OPC model overfitting detection method and device, equipment and storage medium

Abstract

The invention provides an OPC model overfitting detection method, device and equipment and a storage medium, relates to the technical field of integrated circuits, solves the problem that whether an OPC model is overfitted or not is difficult to accurately judge in the related art, and can quickly and simply determine whether the OPC model is overfitted or not based on light intensity calculation of sampling points, thereby improving the detection efficiency. Therefore, the model can be adjusted in time, and the simulation precision of the OPC model can be improved.

Description

Technical Field

[0001] The present application relates to the field of integrated circuit technology, and in particular to an OPC model overfitting detection method, device, equipment and storage medium. Background Art

[0002] Chip manufacturing is a highly complex and demanding process. As a key link in the photolithography process, it determines the precision of the integrated circuits on the chip and affects the overall function and efficiency of the chip. In the photolithography process, the mainstream optical resolution enhancement technologies include off-axis illumination (OAI), optical proximity correction (OPC), phase-shift mask (PSM), polarized illumination, and source mask joint optimization (SMO) and other methods. Among them, optical proximity correction technology is a commonly used photolithography technology.

[0003] Moreover, the correction scheme based on the OPC model is to establish a simulation system to predict the behavior of light on the chip surface during the lithography process and the final circuit pattern through simulation. However, in the traditional scheme, the OPC model may perform poorly on new or unseen data due to over-learning of training data, that is, overfitting occurs. The overfitting phenomenon causes the model to perform well on the training data, but it cannot accurately predict or replicate the results in actual applications, resulting in a decrease in the accuracy of the OPC model simulation, which ultimately affects the lithography effect and causes a decrease in the sophistication of the integrated circuits on the chip. Summary of the invention

[0004] The present application provides an OPC model overfitting detection method, device, equipment and storage medium, which solves the problem in the related art that it is difficult to accurately judge whether the OPC model is overfitting. The present solution can quickly and easily determine whether the OPC model is overfitting based on the light intensity calculation of the sampling point, so as to adjust the model in time, thereby helping to improve the simulation accuracy of the OPC model.

[0005] In a first aspect, the present application provides an OPC model overfitting detection method, which comprises: Obtain model images corresponding to several target graphics through the OPC model, and set a segmentation line on each target graphic; Based on the OPC model, the light intensity of the sampling points on the dividing line is calculated to determine the light intensity value corresponding to each sampling point; According to the light intensity value corresponding to each sampling point, determine the light intensity parameter corresponding to each target graphic; According to the light intensity parameters and the corresponding weights corresponding to each target graphic, the fitting parameters corresponding to the OPC model are determined; According to the fitting parameters, it is determined whether the OPC model is in an overfitting state.

[0006] In a second aspect, the present application also provides an OPC model overfitting detection device, which includes: An image acquisition module is configured to acquire model images corresponding to a plurality of target graphics and set a segmentation line on each target graphic; The light intensity calculation module is configured to perform light intensity calculations on the sampling points on the dividing line based on the OPC model to determine the light intensity value corresponding to each sampling point; A parameter determination module is configured to determine the light intensity parameter corresponding to each target graphic according to the light intensity value corresponding to each sampling point; A fitting detection module is configured to determine fitting parameters corresponding to the OPC model according to light intensity parameters corresponding to each target pattern and corresponding weights; The overfitting judgment module is configured to determine whether the OPC model is in an overfitting state according to the fitting parameters.

[0007] In a third aspect, the present application also provides a detection device, which includes: one or more processors; a storage device for storing one or more programs, When one or more programs are executed by one or more processors, the one or more processors implement the OPC model overfitting detection method of the present application.

[0008] In a fourth aspect, the present application also provides a storage medium storing computer executable instructions, which, when executed by a processor, are used to execute the OPC model overfitting detection method of the present application.

[0009] The present application scheme can simply and quickly determine whether the OPC model is overfitting by calculating the light intensity at the sampling points. The calculation and judgment are fast and accurate, which helps to quickly and accurately determine whether the model is overfitting during the model building process, so that the OPC model can be adjusted in time when overfitting exists, which helps to improve the simulation accuracy of the OPC model. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the steps of an OPC model overfitting detection method provided in one embodiment of the present application.

