below. Figure 3 , Figure 3 FIG. 1 is a flow chart of another method for enhancing a silicon wafer image disclosed in an embodiment of the present application, comprising Figure 3It can be seen that the image signal to be enhanced (original silicon wafer image) can be obtained, the image signal to be enhanced can be transferred to the binary domain and a byte lock can be added, the image signal to be enhanced transferred to the binary domain can be input into the image enhancement algorithm, and various tasks of image enhancement processing are respectively performed in a multi-threaded parallel manner. The tasks include counting the number of pixel values ​​(i.e., the number of target pixels), calculating the original grayscale (i.e., calculating the original grayscale value corresponding to each target pixel), new pixel grayscale = original grayscale * cumulative percentage (i.e., for each target pixel, the product of the original grayscale value corresponding to the target pixel and the preset cumulative percentage of the grayscale level corresponding to the target pixel is used as the grayscale value of the target pixel of the target silicon wafer image after image enhancement), and transforming the pixel value of the new image according to the distribution of the original pixel value (i.e., determining the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image). After image enhancement processing, the byte lock can be closed and transferred to the imaging domain to obtain the enhanced target silicon wafer image.

[0055] It is worth mentioning that compared with image enhancement in other fields (such as vascular images), image enhancement of silicon wafer images focuses more on the enhancement of contrast, brightness and details (while vascular images focus more on the enhancement of vascular structure and texture features). Therefore, the embodiment of the present application determines the target pixel points whose original grayscale values ​​in the original silicon wafer image are within a preset grayscale value range (i.e., the target pixel points related to the silicon wafer information in the original silicon wafer image), and adjusts the original grayscale values ​​of the target pixel points of the original silicon wafer image to within the target grayscale value range. The target grayscale value range is included in the preset grayscale value range, which can pay more attention to the enhancement of contrast, brightness and details corresponding to the silicon wafer, thereby improving the image enhancement effect of the original silicon wafer image.

[0056] In this embodiment, the image enhancement model can be used to determine the target pixel points in the original silicon wafer image whose original grayscale values ​​are within a preset grayscale value range, and the original grayscale values ​​of the target pixel points in the original silicon wafer image are adjusted to within the target grayscale value range to perform image enhancement on the silicon wafer image, wherein the target pixel points are pixel points in the original silicon wafer image that are related to silicon wafer information, and the target grayscale value range is included in the preset grayscale value range. An image enhancement method for silicon wafer images is provided, which can improve the quality of silicon wafer images, reduce overall image noise, and improve target object clarity.

[0057] Secondly, the original grayscale value of the target pixel of the original silicon wafer image sample can be multiplied with the preset cumulative percentage of the grayscale level corresponding to the target pixel of the original silicon wafer image sample to adjust the overall grayscale distribution of the image, making the image clearer and easier to analyze. The contrast of the image can be enhanced, thereby enhancing the ability to distinguish different parts of the image. By adjusting the original grayscale value, the details in the image can be better highlighted to highlight the target (silicon wafer) or area of ​​interest (silicon wafer area) in the image, making it more eye-catching and obvious.

[0058] Next, the grayscale value of the pixel in the original silicon wafer image is converted into a binary domain, and image enhancement processing is performed based on the grayscale value of the binary domain of the pixel in the original silicon wafer image. This can efficiently, accurately and quickly enhance the image and highlight the target, and is not affected by noise. It can provide higher storage efficiency, transmission efficiency and processing efficiency. Compared with the spatial domain and frequency domain, it improves the efficiency of image enhancement and reduces the imaging time of semiconductor measurement equipment.

[0059] Furthermore, byte locks can be used to avoid memory leaks and illegal operations, improving the security and accuracy of image enhancement. Finally, image enhancement processing can be performed in a multi-threaded parallel manner, which can efficiently and quickly process multi-memory data in batches, reaching a speed of 75ms, almost real-time processing, and real-time mapping, thereby improving the efficiency of image enhancement.

