A curve mask extraction method, extraction device and storage medium

Abstract

The present invention relates to the field of computational lithography technology, and in particular to a curve mask extraction method, extraction device and storage medium. The method comprises the following steps: inputting an initial mask outline graphic, and pixelating the initial mask outline graphic to obtain pixel points corresponding to the graphic; obtaining the light transmittance of all pixel points; selecting pixel points whose light transmittance is within a preset range; determining the order of the initial Bezier curve and the number of initial control points based on the selected pixel points; iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve; and outputting the optimized Bezier curve as the final curve mask. The curve mask extraction method provided by the present invention solves the technical problem that the existing curve mask ILT calculation speed is slow and cannot be applied to the whole chip. The device and storage medium of the present invention have the same beneficial effects as the curve mask extraction method.

Description

[Technical field]

[0001] The present invention relates to the technical field of computational lithography, and in particular to a curve mask extraction method, an extraction device and a storage medium. [Background technology]

[0002] In the modern large-scale integrated circuit manufacturing process, the photolithography process occupies a core position. In order to make the graphics on the wafer after exposure as consistent as possible with the designed graphics, inverse lithography technology (ILT) is regarded as a promising project. ILT chooses to use the graphics realized on the wafer as the target to invert and calculate the graphics required on the mask. In the past development history, ILT has solved many advanced node lithography problems and challenges in the process of photolithography or extreme ultraviolet lithography. It can correct the mask image at the pixel level and has been proven to be used to construct a mask layout that maximizes the process window. Currently, ILT is used for local processing of chip bad spots, which can save a lot of computing time. However, due to the running time and mask writing time limitations in the curve mask ILT calculation technology, it cannot be used on the entire chip. Therefore, there is an urgent need for a method to improve the calculation speed of the curve mask ILT so as to realize its application in the whole chip. [Summary of the invention]

[0003] In order to solve the technical problem in the prior art that the curve mask ILT calculation speed is slow and cannot be applied to the whole chip, the present invention provides a curve mask extraction method, extraction device and storage medium.

[0004] In order to solve the above technical problems, the technical solution of the present invention is to provide a curve mask extraction method, which comprises the following steps:

[0005] Inputting an initial mask outline graphic, and performing pixel processing on the initial mask outline graphic to obtain pixel points corresponding to the graphic;

[0006] Get the light transmittance of all pixels;

[0007] Selecting pixels whose light transmittance is within a preset range;

[0008] Determining the order of the initial Bezier curve and the number of initial control points based on the selected pixel points;

[0009] Iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve;

[0010] The optimized Bezier curve is output as the final curve mask.

[0011] Preferably, the pixel points corresponding to the image obtained by pixelating the initial mask outline image are specifically:

[0012] The initial mask outline graphic is pixelated by using a GPU optimization algorithm to obtain an original matrix, thereby obtaining pixel points corresponding to the graphic.

[0013] Preferably, obtaining the light transmittance of all pixels specifically includes the steps of:

[0014] Establishing an imaging model based on the input initial mask contour graphic;

[0015] The light transmittance of all pixels in the pixelated mask outline graphic is obtained based on the imaging model and the gradient optimization algorithm.

[0016] Preferably, iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve specifically comprises the steps of:

[0017] Establish evaluation function;

[0018] Calculating the initial gradient value of the evaluation function and obtaining the direction of the iterative optimization movement of the initial Bezier curve;

[0019] Iteratively moving the initial Bezier curve based on the direction of the iterative optimization movement until the gradient value of the evaluation function meets the iteration termination criterion;

[0020] Get the optimized Bezier curve.

[0021] Preferably, the evaluation function is established based on the difference between the initial Bezier curve and the original mask image function.

[0022] Preferably, the iteration termination criterion is: setting the number of iteration steps, and terminating the iteration after completing the preset number of iteration steps; and / or setting an optimization threshold, and terminating the iteration when the difference between the iterative calculation result and the previous calculation result is less than the preset optimization threshold.