[0011] Figure 2 A schematic diagram of the OPC model output results provided in one embodiment of the present application.

[0012] Figure 3 for Figure 2 The waveform diagram of the first light intensity value corresponding to the OPC model output result is shown.

[0013] Figure 4 for Figure 2The waveform diagram of the second light intensity value corresponding to the OPC model output result is shown.

[0014] Figure 5 A schematic diagram of the steps for calculating light intensity parameters provided in one embodiment of the present application.

[0015] Figure 6 A schematic diagram of the steps for calculating fitting parameters provided in one embodiment of the present application.

[0016] Figure 7 A schematic diagram of the structure of an OPC model overfitting detection device provided in one embodiment of the present application.

[0017] Figure 8 A schematic diagram of the structure of a detection device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] The embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are only used to explain the embodiments of the present application, rather than to limit the embodiments of the present application. It should also be noted that, for ease of description, only the parts related to the embodiments of the present application rather than all structures are shown in the accompanying drawings, and those skilled in the art should be able to think of it after reading the specification of this application that as long as the technical features do not contradict each other, any combination of the technical features can constitute an optional implementation method.

[0019] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable when appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally a class, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated before and after are in an "or" relationship. In the description of the present application, "multiple" means two or more, and "several" means one or more.

[0020] Chip manufacturing is a highly complex and demanding process, in which every tiny step is directly related to the performance and quality of the final product. Among this series of precise processes, the photolithography process is undoubtedly the most critical link. It determines the fineness of the circuit pattern on the chip, which in turn affects the overall function and efficiency of the chip.

[0021] In the photolithography process, when the pattern is transferred from the mask to the chip, it will be limited by the Rayleigh formula, which will affect the minimum size that can be achieved and the photolithography resolution. With the continuous development of integrated circuits, the critical dimensions of chips have gradually shrunk to the level of tens of nanometers. In this regard, improving the photolithography resolution has become a technical problem that needs to be solved urgently. The main means of improving resolution in the early days included shortening the exposure wavelength and increasing the numerical aperture of the projection objective lens, which usually required replacing the exposure equipment and improving the process. However, further increasing the numerical aperture of the projection objective lens and reducing the exposure wavelength not only greatly increased the difficulty of technical implementation, but also sharply increased the cost. Therefore, resolution enhancement technology came into being, aiming to effectively improve the photolithography resolution by reducing the process factor.

[0022] At present, mainstream resolution enhancement technologies include off-axis illumination, optical proximity correction, phase-shift mask, polarized illumination, and light source mask joint optimization. Among them, technologies such as OPC and SMO require the use of numerical modeling to accurately describe the process and optimize related process variables through simulation process results. Such methods are collectively referred to as computational lithography. Optical proximity correction is a commonly used computational lithography technology in the industry, and model-based OPC is a typical representative. Its core is to establish a simulation system. Whether the model can accurately simulate and predict the behavior of light on the chip surface during the lithography process, and how these behaviors affect the final circuit pattern, is the basis for ensuring stable and reliable chip performance.

[0023] However, in traditional solutions, the OPC model may perform poorly on new or unseen data due to over-learning of training data, which is called overfitting. Overfitting causes the model to perform well on training data, but cannot accurately predict or replicate results in actual applications, resulting in reduced accuracy of OPC model simulation, which ultimately affects the lithography effect and reduces the sophistication of integrated circuits on the chip.

[0024] In this regard, the present application provides an OPC model overfitting detection method, which can be applied to detection equipment, such as computers, servers and other electronic devices. It can be imagined that the detection equipment includes a processor to execute a corresponding program / module through the processor to implement the OPC model overfitting detection method of the present application, such as Figure 1 As shown, Figure 1 This is a schematic diagram of the steps of the OPC model overfitting detection method provided by an embodiment of the present application. Accordingly, the specific steps of the OPC model overfitting detection method provided by the present application include steps S110-S150: Step S110: Obtain model images corresponding to a plurality of target graphics through the OPC model, and set a segmentation line on the model image corresponding to each target graphic.

