A method and device for generating a digital image of a curve mask and a computer device

Abstract

The present application relates to the field of photoetching technology, and particularly relates to a digital image generation method, device and computer equipment of a curve mask, and the present application provides a digital image generation method of a curve mask, comprising the following steps: extracting contour points based on pixel points of a preprocessed image and establishing a mathematical curve model of a Bezier curve based on the contour points; preprocessing the preprocessed image based on the mathematical curve model to obtain a digital signal image; and sequentially performing convolution processing and down-sampling processing on the digital signal image to generate a digital image. The present application also provides a device and computer equipment. The present application solves the problems of low efficiency and poor definition in the prior art.

Description

Technical Field

[0001] This invention relates to the field of computational lithography, and particularly to a method, apparatus, and computer device for generating digital images of curve masks. Background Technology

[0002] The physical design of integrated circuits takes performance, power consumption and area as considerations. The design of mask layout files that can be used for manufacturing and the hierarchical design of standard cells provide a good environment for electronic design automation.

[0003] Current curve mask layouts, including those based on optical and extreme ultraviolet lithography techniques, are primarily composed of minute features. Given the characteristics of curve contours, current standard cell image design systems mainly represent curve mask contour layouts in pixelated form. Existing techniques obtain the contour of the curve mask by fitting the pixelated pixels. During the fitting process, the large amount of data per pixel requires significant computational power, thus rendering the digital image takes a considerable amount of time. Furthermore, the poor sharpness of the fitted curve mask contour often results in poor sharpness in the final rendered digital image. Summary of the Invention

[0004] To address the problems of low efficiency and poor clarity in generating digital images in existing technologies, this invention provides a method, apparatus, and computer device for generating digital images using curve masks.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a digital image generation method using curve masks, comprising the following steps: Obtain the preprocessed image; Contour points are extracted from pixels in a preprocessed image, and a mathematical curve model of a Bézier curve is built based on these contour points. Digital signal images are obtained by preprocessing the preprocessed images based on mathematical curve models; The digital signal image is sequentially processed by convolution and downsampling to generate a digital image.

[0006] Preferably, obtaining the preprocessed image includes the following steps: The initial image on the mask is obtained through visual recognition. Pixelate the initial image to obtain the initial image pixels; A preset algorithm is used to reduce noise in the pixels to obtain a preprocessed image.

[0007] Preferably, the preset algorithm is Gaussian smoothing, and the formula for Gaussian smoothing is: ; Where x is the x-axis coordinate of the pixel and y is the y-axis coordinate of the pixel. This represents the confidence interval.

[0008] Preferably, preprocessing the preprocessed image based on a mathematical curve model to obtain a digital signal image includes the following steps: Identify all pixels in the preprocessed image; The preprocessed image is binarized based on a mathematical curve model to assign a first or second value to each pixel to generate a digital signal image.

[0009] Preferably, binarizing the preprocessed image based on a mathematical curve model includes the following steps: Define the mathematical curve model as the boundary and use the ray method to determine whether the pixel is inside the boundary; If a pixel is within the boundary, assign a first value to the pixel; if a pixel is outside the boundary, assign a second value to the pixel.

[0010] Preferably, the step of sequentially performing convolution processing on the digital signal image includes the following steps: processing the digital signal image using the SINC function; the SINC function is: ; Where x is the pixel coordinate value.

[0011] Preferably, contour points are extracted based on the pixels of the preprocessed image using a contour extraction method or a boundary tracking method.

[0012] To solve the above-mentioned technical problems, the present invention provides another technical solution as follows: an apparatus for generating digital images of the above-mentioned curve mask, the apparatus comprising: Recognition module: used to acquire pixels in the preprocessed image; Extraction module: used to extract contour points based on pixels in a preprocessed image; Calculation module: The calculation module includes a model building module and a calculation module; the model building module is used to establish a mathematical curve model of Bézier curves based on contour points; the calculation module is used to preprocess the preprocessed image to obtain a digital signal image.

[0013] Preferably, the device further includes: a digital processing module: used to sequentially perform convolution processing and downsampling processing on the digital signal image to generate a digital image.

[0014] To solve the above-mentioned technical problems, the present invention provides another technical solution as follows: a computer device applied to the above-mentioned digital image generation method of curve mask, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the above-mentioned computer program to implement the steps of the digital image generation method of curve mask.