[0023] Preferably, after iteratively moving the initial Bezier curve based on the direction of the iterative optimization movement, the method further comprises the steps of:

[0024] Determine whether the difference between the gradient values ​​before and after the iteration of the evaluation function is less than a preset value;

[0025] If so, the gradient values ​​of the elements in the original matrix are calculated from a preset direction to adjust the number and / or positions of the control points of the initial Bezier curve.

[0026] Preferably, the preset directions are the x direction and the y direction of the coordinate axis.

[0027] Another technical solution of the present invention to solve the above technical problem is to provide a curve mask extraction device, which is applied to the above curve mask extraction method, comprising:

[0028] Input module: used to input the initial mask outline graphics;

[0029] Processing module: used for pixelation processing of the initial mask outline image to obtain pixel points corresponding to the image;

[0030] Acquisition module: used to obtain the light transmittance passing through all pixels;

[0031] Selection module: used to select pixels whose light transmittance is within a preset range;

[0032] A determination module: used for determining the order of the initial Bezier curve and the number of initial control points based on the selected pixel points;

[0033] Optimization module: used to iteratively optimize the initial Bezier curve to obtain the optimized Bezier curve;

[0034] Output module: Output the optimized Bezier curve as the final curve mask.

[0035] Another technical solution of the present invention to solve the above technical problem is to provide a storage medium, including a processor, and the storage medium implements the above curve mask extraction method when executed by the processor.

[0036] Compared with the prior art, the curve mask extraction method, extraction device and storage medium provided by the present invention have the following advantages:

[0037] 1. The curve mask extraction method provided by the embodiment of the present invention can obtain all the pixels in the input mask contour figure by pixelating the initial mask contour figure. The present invention first obtains the light transmittance of all pixels, and then selects the pixels whose light transmittance is within the preset range, and determines the order of the initial Bezier curve and the number of initial control points based on the selected pixels. In this way, a reference value can be provided for the order of the initial Bezier curve and the number of initial control points, thereby reducing the time of subsequent calculations and preparing for the confidence of subsequent calculation results. Finally, the obtained initial Bezier curve is iteratively optimized to obtain the optimized Bezier curve, and the optimized Bezier curve is output as the final curve mask. Through this design, the degree of fit between the optimized Bezier curve and the input mask contour figure is guaranteed, thereby further reflecting the details in the input mask contour figure. In addition, by representing the Bezier curve as a curve mask, the amount of written data can be reduced, thereby reducing the technical problem of excessive calculation in the reverse lithography technology calculation technology of the curve mask and the technical problem that it cannot be applied to the entire chip.

[0038] 2. The curve mask extraction method provided by the embodiment of the present invention can speed up the pixel processing time of the initial mask outline graphic by pixelating the initial mask outline graphic using the GPU optimization algorithm. The original matrix obtained can facilitate the subsequent calculation of the light transmittance of each pixel.

[0039] 3. The curve mask extraction method provided by the embodiment of the present invention establishes an imaging model based on the input initial mask contour graphic in order to perform mask rule check on the curve mask to ensure that the design of the curve mask does not violate the mask rule. Calculating the light transmittance of all pixels based on the imaging model and the gradient optimization algorithm can globally optimize the calculation of the light transmittance of all pixels and ensure that the subsequent light transmittance values ​​are more accurate, providing data reference values ​​for the subsequent determination of the order of the initial Bezier curve and the number of initial control points and preparing for the subsequent reduction of computing power.

[0040] 4. The curve mask extraction method provided by the embodiment of the present invention can optimize the initial Bezier curve and determine the criterion for iterative termination by establishing the design of the evaluation function. By calculating the design of the initial gradient value of the evaluation function, the moving direction of the iterative optimization of the initial Bezier curve can be obtained, which can provide certain support for subsequent iterative optimization and reduce the time of subsequent iterative optimization, thereby reducing the overall computing power. The design based on whether the gradient value after the first calculation meets the iterative termination criterion can more accurately reflect the details in the input mask contour graphic.