[0025] It is conceivable that the target graphics that need to be exposed and developed during the OPC process, such as the graphics of through holes, connecting lines, etc. in the corresponding circuit diagram, can obtain the corresponding model images after being processed by the OPC model. It is understandable that the OPC model can establish a simulation system, and then simulate the model image corresponding to the target graphics. The OPC model can simulate the imaging of the projection lithography machine illumination and exposure system, and simulate the spatial image formed by the development after the photoresist is applied. For this, a corresponding dividing line is set on the formed model image.

[0026] In one embodiment, in this regard, in the process of setting the segmentation line, it is necessary to determine the center point of the target pattern, such as the target pattern corresponding to the through hole, and the corresponding center point is the center point of the through hole. The OPC model includes an optical model and a photoresist model, the optical model is used to generate a projection image of the corresponding target pattern, and the photoresist model is used to generate a photoresist contour image of the corresponding target pattern.

[0027] Among them, the optical model is usually used to describe the imaging of the projection lithography machine illumination and exposure system. It should be noted that in some embodiments, the optical model used is a TCC model constructed based on the Hopkins optical imaging theory. As for the photoresist model, it is assumed that the concentration of the photoacid in the photoresist reaches a certain threshold and is developed. Since the concentration of the photoacid in the photoresist is proportional to the light intensity of the spatial image, a threshold is used to cut off the light intensity graph of the spatial image, and the edge of the area where the light intensity is higher than this threshold forms the outline of the photoresist.

[0028] like Figure 2 As shown, Figure 2 This is a schematic diagram of the output result of the OPC model provided in an embodiment of the present application. In the figure, region S1 and region S2 overlap so that region S1 only retains the edge, wherein region S1 corresponds to the projection image generated by the optical model, and region S2 corresponds to the photoresist contour image generated by the photoresist model. Based on this, the segmentation line L1 passes through the center point and the above-mentioned projection image and photoresist contour image. For details, refer to Figure 2 In addition, multiple sampling points are set on the dividing line L1 at a preset interval to facilitate the subsequent calculation of the corresponding light intensity values ​​of the sampling points. Therefore, by setting a dividing line passing through the center point, the present solution can more clearly determine the contour of the model image, thereby helping to improve the accuracy of the judgment of whether the projection image and the photoresist contour image are previously fitted.

[0029] Step S120: Based on the OPC model, light intensity calculation is performed on each sampling point on the segmentation line to determine the light intensity value corresponding to each sampling point.

[0030] In one embodiment, a projection image and a photoresist contour image are respectively formed by the optical model and the photoresist model in the OPC model. Accordingly, a first light intensity value corresponding to the sampling point on the dividing line is determined according to the projection image formed by the optical model in the OPC model; and a second light intensity value corresponding to the sampling point on the dividing line is determined according to the photoresist contour image formed by the photoresist model in the OPC model.

[0031] It is conceivable that the calculation of the light intensity of the sampling points in the generated model image can be realized by calling the corresponding software tool or program. For example, Calibre Workbench is used to generate the light intensity value corresponding to each sampling point. It is understandable that after calling the software tool, the output result of the adopted OPC model is imported into the software tool, and optical simulation is performed according to preset optical parameters (such as light source wavelength, intensity distribution, refractive index of optical materials, absorption coefficient, etc.) to complete the calculation of ray tracing and field propagation, and then the light intensity is sampled in the area of ​​interest, and the light intensity value is recorded for each sampling point.

[0032] Figure 3 for Figure 2 The waveform diagram of the first light intensity value corresponding to the output result of the OPC model shown, Figure 4 for Figure 2 The waveform diagram of the second light intensity value corresponding to the output result of the OPC model shown, and Figure 3 and Figure 4 The horizontal axis corresponds to the position of the sampling point on the dividing line, and the vertical axis corresponds to the corresponding light intensity value. Figure 3 and Figure 4 In the figure, the light intensity value corresponding to the corresponding boundary is indicated by a dotted line (for ease of description, this value is defined as the boundary threshold), that is, the light intensity value of the sampling point outside the boundary is less than the boundary threshold, and the light intensity value of the sampling point inside the boundary is greater than the boundary threshold. It can be imagined that the width of the image contour can be determined by the spacing between the sampling points at the corresponding boundary.

[0033] It should be noted that for the calculation of the light intensity at each sampling point, software tools such as SentaurusLithography and Prolith can also be used.

[0034] Step S130: Determine the light intensity parameter corresponding to each target graphic according to the light intensity value corresponding to each sampling point.