[0015] Compared with the prior art, the digital image generation method, apparatus, and computer equipment for curve masks provided by the present invention have the following beneficial effects: 1. This invention provides a method for generating digital images using curve masks, comprising the following steps: acquiring a preprocessed image; extracting contour points based on the pixels of the preprocessed image and establishing a mathematical curve model of a Bézier curve based on the contour points; preprocessing the preprocessed image based on the mathematical curve model to obtain a digital signal image; and sequentially performing convolution and downsampling processing on the digital signal image to generate a digital image. Existing technologies obtain the contour of a curve mask by fitting pixelated pixels. During the fitting process, the large amount of pixel data requires significant computational power, and rendering takes a long time. Furthermore, the generated digital image is prone to poor clarity. This embodiment establishes a mathematical curve model using Bézier curves, preprocesses the preprocessed image based on the mathematical curve model to obtain a digital signal image, and then uses a filter for convolution. That is, the mathematical curve model makes the curve contour edges clear, and the convolution processing extracts the image edges, making the extracted local features clearer. Downsampling processing eliminates the numerical errors caused by numerical conversion after convolution processing, allowing the digital signal image to be rendered. In addition, downsampling processing speeds up rendering and improves rendering efficiency.

[0016] 2. The method of obtaining a preprocessed image in this embodiment of the invention includes the following steps: obtaining an initial image on a mask by visual recognition; pixelating the initial image to obtain initial image pixels; performing noise reduction processing on the pixels using a preset algorithm to obtain a preprocessed image, thereby removing the influence of noise on the subsequent determination of control points by performing noise reduction processing on the pixels using a preset algorithm.

[0017] 3. The embodiment of the present invention preprocesses a preprocessed image to obtain a digital signal image based on a mathematical curve model, including the following steps: identifying all pixels in the preprocessed image; and performing binarization processing on the preprocessed image based on the mathematical curve model to assign a first value or a second value to the pixels to generate a digital signal image. By binarizing the preprocessed image to convert it into a digital image, the diffraction limit caused by angular resolution is overcome, resulting in better image clarity in subsequent rendering.

[0018] 4. The embodiment of the present invention performs binarization processing on the preprocessed image based on the mathematical curve model, including the following steps: defining the mathematical curve model as the boundary and using the ray casting method to determine whether the pixel is within the boundary; if the pixel is within the boundary, assigning a first value to the pixel; if the pixel is outside the boundary, assigning a second value to the pixel. By assigning values ​​to points within and outside the mathematical curve model on the mask, the image is numerically represented as the mask outline.

[0019] 5. The present invention also provides an apparatus that has the same beneficial effects as the above-described method for analyzing the manufacturability of overlay marks, which will not be described in detail here.

[0020] 6. The apparatus of this embodiment further includes: a digital processing module: used to sequentially perform convolution processing and downsampling processing on the digital signal image to generate a digital image.

[0021] 7. The present invention also provides a computer device that has the same beneficial effects as the above-described method for analyzing the manufacturability of overlay marks, which will not be described in detail here. Attached Figure Description

[0022] Figure 1 This is a schematic flowchart of a digital image generation method using curve masks provided in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the process of obtaining a preprocessed image in a digital image generation method using curve masks provided in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram illustrating the process of extracting contour points from a preprocessed image in a digital image generation method using curve masks, as provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram illustrating the process of establishing a mathematical curve model of a Bézier curve in a digital image generation method for curve masks provided in an embodiment of the present invention.

[0026] Figure 5 This is a detailed flowchart illustrating the process of establishing a mathematical curve model of a Bézier curve in a digital image generation method using curve masks, as provided in an embodiment of the present invention.

[0027] Figure 6 This is a schematic diagram illustrating the process of obtaining a digital signal image using a curve mask digital image generation method provided in an embodiment of the present invention.

[0028] Figure 7 This is a schematic diagram of an SEM image representing a curve mask digital image generation method provided in an embodiment of the present invention. Figure 2 .

[0029] Figure 8 This is a schematic diagram illustrating the state changes when extracting M contour points from a SEM image using a curve mask digital image generation method provided in an embodiment of the present invention.