[0060] The above describes the image enhancement method of the silicon wafer image in the embodiment of the present application. The following describes the image processing device in the embodiment of the present application. Figure 4 , an embodiment of the image processing device in the embodiment of the present application includes:

[0061] An obtaining unit 401 is used to obtain an original silicon wafer image;

[0062] The image enhancement processing unit 402 is used to input the original silicon wafer image into a pre-trained image enhancement model, determine the original grayscale value corresponding to each pixel in the original silicon wafer image by the image enhancement model, determine the target pixel in the original silicon wafer image whose original grayscale value is within a preset grayscale value range, adjust the original grayscale value of the target pixel in the original silicon wafer image to within a target grayscale value range, and determine the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0063] In an embodiment of the present application, an image enhancement model can be used to determine target pixel points in the original silicon wafer image whose original grayscale values ​​are within a preset grayscale value range, and the original grayscale values ​​of the target pixel points in the original silicon wafer image are adjusted to within the target grayscale value range to perform image enhancement on the silicon wafer image, wherein the target pixel points are pixel points in the original silicon wafer image that are related to silicon wafer information, and the target grayscale value range is included in the preset grayscale value range. An image enhancement method for silicon wafer images is provided, which can improve the quality of silicon wafer images, reduce overall image noise, and improve target object clarity.

[0064] The following is a detailed description of the image processing device in the embodiment of the present application. Figure 5 Another embodiment of the image processing device in the embodiment of the present application includes:

[0065] An obtaining unit 501 is used to obtain an original silicon wafer image;

[0066] The image enhancement processing unit 502 is used to input the original silicon wafer image into a pre-trained image enhancement model, determine the original grayscale value corresponding to each pixel in the original silicon wafer image by the image enhancement model, determine the target pixel in the original silicon wafer image whose original grayscale value is within a preset grayscale value range, adjust the original grayscale value of the target pixel in the original silicon wafer image to within a target grayscale value range, and determine the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

[0067] The image processing device further includes: a calculation unit 503;

[0068] The obtaining unit 501 is further used to obtain an original silicon wafer image sample; wherein the original silicon wafer image sample is annotated with an image of the silicon wafer after image enhancement; and the image clarity of the annotated silicon wafer image meets a preset clarity standard;

[0069] The image enhancement processing unit 502 is further used to input the original silicon wafer image sample into an image enhancement model, determine the original grayscale value corresponding to each pixel in the original silicon wafer image sample by the image enhancement model, determine the target pixel whose original grayscale value in the original silicon wafer image sample is within a preset grayscale value range, adjust the original grayscale value of the target pixel of the original silicon wafer image sample to within a target grayscale value range, and determine the grayscale value of each pixel of the target silicon wafer image after image enhancement of the original silicon wafer image sample based on the adjusted grayscale value of the target pixel of the original silicon wafer image sample, so as to obtain the predicted silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel of the original silicon wafer image sample is the pixel related to the silicon wafer information in the original silicon wafer image sample; and the target grayscale value range is included in the preset grayscale value range;

[0070] The calculation unit 503 is used to calculate the loss between the predicted silicon wafer image and the labeled silicon wafer image according to the regression loss function, and when the loss meets the convergence condition, a trained image enhancement model is obtained.

[0071] The image enhancement processing unit 502 is specifically used to multiply the original grayscale value of the target pixel point of the original silicon wafer image sample by a preset cumulative percentage of the grayscale level corresponding to the target pixel point of the original silicon wafer image sample for each target pixel point, so as to adjust the original grayscale value of the target pixel point of the original silicon wafer image sample to within the target grayscale value range.

[0072] The image processing device further comprises:

[0073] A determination unit 504 is used to determine the number of pixels corresponding to each gray level in the original silicon wafer image sample and the total number of pixels corresponding to all gray levels in the original silicon wafer image sample;

[0074] The determining unit 504 is further configured to, for each gray level, use a ratio of the number of pixels corresponding to the gray level to the total number of pixels corresponding to all gray levels as a preset cumulative percentage corresponding to the gray level.

[0075] The image processing device further comprises:

[0076] A conversion unit 505 is used to convert the grayscale value of the pixel point in the original silicon wafer image into a binary domain to obtain the grayscale value of the pixel point in the original silicon wafer image after the conversion;

[0077] The image enhancement processing unit 502 is specifically used to input the grayscale values ​​of the converted representation forms of the pixels in the original silicon wafer image into a pre-trained image enhancement model.