[0035] For each target graphic, after the light intensity calculation of the sampling points is completed, the corresponding light intensity parameter can be calculated according to the light intensity value corresponding to the sampling point on the target graphic. For example, in some embodiments, the average value of the light intensity value can be used as the light intensity parameter of the target graphic. For another example, in some embodiments, the corresponding light intensity parameter can be calculated using the difference in the light intensity value. It is conceivable that by calculating the light intensity parameter of each target graphic, this can be used as an overfitting indicator to calculate the fitting parameters of the model based on the light intensity parameter.

[0036] Step S140: Determine the fitting parameters corresponding to the OPC model according to the light intensity parameters and the corresponding weights corresponding to each target pattern.

[0037] In the case where multiple target graphics are sampled, a corresponding weight is set for each target graphic, and then the corresponding fitting parameters are calculated according to the weight and the light intensity parameters corresponding to each target graphic, such as calculating the corresponding fitting parameters by weighted averaging of each light intensity parameter according to the corresponding weight, so as to judge whether the OPC model is overfitting. It is conceivable that in the case where only one target graphic is sampled, the corresponding fitting parameters can also be calculated using the weight set for the target graphic.

[0038] Step S150: Determine whether the OPC model is in an overfitting state according to the fitting parameters.

[0039] After determining the fitting parameters corresponding to the OPC model, the fitting parameters are compared and judged based on a preset threshold, and then it is determined whether there is an overfitting risk, so as to determine that the OPC model is in an overfitting state when there is an overfitting risk. Optionally, in one embodiment, a preset risk threshold is set for the overfitting state, and if the fitting parameter is greater than or equal to the preset risk threshold, it is determined that the OPC model is in an overfitting state; if the fitting parameter is less than the preset risk threshold, it is determined that the OPC model is not in an overfitting state.

[0040] It can be seen from the above scheme that this scheme can simply and quickly judge whether the OPC model is overfitting by calculating the light intensity of the sampling points. The calculation and judgment are fast and accurate, which helps to quickly and accurately judge whether the model is overfitting during the model establishment process, so that the OPC model can be adjusted in time when overfitting exists, which helps to improve the simulation accuracy of the OPC model.

[0041] Figure 5A schematic diagram of the steps for calculating light intensity parameters provided for one embodiment of the present application. In one embodiment, a projection image can be obtained through the optical model in the OPC model, and a photoresist projection pattern can be obtained through the photoresist model in the OPC model, and then the light intensity values ​​of the sampling points on the projection image and the photoresist image are calculated respectively, and then the light intensity difference of the sampling point can be calculated based on the light intensity value of the same sampling point. Among them, the light intensity value of the corresponding target sampling point determined based on the optical model is the first light intensity value, and the light intensity value of the corresponding target sampling point determined based on the photoresist model is the second light intensity value. Then the light intensity parameter corresponding to the target pattern is determined by the light intensity difference, and the specific steps are as follows: Step S210: determining a light intensity difference according to a first light intensity value and a second light intensity value corresponding to the same target sampling point.

[0042] Step S220: Determine a first cumulative sum of light intensity differences corresponding to all target sampling points and a second cumulative sum of first light intensity values ​​corresponding to all target sampling points.

[0043] Step S230: Based on the values ​​of the first cumulative sum and the second cumulative sum, determine the ratio of the first cumulative sum to the second cumulative sum as the light intensity parameter corresponding to the current target pattern.

[0044] It can be understood that the target sampling point is the sampling point selected during the current calculation. It can be imagined that in some embodiments, all sampling points on the dividing line can be selected as target sampling points, or some sampling points can be selected as target sampling points. For the same sampling point, the difference between the first light intensity value and the second light intensity value corresponding to the target sampling point is calculated as the corresponding light intensity difference. After determining the light intensity difference corresponding to all target sampling points, the corresponding cumulative sum is accumulated, that is, the light intensity difference corresponding to all sampling points is accumulated to determine the first cumulative sum. And the cumulative sum is also calculated for the first light intensity value, that is, the first light intensity values ​​of all target sampling points are accumulated to determine the second cumulative sum. Then, the ratio of the first cumulative sum to the second cumulative sum is used as the light intensity parameter corresponding to the current target graphic.