[0041] 5. In the curve mask extraction method provided by the embodiment of the present invention, the two iteration termination modes are used in combination to ensure the efficiency and accuracy of the iteration process.

[0042] 6. In the curve mask extraction method provided by the embodiment of the present invention, if the initial Bezier curve is moved based on the determined optimized moving direction, it is necessary to determine whether the difference in the gradient value before and after the evaluation function iteration is less than the preset value. If so, that is, when the numerical change of the current gradient value relative to the next gradient value is less than the preset value, it indicates that the number of selected initial control points is inaccurate and the number of initial control points needs to be re-determined. Therefore, the gradient values ​​of all elements in the original matrix can be calculated from the preset direction, and then the number and / or position of the initial control points can be adjusted.

[0043] 7. The curved mask extraction device provided in the embodiment of the present invention has the same beneficial effects as the above-mentioned mask extraction method, which will not be described in detail here.

[0044] 8. The storage medium provided in the embodiment of the present invention has the same beneficial effects as the above-mentioned mask extraction method, which will not be described in detail here.

Brief Description of the Drawings

[0045] Figure 1 is a schematic diagram of the overall process of the curve mask extraction method provided by the first embodiment of the present invention;

[0046] Figure 2 The detailed steps of the curve mask extraction method provided by the first embodiment of the present invention are shown in FIG. Figure 1 ;

[0047] Figure 3 The detailed steps of the curve mask extraction method provided by the first embodiment of the present invention are shown in FIG. Figure 2 ;

[0048] Figure 4 The detailed steps of the curve mask extraction method provided by the first embodiment of the present invention are shown in FIG. Figure 3 ;

[0049] Figure 5 It is a schematic diagram of the structure of a curve mask extraction device provided by the second embodiment of the present invention.

[0050] Description of the accompanying drawings:

[0051] 1. Curve mask extraction method; 2. Curve mask extraction device;

[0052] 21. Input module; 22. Processing module; 23. Acquisition module; 24. Selection module; 25. Determination module; 26. Optimization module; 27. Output module. [Specific implementation method]

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and implementation examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] The optimal mask layout produced by the application of inverse lithography technology (ILT) is usually curved. Currently, multi-beam mask writers can produce high-quality masks with curved patterns and standardize the data format of curved features on the mask. However, the running time required for its calculation is relatively long and it cannot be applied to the entire chip. In view of this, the present invention provides a curved mask extraction method.

[0055] See also Figure 1 The first embodiment of the present invention provides a curve mask extraction method 1, comprising the following steps:

[0056] S1: input an initial mask outline graphic, and perform pixel processing on the initial mask outline graphic to obtain pixel points corresponding to the graphic;

[0057] S2: Obtain the light transmittance of all pixels;

[0058] S3: Selecting pixel points whose light transmittance is within a preset range;

[0059] S4: determining the order of the initial Bezier curve and the number of initial control points based on the selected pixel points;

[0060] S5: iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve;

[0061] S6: Output the optimized Bezier curve as the final curve mask.

[0062] It can be understood that all the pixels in the mask contour graphic can be obtained by pixelating the initial mask contour graphic. Since the n-order Bezier curve can smoothly fit the target contour according to the limited control points, the present invention represents the Bezier curve as the final curve mask output, which can greatly reduce the overall computing power. If the light transmittance of each pixel is not clear, the order of the initial Bezier curve and the number of initial control points can only be selected blindly. If the order of the initial Bezier curve and the number of control points are determined by blind selection, the subsequent calculation will be redundant or insufficient. Therefore, the present invention first obtains the light transmittance of all pixels, and then selects the pixels whose light transmittance is within the preset range, and determines the order of the initial Bezier curve and the number of initial control points based on the selected pixels. In this way, a reference value can be provided for generating the order of the initial Bezier curve and the number of initial control points, thereby reducing the time of subsequent calculations and preparing for the confidence of subsequent calculation results. Finally, the initial Bezier curve is iteratively optimized to obtain the optimized Bezier curve, and the optimized Bezier curve is used as the final curve mask output. This design ensures the degree of fit between the optimized Bezier curve and the input mask contour graphic, thereby further reflecting the details in the input mask contour graphic. In addition, by representing the Bezier curve as a curve mask, the amount of written data can be reduced, thereby reducing the technical problem of excessive calculation in the reverse lithography technology calculation technology of the curve mask and the technical problem that it cannot be applied to the entire chip.