[0030] Figure 9 This is a schematic diagram of a digital image generated by convolution processing using a curve mask, as provided in an embodiment of the present invention.

[0031] Figure 10 This is a schematic diagram of a digital image generated by curve masking according to an embodiment of the present invention, after downsampling processing.

[0032] Figure 11 This is a schematic diagram of a device structure provided in an embodiment of the present invention.

[0033] Figure 12 This is a schematic diagram of a computer device structure provided in an embodiment of the present invention.

[0034] Explanation of reference numerals in the attached diagram: 1. Device; 2. Computer equipment; 11. Identification module; 12. Extraction module; 13. Calculation module; 14. Calibration module; 15. Digital processing module; 21. Memory; 22. Processor; 23. Computer program. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0036] In the embodiments provided by this invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0037] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Those skilled in the art should also recognize that the embodiments described in the specification are optional embodiments, and the actions and modules involved are not necessarily essential to the invention.

[0038] In various embodiments of the present invention, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0039] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It is particularly important to note that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0040] Please see Figure 1 This invention provides a method for generating digital images using curve masks, comprising the following steps: S1, Obtain the preprocessed image; S2, extract contour points based on the pixels of the preprocessed image and establish a mathematical curve model of the Bézier curve based on the contour points. S3, based on the mathematical curve model, preprocesses the preprocessed image to obtain a digital signal image; S4 performs convolution and downsampling processes on the digital signal image sequentially to generate a digital image.

[0041] In this embodiment, contour points are first extracted from the preprocessed image using a preset method, and a mathematical curve contour model is established based on these contour points. It should be understood that existing SEM image contour extraction methods extract and save the coordinates of a series of discrete points on the image contour, resulting in a large amount of data and often failing to cover the entire contour line. Bézier curves, a mathematical curve, can be described by a finite number of control points, eliminating the need to extract the coordinates of all discrete points during contour extraction, thus greatly saving computational power and increasing extraction efficiency. Furthermore, compared to traditional techniques, Bézier curves can clearly depict the mask contour, significantly improving the clarity of subsequent digital signal image rendering into a digital image.

[0042] Furthermore, this embodiment preprocesses the preprocessed image based on a mathematical curve model, converting it into a digital signal image. Then, it sequentially performs convolution and downsampling on the digital signal image, rendering it to generate a digital image. It should be understood that existing technologies obtain the contour of a curve mask by fitting pixelated pixels. During the fitting process, the large amount of data per pixel requires significant computational power, and rendering is time-consuming. Additionally, the resulting digital image is prone to poor clarity. This embodiment preprocesses the curve mask image based on a mathematical curve model, converting it into a digital signal image. Then, convolution and downsampling are used to convert the digital signal image back into a digital image, resulting in a clearer contour of the curve mask on the image. This also allows the generated curve mask to exhibit more small and complex characteristics. It should be noted that this embodiment uses filters for convolution, extracting image edges through convolution processing, making the extracted local features clearer. The numerical conversion after convolution can introduce numerical errors, which are eliminated through downsampling, allowing the digital signal image to be rendered. Meanwhile, downsampling speeds up rendering and improves rendering efficiency.

[0043] In step S1 above, please combine Figure 1 and Figure 2 Obtaining a preprocessed image includes the following steps: S11, the initial image on the mask is obtained through visual recognition; S12, pixelate the initial image to obtain the initial image pixels; S13 uses a preset algorithm to perform noise reduction on the pixels to obtain a preprocessed image.

[0044] It should be understood that the initial image is a SEM (Scanning Electron Microscope) image, which is obtained by scanning photoresist using SEM technology. SEM images are an important technique in the semiconductor manufacturing field and can be used for processes such as measuring key dimensions of images. The initial image is identified by a preset recognition module. After identification, the recognition module analyzes the initial image and pixelates it to obtain the initial image pixels. Then, a preset algorithm is used to reduce noise in the pixels to obtain a pre-processed image. Compared to the initial image, the pre-processed image has less noise and a clearer image outline.

[0045] In step S13 above, the preset algorithm is Gaussian smoothing, and the formula for Gaussian smoothing is: ; Where x is the x-axis coordinate of a pixel in the initial image, and y is the y-axis coordinate of a pixel in the initial image. This represents the confidence interval.