[0078] The image processing device further comprises:

[0079] A locking unit 506, used for saving the grayscale value of the converted pixel point in the original silicon wafer image to the memory, and locking the grayscale value of the converted pixel point in the original silicon wafer image through a byte lock;

[0080] The closing unit 507 is used to close the byte lock and execute the step of determining the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image to obtain the target silicon wafer image after image enhancement output by the image enhancement model.

[0081] The image enhancement processing unit 502 is specifically used to execute a first task through a first thread, wherein the first task is a task corresponding to the original grayscale value corresponding to each pixel in the original silicon wafer image determined by the image enhancement model; execute a second task through a second thread, wherein the second task is a task corresponding to a target pixel in the original silicon wafer image whose original grayscale value is within a preset grayscale value range determined by the image enhancement model; execute a third task through a third thread, wherein the third task is a task corresponding to adjusting the original grayscale value of the target pixel of the original silicon wafer image to within a target grayscale value range by the image enhancement model; execute a fourth task through a fourth thread, wherein the fourth task is a task corresponding to determining the grayscale value of each pixel of the target silicon wafer image after image enhancement by the image enhancement model based on the adjusted grayscale value of the target pixel of the original silicon wafer image, so as to obtain the image enhanced target silicon wafer image output by the image enhancement model; and the first thread, the second thread, the third thread and the fourth thread are executed in parallel.

[0082] In this embodiment, each unit in the image processing device performs the above Figure 2 The operation of the image processing device in the illustrated embodiment will not be described in detail herein.

[0083] See below Figure 6 Another embodiment of the image processing device 600 in the embodiment of the present application includes:

[0084] CPU 601, memory 605, input / output interface 604, wired or wireless network interface 603 and power supply 602;

[0085] The memory 605 is a temporary storage memory or a permanent storage memory;

[0086] The CPU 601 is configured to communicate with the memory 605 and execute the instructions in the memory 605 to perform the aforementioned Figure 2 The method in the illustrated embodiment.

[0087] The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium includes instructions, when the instructions are executed on a computer, the computer executes the aforementioned Figure 2 The method in the illustrated embodiment.

[0088] The present application also provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the aforementioned Figure 2 The method in the illustrated embodiment.

[0089] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0090] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0091] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0094] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A method for enhancing a silicon wafer image, characterized in that: include: Obtaining original silicon wafer images; The original silicon wafer image is input into a pre-trained image enhancement model, and the image enhancement model determines the original grayscale value corresponding to each pixel in the original silicon wafer image, determines the target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range, adjusts the original grayscale value of the target pixel of the original silicon wafer image to within a target grayscale value range, and determines the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

2. The method according to any one of claim 1, characterized in that Before inputting the original silicon wafer image into a pre-trained image enhancement model, the method further includes: Obtaining an original silicon wafer image sample; wherein the original silicon wafer image sample is annotated with an image of the silicon wafer after image enhancement; and the image clarity of the annotated silicon wafer image reaches a preset clarity standard; Input the original silicon wafer image sample into an image enhancement model, determine the original grayscale value corresponding to each pixel in the original silicon wafer image sample by the image enhancement model, determine the target pixel whose original grayscale value is within a preset grayscale value range in the original silicon wafer image sample, adjust the original grayscale value of the target pixel of the original silicon wafer image sample to within a target grayscale value range, and determine the grayscale value of each pixel of the target silicon wafer image after image enhancement of the original silicon wafer image sample based on the adjusted grayscale value of the target pixel of the original silicon wafer image sample, so as to obtain the predicted silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel of the original silicon wafer image sample is the pixel related to the silicon wafer information in the original silicon wafer image sample; and the target grayscale value range is included in the preset grayscale value range; The loss between the predicted silicon wafer image and the labeled silicon wafer image is calculated according to the regression loss function, and when the loss meets the convergence condition, a trained image enhancement model is obtained.

3. The method according to any one of claim 2, characterized in that The step of adjusting the original grayscale value of the target pixel point of the original silicon wafer image sample to a target grayscale value range includes: For each target pixel, the original grayscale value of the target pixel of the original silicon wafer image sample is multiplied by the preset cumulative percentage of the grayscale level corresponding to the target pixel of the original silicon wafer image sample to adjust the original grayscale value of the target pixel of the original silicon wafer image sample to within the target grayscale value range.