[0063] It should be noted that full-chip application means that a curved design mask is applied to the entire chip layout. However, due to the extremely low computational efficiency of the curve feature data, the existing technology requires about three days of computing time to achieve full-chip application. After adopting this solution, the computing time will be greatly reduced. The current longest test time is within 42 hours. It can be seen that the computing time will be greatly reduced after adopting this solution, thereby achieving full-chip application.

[0064] Specifically, selecting the pixel points whose light transmittance is within the preset range in step S3 means selecting the pixel points whose light transmittance has a larger value. It can be understood that by selecting the pixel points whose light transmittance has a larger value, the degree of fit between the initial Bezier curve obtained subsequently and the original mask contour image can be ensured, thereby further reducing the subsequent computing power and being able to feedback more details on the original mask contour image.

[0065] Furthermore, the initial mask outline pattern is pixelated to obtain the pixel points corresponding to the pattern:

[0066] The GPU optimization algorithm is used to pixelate the initial mask outline image and obtain the original matrix to obtain the pixel points corresponding to the image.

[0067] It can be understood that the design of pixelizing the initial mask outline pattern through the GPU optimization algorithm can speed up the pixelization process, and the original matrix obtained can facilitate the subsequent calculation of the light transmittance of each pixel.

[0068] It should be noted that the CPU optimization algorithm can also be improved to perform pixelation processing on the initial mask outline graphic, as long as the time for performing pixelation processing on the initial mask outline graphic can be accelerated through the corresponding algorithm, and no limitation is made here.

[0069] For further information, see Figure 2 , obtaining the light transmittance of all pixels specifically includes the steps of:

[0070] S21: establishing an imaging model based on the input initial mask contour graphic;

[0071] S22: Obtain the light transmittance of all pixels in the pixelated mask contour graphic based on the imaging model and the gradient optimization algorithm.

[0072] It should be noted that the light transmittance of each pixel is calculated based on the uniform distribution of the light intensity of the coherent light source. This design can ensure the accuracy of the subsequent light transmittance of all pixels, avoid errors in the light transmittance of all pixels due to uneven light intensity distribution, and thus avoid differences in the initial control points and orders, thereby avoiding an increase in subsequent calculation time.

[0073] It is understandable that the imaging model is established based on the input initial mask contour graphic in order to check the mask rules of the curve mask to ensure that the design of the curve mask does not violate the mask rules. Calculating the light transmittance of all pixels based on the imaging model and gradient optimization algorithm can globally optimize the calculation of the light transmittance of all pixels and ensure that the subsequent light transmittance values ​​are more accurate, providing data reference values ​​for the subsequent determination of the order of the initial Bezier curve and the number of initial control points, and preparing for the subsequent reduction of computing power.

[0074] For further information, see Figure 3 , iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve specifically includes the steps of:

[0075] S51: Establishing an evaluation function;

[0076] S52: Calculate the initial gradient value of the evaluation function and obtain the direction of the initial Bezier curve iterative optimization movement;

[0077] S53: iteratively moving the initial Bezier curve based on the iterative optimization moving direction until the gradient value of the evaluation function meets the iteration termination criterion;

[0078] S54: Obtain an optimized Bezier curve.