[0046] It should be understood that this embodiment processes the image based on a Standard Unit Graphics Design System (GDS). A coordinate system is established based on GDS, and the initial image is substituted into this coordinate system. Specifically, after step S12, i.e., after pixelating the initial image, the coordinates of the obtained pixels in the coordinate system are (x, y). By substituting the obtained pixels into the Gaussian smoothing formula, noise is removed, resulting in an optimized SEM image.

[0047] Optionally, in step S2 above, contour points are extracted based on the pixels of the preprocessed image using either a contour extraction method or a boundary tracking method; preferably, the preset method is the contour extraction method. Specifically, please refer to... Figure 1 and Figure 3 Extracting contour points from pixels in a preprocessed image includes the following steps: S201, The preprocessed image is pixelated to obtain the original matrix; S202, after performing Gaussian filtering on the original matrix, the image matrix is ​​obtained and the gradient value of each pixel in the image matrix is ​​calculated; S203, obtain all pixels with gradient values ​​within a preset range, and find the peak point along the gradient direction among all selected pixels; S204, the peak points are the M discrete contour points extracted from the preprocessed image.

[0048] It should be noted that acquiring all pixels with gradient values ​​within the preset range refers to acquiring the pixel with the largest gradient value. This design enables rapid identification, thereby improving the speed and accuracy of subsequent contour point extraction.

[0049] Understandably, the preprocessed image is the optimized SEM image. This method for extracting contour points from the preprocessed image allows for the rapid and accurate extraction of M contour points based solely on the preprocessed image, even without a GDS map file. Furthermore, by first acquiring all pixels within a preset range and then searching for peak points among the selected pixels, the contour point extraction speed is increased, and the extracted contour points become more accurate, thus facilitating subsequent contour extraction.

[0050] For example, first randomly select a preprocessed image to be processed (such as...). Figure 7 As shown in the diagram, the preprocessed image is then pixelated, and the original matrix is ​​Gaussian filtered to obtain the image matrix. The gradient value of each pixel in the image matrix is ​​then calculated. Next, all pixels with gradient values ​​within a preset range are obtained, and peak points are found along the gradient direction among all selected pixels. It can be seen that the contour points obtained in this way are more accurate, which facilitates subsequent contour extraction.

[0051] Please combine Figure 1 and Figure 4 In step S2 above, establishing the mathematical curve model of the Bézier curve based on the contour points specifically includes the following steps: S21, extract M discrete contour points from the preprocessed image, where M is an integer; S22, randomly select N initial control points, and obtain an initial Bézier curve based on the N initial control points, where N is an integer and N≤M; S23, calibrate the N initial control points to gradually fit the initial Bézier curve with the M contour points to obtain the calibrated Bézier curve; S24, the calibrated Bézier curve is output as the contour information extracted from the preprocessed image to obtain a mathematical curve model.

[0052] It should be understood that, firstly, M discrete contour points are extracted from the preprocessed image using the aforementioned contour extraction method. Then, N initial control points are randomly selected to calculate the Bézier curve. Since the Bézier curve is a differentiable mathematical curve, it can describe a continuous and smooth curve using a finite number of initial control points. The N initial control points are then calibrated using a preset calibration method, thereby ensuring the Bézier curve fits the M contour points. In other words, by first identifying the contour points and then establishing a mathematical curve model of the Bézier curve using these points, the contour information of the preprocessed image can be represented better and more accurately. Furthermore, since this invention represents the calibrated Bézier curve as the contour in the preprocessed image, the resulting data volume of the preprocessed image contour extraction is smaller, making it easier for users to store. Compared to the existing technology that first identifies all contour points and then connects them to obtain the contour edges, the method of establishing a Bézier mathematical curve model requires less computation and is more efficient.

[0053] Please combine Figure 1 , Figure 4 and Figure 5 In step S23 above, calibrating the N initial control points to gradually fit the Bézier curve with the contour points specifically includes the following steps: S231, Generate a preset loss function based on the M contour points and the Bézier curve; S232, the initial control point is iteratively moved based on a preset gradient algorithm until the loss function satisfies a preset condition; S233, obtain the final position information of the current control point after the change value of the optimization degree that satisfies the loss function is less than a preset threshold; S234, Based on the final position information, obtain the calibrated Bezier curve.