[0045] For example, when n sampling points are set on the dividing line, for the n sampling points, the first light intensity value Io i (i=1, 2, ..., n), the second light intensity value Ir j (j=1, 2, ..., n), calculate the light intensity parameter I single The calculation formula is as follows:

[0046] Therefore, by calculating the light intensity difference at the target sampling point, this scheme can accurately determine the difference between the output results of the OPC model, and then use it to judge the fitting of the OPC model, which helps to more efficiently detect whether the OPC model is in an overfitting state.

[0047] Figure 6 A schematic diagram of the steps for calculating the fitting parameters provided in one embodiment of the present application. In one embodiment, a corresponding weight is set for each target pattern, and then the obtained light intensity parameters are weighted averaged according to the weight to determine the corresponding fitting parameters. The specific steps are as follows: Step S310: Determine the weight corresponding to the target graph according to the number and importance values ​​of graphs of the same type as the target graph among all graphs.

[0048] Step S320: Calculate the corresponding weighted average value according to the light intensity parameter and the corresponding weight of each target graphic, and use the weighted average value as the fitting parameter.

[0049] It can be understood that the weight set for each target graphic is determined based on the number of graphics of the same type as the target graphic in all graphics and the importance value. For example, there are 100 graphics in total, and the number of graphics of the same type as the target graphic is 10. Correspondingly, the number weight of the target graphic is 0.1. In addition, different importance values ​​are set for different types of graphics on different layout layers, that is, the importance value is related to the type of target graphic and the layout layer where the target graphic is located. For example, for the layout layer of the through-hole layer, the importance of the square type of graphics is high, and accordingly, its importance value is also high; and for the layout layer of the AA layer (Active Area Layer, active area layer), the importance of the rectangular type of graphics is higher than that of the square type. It should be noted that the specific value of the importance value can be preset according to the actual situation. Exemplarily, after determining the type of the target graphic and the layout layer it is in, its corresponding importance value can also be determined. For example, the importance value corresponding to the target graphic is 0.5, and the corresponding quantity weight is 0.1. Based on this, the weight corresponding to the target graphic is 0.1*0.5=0.05.

[0050] Therefore, after determining the light intensity parameters and weights of each target pattern, the light intensity parameters are weighted averaged to determine the corresponding weighted average value and use this value as the fitting parameter of the current OPC model. Based on this, the calculation formula for calculating the fitting parameter OverFitIndex is as follows:

[0051] Among them, W nis the weight corresponding to the nth target graph.

[0052] Furthermore, this scheme sets corresponding weights for the sampled target graphics and performs weighted averaging on the determined light intensity parameters to determine the fitting parameters, and then uses the fitting parameters as over-fitting judgment indicators to make over-fitting judgments on the OPC model, which helps to more accurately judge the fitting of the OPC model and is conducive to generating a more accurate OPC model.

[0053] Figure 7 The schematic diagram of the structure of the OPC model overfitting detection device provided in one embodiment of the present application is used to execute the OPC model overfitting detection method provided in the above embodiment, and has the corresponding functional modules and beneficial effects of the execution method. As shown in the figure, the device includes an image acquisition module 401, a light intensity calculation module 402, a parameter determination module 403, a fitting detection module 404 and an overfitting judgment module 405.

[0054] Among them, the image acquisition module 401 is configured to acquire model images corresponding to several target graphics through the OPC model, and set a dividing line on each target graphic; the light intensity calculation module 402 is configured to perform light intensity calculations on the sampling points on the dividing line based on the OPC model to determine the light intensity value corresponding to each sampling point; the parameter determination module 403 is configured to determine the light intensity parameters corresponding to each target graphic according to the light intensity value corresponding to each sampling point; the fitting detection module 404 is configured to determine the fitting parameters corresponding to the OPC model according to the light intensity parameters corresponding to each target graphic and the corresponding weights; the overfitting judgment module 405 is configured to determine whether the OPC model is in an overfitting state based on the fitting parameters.