[0079] Specifically, an evaluation function is established based on the difference between the initial Bezier curve and the original mask image function.

[0080] It can be understood that by establishing the design of the evaluation function, the initial Bezier curve can be optimized and the criteria for iterative termination can be determined. By calculating the design of the initial gradient value of the evaluation function, the moving direction of the iterative optimization of the initial Bezier curve can be obtained, which can provide certain support for subsequent iterative optimization, and can also reduce the time of subsequent iterative optimization, thereby reducing the overall computing power. In addition, based on the design of calculating the gradient value of the evaluation function until the iterative termination criteria are met, iterative optimization ensures that the final Bezier curve can fit the original mask contour figure, and can also more accurately reflect the details in the mask contour figure.

[0081] Specifically, the evaluation function is expressed as:

[0082]

[0083] Where m represents the number of control points on the Bezier curve, f represents the mask-related intensity function, Q j represents the jth point on the curve, C n represents the nth control point. represents the mask-dependent intensity function of the jth point on the Bezier curve, Represents an intermediate function connecting the light intensity and the control points of the Bezier curve.

[0084] Specifically, the iteration termination criteria are: setting the number of iteration steps, and terminating the iteration after completing the preset number of iteration steps; and / or setting the optimization threshold, and terminating the iteration when the difference between the iterative calculation result and the previous calculation result is less than the preset optimization threshold. It can be understood that the two iteration termination methods can be used together to ensure the efficiency and accuracy of the iteration process.

[0085] For further information, see Figure 4 , after iteratively moving the initial Bezier curve based on the direction of iterative optimization movement, it also includes the steps of;

[0086] S531: Determine whether the difference between the gradient values ​​before and after the iteration of the evaluation function is less than a preset value;

[0087] S532: If yes, calculate the gradient value of the element in the original matrix from a preset direction to adjust the number and / or position of the control points of the initial Bezier curve.

[0088] S533: If not, continue to calculate the gradient value of the evaluation function until the gradient value meets the iteration termination criterion.

[0089] Specifically, the preset directions are the x direction and the y direction of the coordinate axis.

[0090] It can be understood that if the initial Bezier curve is moved based on the determined optimization moving direction, it is necessary to determine whether the difference in the gradient value before and after the evaluation function iteration is less than the preset value. If so, that is, when the numerical change of the current gradient value relative to the next gradient value is less than the preset value, it indicates that the number of selected initial control points is inaccurate, and the number and / or position of the initial control points need to be re-determined to avoid the technical problem of large computing time or poor fit between the final output Bezier curve and the original mask contour graphic in the subsequent iterative optimization process. Therefore, the gradient values ​​of all elements in the original matrix can be calculated from the preset direction, and then the number and / or position of the initial control points can be adjusted, thereby further reducing the computing power and improving the fit between the final output Bezier curve and the original mask contour graphic.

[0091] It should also be noted that if the calculated gradient data values ​​in the x and y directions are optimized well enough, that is, the calculated current gradient value relative to the previous gradient value is less than the preset threshold, even if the global optimization does not have a more obvious decline, then there is no need for global element optimization, and the iteration can be terminated early to further improve the running speed.

[0092] See also Figure 5The second embodiment of the present invention provides a curve mask extraction device 2, which is applied to the curve mask extraction method provided in the first embodiment of the present invention, comprising:

[0093] Input module 21: used to input an initial mask outline graphic;

[0094] Processing module 22: used for performing pixel processing on the initial mask outline pattern to obtain pixel points corresponding to the pattern;

[0095] Acquisition module 23: used to acquire light transmittance of all pixels in the mask outline pattern after pixelation processing;

[0096] Selection module 24: used to select pixel points whose light transmittance is within a preset range;

[0097] Determination module 25: used to determine the order of the initial Bezier curve and the number of initial control points based on the selected pixel points;

[0098] Optimization module 26: used for iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve;

[0099] Output module 27: outputs the optimized Bezier curve as the final curve mask.