[0054] It should be understood that in step S231, the preset loss function satisfies the following condition: the preset loss function is established based on the sum of the squares of the distances between the M contour points and the Bézier curve. In step S233, the optimization degree of the loss function refers to the degree of change of the value of each iteration relative to the value of the previous iteration. Iteration can stop when the current value of the loss function no longer decreases significantly relative to the previous value after several consecutive iterations, thus meeting the preset condition. Furthermore, the preset threshold can be set according to actual needs.

[0055] Please combine Figure 1 and Figure 6 In step S3 above, preprocessing the preprocessed image based on the mathematical curve model to obtain the digital signal image includes the following steps: S31, identifies all pixels in the preprocessed image; S32, based on the mathematical curve model, performs binarization processing on the preprocessed image to assign a first value or a second value to the pixels to generate a digital signal image.

[0056] Please understand that this should also be consulted. Figure 9 The system identifies all pixels in the preprocessed image based on the preset recognition module 11, and then performs binarization on the preprocessed image based on a mathematical curve model to assign values ​​to the pixels. By establishing the mathematical curve model, the outline of the curve mask is clearly displayed. The binarization process involves finding the pixel values ​​of the two highest peaks in the image, then taking the curve edge value of the curve mask image as a threshold K. For each pixel value scanned in the image, if the pixel value is greater than K, the pixel value is set to 255 (white); if the value is less than or equal to K, the pixel value is set to 0 (black). By binarizing the preprocessed image, the image is converted into a digital image, offering greater freedom than the traditional GDS-to-mask conversion.

[0057] Furthermore, the binarization process of the preprocessed image based on the mathematical curve model includes the following steps: S321, Define the mathematical curve model as the boundary and use the ray method to determine whether the pixel is inside the boundary; S322, if the pixel is within the boundary, perform the first value assignment on the pixel; if the pixel is outside the boundary, perform the second value assignment on the pixel.

[0058] Specifically, the ray casting method refers to the following: Take any pixel in the image. If a ray is drawn from this pixel in any direction, it will intersect a polygon p at several points p0, p1, ..., pn. If the number of intersection points is odd, it indicates that the pixel is inside the polygon (including pixels on the polygon), and the pixel is assigned a first value of 1. Conversely, if the number of intersection points is odd, the pixel is outside the polygon, and the pixel is assigned a second value of 2. When the ray passes through an edge of the polygon, the following situations may occur: 1. If the ray from the pixel coincides with a horizontal edge of the polygon, the point of coincidence is considered an intersection point. 2. If the ray from the pixel passes through an endpoint of the top edge of the polygon only once, the intersection point is a valid intersection point. 3. If the ray from the pixel intersects the polygon again after passing through an endpoint of the top edge, the point where it intersects the top edge is considered an invalid point. By using the ray casting method to determine whether a pixel is within the boundary and then assigning a value to the pixel, the preprocessed image is converted into a digital signal image, that is, the image is digitized, making the edge information of the processed curve mask more detailed and clearer.

[0059] In step S4 above, the convolution processing of the digital signal image includes the following steps: processing the digital signal image using the SINC function; the SINC function is: ; Where x is the pixel coordinate value.

[0060] Understandably, convolution processing can convert the assigned digital signal image into a digital image, and then downsampling processing can obtain the final rendered digital image. Downsampling involves first sampling multiple points within a preset area to form a mean point, and then connecting adjacent mean points to obtain a clear curve outline. It should be noted that the preset area is a circle drawn with a preset radius, using points on a mathematical curve model as the radius. The size of the preset area and the preset radius are not limited and can be freely adjusted according to the user's specific needs. For example, downsampling 100 points by a factor of 5 means sampling a mean point every 5 points within the preset range, ultimately forming 20 points. Then, connecting any two adjacent points within these 20 points forms the curve outline. Please refer to... Figure 9 and Figure 10 The numerical conversion process after convolution can cause numerical conversion errors. Downsampling eliminates these errors, resulting in a clearer digital image with a smaller difference from the target image.