4. The method according to any one of claim 3, characterized in that Before the method multiplies the original grayscale value of the target pixel of the original silicon wafer image sample by the preset cumulative percentage of the grayscale level corresponding to the target pixel of the original silicon wafer image sample for each target pixel to adjust the original grayscale value of the target pixel of the original silicon wafer image sample to within the target grayscale value range, the method further includes: Determine the number of pixels corresponding to each gray level in the original silicon wafer image sample and the total number of pixels corresponding to all gray levels in the original silicon wafer image sample; For each gray level, the ratio of the number of pixel points corresponding to the gray level to the total number of pixel points corresponding to all gray levels is used as the preset cumulative percentage corresponding to the gray level.

5. The method according to claim 1, characterized in that After obtaining the original silicon wafer image, the method further comprises: Converting the grayscale value of the pixel in the original silicon wafer image into a binary domain to obtain the grayscale value of the pixel in the original silicon wafer image after the conversion; The step of inputting the original silicon wafer image into a pre-trained image enhancement model comprises: The grayscale values ​​of the converted representation of the pixels in the original silicon wafer image are input into a pre-trained image enhancement model.

6. The method according to claim 5, characterized in that After converting the grayscale value of the pixel in the original silicon wafer image into a binary domain, the method further includes: The grayscale values ​​of the converted pixel points in the original silicon wafer image are saved in the memory, and the grayscale values ​​of the converted pixel points in the original silicon wafer image are locked by byte lock; After adjusting the original grayscale value of the target pixel point of the original silicon wafer image to within the target grayscale value range, the method further includes: The byte lock is closed, and the step of determining the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image is executed to obtain the target silicon wafer image after image enhancement output by the image enhancement model.

7. The method according to claim 1, characterized in that The method of determining the original grayscale value corresponding to each pixel in the original silicon wafer image by the image enhancement model, determining the target pixel whose original grayscale value in the original silicon wafer image is within a preset grayscale value range, adjusting the original grayscale value of the target pixel in the original silicon wafer image to within a target grayscale value range, and determining the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image to obtain the target silicon wafer image after image enhancement output by the image enhancement model, includes: Execute a first task through a first thread; wherein the first task is a task corresponding to determining the original grayscale value corresponding to each pixel in the original silicon wafer image by the image enhancement model; Execute a second task through a second thread; wherein the second task is a task corresponding to a target pixel point whose original gray value in the original silicon wafer image is within a preset gray value range determined by the image enhancement model; Execute a third task through a third thread; wherein the third task is a task corresponding to adjusting the original grayscale value of the target pixel point of the original silicon wafer image to a target grayscale value range by the image enhancement model; The fourth task is executed through the fourth thread; wherein the fourth task is to determine the grayscale value of each pixel of the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel of the original silicon wafer image by the image enhancement model, so as to obtain the task corresponding to the target silicon wafer image after image enhancement output by the image enhancement model; and the first thread, the second thread, the third thread and the fourth thread are executed in parallel.

8. An image processing device, characterized in that: include: An acquisition unit, used for acquiring an original silicon wafer image; An image enhancement processing unit is used to input the original silicon wafer image into a pre-trained image enhancement model, determine the original grayscale value corresponding to each pixel in the original silicon wafer image by the image enhancement model, determine the target pixel in the original silicon wafer image whose original grayscale value is within a preset grayscale value range, adjust the original grayscale value of the target pixel in the original silicon wafer image to within a target grayscale value range, and determine the grayscale value of each pixel in the target silicon wafer image after image enhancement based on the adjusted grayscale value of the target pixel in the original silicon wafer image, so as to obtain the target silicon wafer image after image enhancement output by the image enhancement model; wherein the target pixel is a pixel in the original silicon wafer image related to silicon wafer information; and the target grayscale value range is included in the preset grayscale value range.

9. An image processing device, characterized in that: include: CPU and memory; The memory is a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium comprises instructions, and when the instructions are executed on a computer, the computer is caused to perform the method according to any one of claims 1 to 7.

Images