[0100] It can be understood that the second embodiment of the present invention provides a curved mask extraction device 2 having the same beneficial effects as the curved mask extraction method provided by the first embodiment of the present invention, which will not be described in detail herein.

[0101] A third embodiment of the present invention provides a storage medium, including a processor. When the storage medium is executed by the processor, the curve mask extraction method provided by the first embodiment of the present invention is implemented.

[0102] It can be understood that the third embodiment of the present invention provides a storage medium having the same beneficial effects as the curve mask extraction method provided by the first embodiment of the present invention, which will not be described in detail herein.

[0103] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. Those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required for the present invention.

[0104] In various embodiments of the present invention, it should be understood that the size of the serial numbers of the above-mentioned processes does not mean the necessary 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.

[0105] The flowchart and block diagram in the accompanying drawings of the present invention illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementation schemes, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which is determined based on the functions involved. It should be particularly noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0106] Compared with the prior art, the curve mask extraction method, extraction device and storage medium provided by the present invention have the following advantages:

[0107] 1. The curve mask extraction method provided by the embodiment of the present invention can obtain all the pixels in the input mask contour figure by pixelating the initial mask contour figure. The present invention first obtains the light transmittance of all pixels, and then selects the pixels whose light transmittance is within the preset range, and determines the order of the initial Bezier curve and the number of initial control points based on the selected pixels. In this way, a reference value can be provided for the order of the initial Bezier curve and the number of initial control points, thereby reducing the time of subsequent calculations and preparing for the confidence of subsequent calculation results. Finally, the obtained initial Bezier curve is iteratively optimized to obtain the optimized Bezier curve, and the optimized Bezier curve is output as the final curve mask. Through this design, the degree of fit between the optimized Bezier curve and the input mask contour figure is guaranteed, thereby further reflecting the details in the input mask contour figure. In addition, by representing the Bezier curve as a curve mask, the amount of written data can be reduced, thereby reducing the technical problem of excessive calculation in the reverse lithography technology calculation technology of the curve mask and the technical problem that it cannot be applied to the entire chip.

[0108] 2. The curve mask extraction method provided by the embodiment of the present invention can speed up the pixel processing time of the initial mask outline graphic by pixelating the initial mask outline graphic using the GPU optimization algorithm. The original matrix obtained can facilitate the subsequent calculation of the light transmittance of each pixel.

[0109] 3. The curve mask extraction method provided by the embodiment of the present invention establishes an imaging model based on the input initial mask contour graphic in order to perform mask rule check on the curve mask to ensure that the design of the curve mask does not violate the mask rule. Calculating the light transmittance of all pixels based on the imaging model and the gradient optimization algorithm can globally optimize the calculation of the light transmittance of all pixels and ensure that the subsequent light transmittance values ​​are more accurate, providing data reference values ​​for the subsequent determination of the order of the initial Bezier curve and the number of initial control points and preparing for the subsequent reduction of computing power.

[0110] 4. The curve mask extraction method provided by the embodiment of the present invention can optimize the initial Bezier curve and determine the criterion for iterative termination by establishing the design of the evaluation function. By calculating the design of the initial gradient value of the evaluation function, the moving direction of the iterative optimization of the initial Bezier curve can be obtained, which can provide certain support for subsequent iterative optimization and reduce the time of subsequent iterative optimization, thereby reducing the overall computing power. The design based on whether the gradient value after the first calculation meets the iterative termination criterion can more accurately reflect the details in the input mask contour graphic.

[0111] 5. In the curve mask extraction method provided by the embodiment of the present invention, the two iteration termination modes are used in combination to ensure the efficiency and accuracy of the iteration process.