[0061] Please see Figure 11 The present invention also provides an apparatus 1 for use in the above-mentioned digital image generation method for curve masks, the apparatus 1 comprising: Recognition module 11: Used to acquire pixels of the preprocessed image; Extraction module 12: used to extract contour points based on pixels in the preprocessed image; Calculation module 13: The calculation module includes a model building module and a calculation module; the model building module is used to establish a mathematical curve model of Bézier curves based on contour points; the calculation module is used to preprocess the preprocessed image to obtain a digital signal image; Calibration module 14: Used to calibrate N initial control points.

[0062] Digital processing module 15: used to perform convolution and downsampling processing on digital signal images in sequence to generate digital images.

[0063] The device 1 provided in this embodiment of the invention has the same beneficial effects as the above-described rapid screening method for manufacturability of overlay markings, and will not be described in detail here.

[0064] Please see Figure 12 A computer device 2, applied to the above-mentioned digital image generation method of curve mask, includes a memory 21, a processor 22 and a computer program 23 stored in the memory 21, wherein the processor 22 executes the computer program 23 to implement the steps of the rapid screening method for manufacturability of overlay marks.

[0065] The computer device 2 provided in this embodiment of the invention has the same beneficial effects as the above-described rapid screening method for manufacturability of overlay markings, and will not be described in detail here.

[0066] Compared with the prior art, the digital image generation method, apparatus, and computer equipment for curve masks provided by the present invention have the following beneficial effects: 1. This invention provides a method for generating digital images using curve masks, comprising the following steps: acquiring a preprocessed image; extracting contour points from the pixels of the preprocessed image and establishing a mathematical curve model of a Bézier curve based on the contour points; preprocessing the preprocessed image based on the mathematical curve model to obtain a digital signal image; and sequentially performing convolution and downsampling processing on the digital signal image to generate a digital image. Existing technologies obtain the contour of a curve mask by fitting pixelated pixels. During the fitting process, the large amount of pixel data requires significant computational power, and rendering takes a long time. Furthermore, the generated digital image is prone to poor clarity. This embodiment establishes a mathematical curve model using Bézier curves, preprocesses the preprocessed image based on the mathematical curve model to obtain a digital signal image, and then uses a filter for convolution. The edges of the mathematical curve model are clear, and the edges extracted through convolution processing make the extracted local features of the image clearer. Downsampling processing eliminates the numerical errors caused by numerical conversion after convolution processing, ensuring that the difference between the convolved digital image and the target image is within the confidence interval, thus enabling the rendering of the digital signal image. In addition, downsampling processing speeds up rendering and improves rendering efficiency.

[0067] 2. The method of obtaining a preprocessed image in this embodiment of the invention includes the following steps: obtaining an initial image on a mask by visual recognition; pixelating the initial image to obtain initial image pixels; performing noise reduction processing on the pixels using a preset algorithm to obtain a preprocessed image, thereby removing the influence of noise on the subsequent determination of control points by performing noise reduction processing on the pixels using a preset algorithm.

[0068] 3. The embodiment of the present invention preprocesses a preprocessed image to obtain a digital signal image based on a mathematical curve model, including the following steps: identifying all pixels in the preprocessed image; and performing binarization processing on the preprocessed image based on the mathematical curve model to assign a first value or a second value to the pixels to generate a digital signal image. By binarizing the preprocessed image to convert it into a digital image, the diffraction limit caused by angular resolution is overcome, resulting in better image clarity in subsequent rendering.

[0069] 4. The embodiment of the present invention performs binarization processing on the preprocessed image based on the mathematical curve model, including the following steps: defining the mathematical curve model as the boundary and using the ray casting method to determine whether the pixel is within the boundary; if the pixel is within the boundary, assigning a first value to the pixel; if the pixel is outside the boundary, assigning a second value to the pixel. By assigning values ​​to points within and outside the mathematical curve model on the mask, the image is numerically represented as the mask outline.

[0070] 5. The present invention also provides an apparatus that has the same beneficial effects as the above-described method for analyzing the manufacturability of overlay marks, which will not be described in detail here.

[0071] 6. The apparatus of this embodiment further includes: a digital processing module: used to sequentially perform convolution processing and downsampling processing on the digital signal image to generate a digital image.

[0072] 7. The present invention also provides a computer device that has the same beneficial effects as the above-described method for analyzing the manufacturability of overlay marks, which will not be described in detail here.