[0055] On the basis of the above embodiment, the OPC model includes an optical model and a photoresist model, the optical model is used to generate a projection image corresponding to the target pattern, and the photoresist model is used to generate a photoresist contour image corresponding to the target pattern. The image acquisition module 401 is specifically configured as follows: Determine the center point of each target shape; Generate a projection image and a photoresist outline image corresponding to each target pattern, and set a segmentation line passing through the projection image and the photoresist outline image corresponding to each target pattern, and the center point of the target pattern is located on the segmentation line; Set multiple sampling points on the dividing line according to the preset interval.

[0056] Based on the above embodiment, the light intensity calculation module 402 is specifically configured as follows: Determine a first light intensity value corresponding to a sampling point on a segmentation line according to a projection image formed by an optical model in an OPC model; According to the photoresist contour image formed by the photoresist model in the OPC model, the second light intensity value corresponding to the sampling point on the segmentation line is determined.

[0057] On the basis of the above embodiment, the optical model is a TCC model constructed based on Hopkins optical imaging theory.

[0058] On the basis of the above embodiment, the OPC model includes an optical model and a photoresist model, and the parameter determination module 403 is specifically configured as follows: Determine a light intensity difference according to a first light intensity value and a second light intensity value corresponding to the same target sampling point, wherein the first light intensity value is a light intensity value of the corresponding target sampling point determined based on the optical model, and the second light intensity value is a light intensity value of the corresponding target sampling point determined based on the photoresist model; Determine a first cumulative sum of light intensity differences corresponding to all target sampling points, and a second cumulative sum of first light intensity values ​​corresponding to all target sampling points; Based on the values ​​of the first cumulative sum and the second cumulative sum, a ratio of the first cumulative sum to the second cumulative sum is determined as a light intensity parameter corresponding to the current target pattern.

[0059] Based on the above embodiment, the fitting detection module 404 is specifically configured as follows: Determine the weight corresponding to the target graphic according to the number of graphics of the same type as the target graphic among all graphics and their importance values, where the importance value is associated with the type of the target graphic and the layout layer where the target graphic is located; According to the light intensity parameters and weights corresponding to each target graphic, the corresponding weighted average is calculated, and the weighted average is used as the fitting parameter.

[0060] Based on the above embodiment, the overfitting judgment module 405 is specifically configured as follows: When the fitting parameter is greater than or equal to the preset risk threshold, it is determined that the OPC model is in an overfitting state; When the fitting parameter is less than the preset risk threshold, it is determined that the OPC model is not in an overfitting state.

[0061] It is worth noting that in the embodiment of the above-mentioned device, the modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present application.

[0062] Figure 8The structural diagram of the detection device provided for an embodiment of the present application is used to execute the OPC model overfitting detection method provided in the above embodiment, and has the functional modules and beneficial effects corresponding to the execution method. As shown in the figure, the detection device includes a processor 501, a memory 502, an input device 503 and an output device 504. The number of processors 501 can be one or more, and one processor 501 is taken as an example in the figure; the processor 501, the memory 502, the input device 503 and the output device 504 can be connected by a bus or other means, and the figure takes the connection through a bus as an example. The memory 502, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the OPC model overfitting detection method in the embodiment of the present application. The processor 501 executes the corresponding various functional applications and data processing by running the software programs, instructions and modules stored in the memory 502, that is, the above-mentioned OPC model overfitting detection method is realized.

[0063] The memory 502 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data recorded or created during use, etc. In addition, the memory 502 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 502 may further include a memory remotely arranged relative to the processor 501, and these remotely arranged memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0064] The input device 503 can be used to input corresponding digital or character information to the processor 501, and to generate key signal input related to the user settings and function control of the device; the output device 504 can be used to send or display key signal output related to the user settings and function control of the device.

[0065] An embodiment of the present application also provides a storage medium storing computer executable instructions, which, when executed by a processor, are used to perform relevant operations in the OPC model overfitting detection method provided in any embodiment of the present application.

[0066] Computer-readable storage media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0067] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0068] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A method for detecting overfitting of an OPC model, characterized in that: include: Acquire model images corresponding to a plurality of target graphics through the OPC model, and set a segmentation line on the model image corresponding to each of the target graphics; Based on the OPC model, light intensity calculation is performed on the sampling points on the segmentation line to determine the light intensity value corresponding to each sampling point; Determine the light intensity parameter corresponding to each target graphic according to the light intensity value corresponding to each sampling point; Determine the fitting parameters corresponding to the OPC model according to the light intensity parameters and the corresponding weights corresponding to each target graphic; It is determined whether the OPC model is in an overfitting state according to the fitting parameters.