[0112] 6. In the curve mask extraction method provided by the embodiment of the present invention, if the initial Bezier curve is moved based on the determined optimized moving direction, it is necessary to determine whether the difference in the gradient value before and after the evaluation function iteration is less than the preset value. If so, that is, when the numerical change of the current gradient value relative to the next gradient value is less than the preset value, it indicates that the number of selected initial control points is inaccurate and the number of initial control points needs to be re-determined. Therefore, the gradient values ​​of all elements in the original matrix can be calculated from the preset direction, and then the number and / or position of the initial control points can be adjusted.

[0113] 7. The curved mask extraction device provided in the embodiment of the present invention has the same beneficial effects as the above-mentioned mask extraction method, which will not be described in detail here.

[0114] 8. The storage medium provided in the embodiment of the present invention has the same beneficial effects as the above-mentioned mask extraction method, which will not be described in detail here.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A curve mask extraction method, characterized in that: The curve mask extraction method comprises the following steps: Inputting an initial mask outline graphic, and performing pixel processing on the initial mask outline graphic to obtain pixel points corresponding to the graphic; Get the light transmittance of all pixels; Selecting pixels whose light transmittance is within a preset range; Determining the order of the initial Bezier curve and the number of initial control points based on the selected pixel points; Iteratively optimizing the initial Bezier curve to obtain an optimized Bezier curve; Establish evaluation function; Calculating the initial gradient value of the evaluation function and obtaining the direction of the iterative optimization movement of the initial Bezier curve; Iteratively moving the initial Bezier curve based on the direction of the iterative optimization movement until the gradient value of the evaluation function meets the iteration termination criterion; Get the optimized Bezier curve; The optimized Bezier curve is output as the final curve mask.

2. The curve mask extraction method according to claim 1, characterized in that: The pixel points corresponding to the image obtained by pixelating the initial mask outline image are specifically: The initial mask outline graphic is pixelated by using a GPU optimization algorithm to obtain an original matrix, thereby obtaining pixel points corresponding to the graphic.

3. The curve mask extraction method according to claim 1, characterized in that: Obtaining the light transmittance of all pixels specifically includes the steps of: Establishing an imaging model based on the input initial mask contour graphic; The light transmittance of all pixels in the pixelated mask outline graphic is obtained based on the imaging model and the gradient optimization algorithm.

4. The curve mask extraction method according to claim 1, characterized in that: The evaluation function is established based on the difference between the initial Bezier curve and the original mask image function.

5. The curve mask extraction method according to claim 1, characterized in that: The iteration termination criteria are: setting the number of iteration steps, and terminating the iteration after completing the preset number of iteration steps; and / or setting an optimization threshold, and terminating the iteration when the difference between the iterative calculation result and the previous calculation result is less than the preset optimization threshold.

6. The curve mask extraction method according to claim 2, characterized in that: After iteratively moving the initial Bezier curve based on the direction of the iterative optimization movement, the following steps are also included: Determine whether the difference between the gradient values ​​before and after the iteration of the evaluation function is less than a preset value; If so, the gradient values ​​of the elements in the original matrix are calculated from a preset direction to adjust the number and / or positions of the control points of the initial Bezier curve.

7. The curve mask extraction method according to claim 6, characterized in that: The preset directions are the x direction and the y direction of the coordinate axis.

8. A curved mask extraction device, applied to the curved mask extraction method according to any one of claims 1 to 7, characterized in that: include Input module: used to input the initial mask outline graphics; Processing module: used for pixelation processing of the initial mask outline image to obtain pixel points corresponding to the image; Acquisition module: used to obtain the light transmittance passing through all pixels; Selection module: used to select pixels whose light transmittance is within a preset range; A determination module: used for determining the order of the initial Bezier curve and the number of initial control points based on the selected pixel points; Optimization module: used to iteratively optimize the initial Bezier curve to obtain the optimized Bezier curve; Output module: Output the optimized Bezier curve as the final curve mask.

9. A storage medium, characterized in that: It includes a processor, and when the storage medium is executed by the processor, it implements the curve mask extraction method according to any one of claims 1 to 7.

Images