[0073] The foregoing has provided a detailed description of a method, apparatus, and computer device for generating digital images using curve masks according to embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention. Any modifications, equivalent substitutions, and improvements made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for generating digital images using curve masks, characterized in that: Includes the following steps: Obtain the preprocessed image; Contour points are extracted from pixels in a preprocessed image, and a mathematical curve model of a Bézier curve is built based on these contour points. Digital signal images are obtained by preprocessing the preprocessed images based on mathematical curve models; The digital signal image is sequentially convolved and downsampled to generate a digital image; The step of extracting contour points from pixels in a preprocessed image and establishing a mathematical curve model for a Bézier curve based on these contour points includes: Multiple contour points are extracted from the preprocessed image; Multiple initial control points are selected from the multiple contour points, and an initial Bézier curve is obtained based on the multiple initial control points; A loss function is generated based on the sum of the squares of the distances between the plurality of contour points and the initial Bézier curve; Based on the gradient algorithm, the initial control point is iteratively moved until the loss function meets the preset conditions; After the change in the optimization degree of the loss function is less than a preset threshold, the position information of the current control point is obtained; Based on the location information, a calibrated Bézier curve is obtained, which is used as the contour information extracted from the preprocessed image, and the mathematical curve model is output.

2. The digital image generation method using curve masks as described in claim 1, characterized in that: Obtaining a preprocessed image involves the following steps: The initial image on the mask is obtained through visual recognition. Pixelate the initial image to obtain the initial image pixels; A preset algorithm is used to reduce noise in the pixels to obtain a preprocessed image.

3. The digital image generation method using curve masks as described in claim 2, characterized in that: The preset algorithm is Gaussian smoothing, and the formula for Gaussian smoothing is: ； Where x is the x-axis coordinate of the pixel and y is the y-axis coordinate of the pixel. This represents the confidence interval.

4. The digital image generation method using curve masks as described in claim 1, characterized in that: Preprocessing a preprocessed image to obtain a digital signal image based on a mathematical curve model includes the following steps: Identify all pixels in the preprocessed image; The preprocessed image is binarized based on a mathematical curve model to assign a first or second value to each pixel to generate a digital signal image.

5. The digital image generation method using curve masks as described in claim 4, characterized in that: Binarization of the preprocessed image based on the mathematical curve model includes the following steps: Define the mathematical curve model as the boundary and use the ray method to determine whether the pixel is inside the boundary; If a pixel is within the boundary, assign a first value to the pixel; if a pixel is outside the boundary, assign a second value to the pixel.

6. The digital image generation method using curve masks as described in claim 5, characterized in that: The convolution processing of the digital signal image includes the following steps: processing the digital signal image using the SINC function; the SINC function is: ； Where x is the pixel coordinate value.

7. The digital image generation method using curve masks as described in claim 1, characterized in that: Contour points are extracted from pixels in a preprocessed graphic using either contour extraction or boundary tracing methods.

8. A digital image generation apparatus for a curve mask, applied to the digital image generation method for a curve mask as described in any one of claims 1-7, characterized in that: The digital image generation device for the curve mask includes: Recognition module: used to acquire pixels in the preprocessed image; Extraction module: used to extract contour points based on pixels in a preprocessed image; The calculation module includes a model building module and a computation module; the model building module is used to establish a mathematical curve model of Bézier curves based on contour points; the computation module is used to preprocess the preprocessed image to obtain a digital signal image. The calculation module is further configured to extract multiple contour points from the preprocessed image; select multiple initial control points from the multiple contour points; obtain an initial Bézier curve based on the multiple initial control points; generate a loss function based on the sum of the squares of the distances between the multiple contour points and the initial Bézier curve; iteratively move the initial control points based on a gradient algorithm until the loss function meets a preset condition; obtain the position information of the current control point after the change in the optimization degree of the loss function is less than a preset threshold; obtain the calibrated Bézier curve based on the position information, use it as the contour information extracted from the preprocessed image, and output the mathematical curve model.

9. The digital image generation apparatus for curve masks as described in claim 8, characterized in that: The digital image generation device for the curve mask further includes: a digital processing module: used to sequentially perform convolution and downsampling processing on the digital signal image to generate a digital image.

10. A computer device, applied to the digital image generation method using a curve mask as described in any one of claims 1-7, characterized in that: The method includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the digital image generation method for the curve mask.

Images