2. The OPC model overfitting detection method according to claim 1, characterized in that: The OPC model includes an optical model and a photoresist model, wherein the optical model is used to generate a projection image corresponding to the target pattern, and the photoresist model is used to generate a photoresist contour image corresponding to the target pattern; The method of obtaining model images corresponding to a plurality of target graphics through the OPC model and setting a segmentation line on each of the target graphics includes: Determine the center point of each target graphic; Generate a projection image and a photoresist contour image corresponding to each target pattern, and set a segmentation line passing through the projection image and the photoresist contour image corresponding to each target pattern, and the center point of the target pattern is located on the segmentation line; A plurality of sampling points are arranged on the dividing line according to a preset interval.

3. The OPC model overfitting detection method according to claim 2, characterized in that: The step of performing light intensity calculation on the sampling points on the segmentation line based on the OPC model to determine the light intensity value corresponding to each sampling point includes: Determine a first light intensity value corresponding to a sampling point on the segmentation line according to a projection image formed by an optical model in the OPC model; According to the photoresist contour image formed by the photoresist model in the OPC model, a second light intensity value corresponding to the sampling point on the segmentation line is determined.

4. The OPC model overfitting detection method according to claim 2 or 3, characterized in that: The optical model is a TCC model constructed based on Hopkins optical imaging theory.

5. The OPC model overfitting detection method according to claim 1 or 2, characterized in that: The OPC model includes an optical model and a photoresist model. The light intensity parameter corresponding to each target pattern is determined according to the light intensity value corresponding to each sampling point, including: Determine a light intensity difference according to a first light intensity value and a second light intensity value corresponding to the same target sampling point, wherein the first light intensity value is a light intensity value corresponding to the target sampling point determined based on the optical model, and the second light intensity value is a light intensity value corresponding to the target sampling point determined based on the photoresist model; Determine a first cumulative sum of light intensity differences corresponding to all target sampling points, and a second cumulative sum of first light intensity values ​​corresponding to all target sampling points; Based on the values ​​of the first cumulative sum and the second cumulative sum, a ratio of the first cumulative sum to the second cumulative sum is determined as a light intensity parameter corresponding to the current target pattern.

6. The OPC model overfitting detection method according to claim 1, characterized in that: Determining the fitting parameters corresponding to the OPC model according to the light intensity parameters and the corresponding weights corresponding to each target graphic includes: Determine the weight corresponding to the target graphic according to the number of graphics of the same type as the target graphic among all graphics and the importance value, wherein the importance value is associated with the type of the target graphic and the layout layer where the target graphic is located; According to the light intensity parameter and the corresponding weight corresponding to each target graphic, the corresponding weighted average value is calculated, and the weighted average value is used as the fitting parameter.

7. The OPC model overfitting detection method according to claim 1 or 6, characterized in that: Determining whether the OPC model is in an overfitting state according to the fitting parameters includes: When the fitting parameter is greater than or equal to a preset risk threshold, determining that the OPC model is in an overfitting state; When the fitting parameter is less than the preset risk threshold, it is determined that the OPC model is not in an overfitting state.

8. An OPC model overfitting detection device, characterized in that: include: An image acquisition module is configured to acquire model images corresponding to a plurality of target graphics through the OPC model, and set a segmentation line on each of the target graphics; A light intensity calculation module is configured to perform light intensity calculations on the sampling points on the dividing line based on the OPC model to determine the light intensity value corresponding to each sampling point; A parameter determination module, configured to determine the light intensity parameter corresponding to each of the target graphics according to the light intensity value corresponding to each sampling point; A fitting detection module, configured to determine fitting parameters corresponding to the OPC model according to light intensity parameters and corresponding weights corresponding to each target graphic; The overfitting judgment module is configured to determine whether the OPC model is in an overfitting state based on the fitting parameters.

9. A detection device, characterized in that: include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the OPC model overfitting detection method as described in any one of claims 1-7.

10. A storage medium storing computer executable instructions, characterized in that: When the computer executable instructions are executed by a processor, they are used to perform the OPC model overfitting detection method according to any one of claims 1 to 7.

Images