A feature region recognition method and system based on atomic force microscope topography
By using an automated AFM morphology data processing method, batch and high-precision identification of characteristic regions in nanofabrication was achieved, solving the problems of large errors and low efficiency in manual measurement in existing technologies, and improving the accuracy and consistency of process parameters.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA COAL SCIENCE & TECHNOLOGY (TIANJIN) ROCK FORMATION INTELLIGENT CONTROL TECHNOLOGY CO LTD
- Filing Date
- 2026-02-12
- Publication Date
- 2026-05-29
Smart Images
Figure CN122108018A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of nanofabrication feature region identification technology, and in particular to a method and system for feature region identification based on atomic force microscopy morphology. Background Technology
[0002] Scanning probe direct writing fabrication (such as scanning probe mechanical etching) is a cost-effective method for achieving high-resolution, customized patterns at the nanoscale. In this technique, the quality of the nanofabrication (e.g., feature size) is significantly affected by various process parameters (e.g., normal force, amplitude, fabrication speed). To optimize these parameters, accurate evaluation of the fabrication results is necessary, typically achieved through atomic force microscopy (AFM) characterization of the post-fabrication morphology. However, existing evaluation methods mainly rely on manual measurement, i.e., selecting one or more cross-sections to measure feature size. This method is time-consuming, subjective, and inaccurate. Since nanofabrication results often vary at different locations, manual measurement struggles to achieve a global and unified evaluation, resulting in a lack of clear direction for process parameter optimization and hindering the development of nanofabrication technology. Therefore, there is an urgent need to propose an automatic, global, and accurate AFM method for identifying morphological feature regions to establish a unified evaluation basis for nanofabrication results. Summary of the Invention
[0003] This application provides a method and system for identifying feature regions based on atomic force microscopy (AFM) morphology, which at least solves the technical problems of large subjective errors, low processing efficiency, inability to perform batch global analysis, and disconnect between identification results and process parameters caused by the reliance on manual feature identification of AFM morphology in the prior art.
[0004] The first aspect of this application proposes a method for identifying feature regions based on atomic force microscopy morphology, the method comprising:
[0005] AFM morphology data matrix of the nano-fabricated region is obtained, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; The AFM morphology data matrix is divided into multiple sub-regions corresponding to each prepared structure based on the number of prepared structures. Perform a region localization step for each sub-region to determine the start and end range of the characteristic regions representing nanostructures in the vertical direction within each sub-region; A centerline fitting step is performed within the start and end range of the feature region. By fitting the feature points in multiple rows of data, a feature centerline characterizing the extension direction of the nanostructure is obtained. Along the feature center line, perform a line-by-line recognition operation on multiple rows of topographic data within the feature region to identify the feature region boundary points in each row of data.
[0006] Preferably, the area positioning step includes: Starting from the middle row of the sub-region, perform row scanning upwards and downwards respectively; While scanning each row, calculate the difference between the maximum and minimum values of the current row's topography data; The difference is compared with a preset threshold, which is determined based on the morphological statistics of the non-etched region. Based on the comparison results, the starting and ending row positions of the feature region are determined and marked.
[0007] Furthermore, the centerline fitting step includes: Within the start and end range of the feature region, predetermined feature points are extracted from each row of morphological data. The predetermined feature points include the minimum or maximum value points in the current row of data. A first linear fit is performed on the coordinates of the predetermined feature points extracted from each row; Remove outlier feature points whose deviation from the first fitting result is greater than a preset range; A second linear fit is performed using the coordinates of the proposed feature points, and the straight line obtained from the second fit is taken as the feature center line.
[0008] Furthermore, the step of performing line-by-line recognition of multiple rows of morphological data within the feature region along the feature center line includes: For each row of morphological data, determine the dominant morphological features of each row; If it is determined that etching is dominant, then the left boundary point and the right boundary point of the etching region in the current row of data are identified; If the data is determined to be dominated by stacking, then the left and right boundary points of the stacking region in the current row of data are identified.
[0009] Furthermore, identifying the left and right boundary points of the etched region in the current row of data includes: Based on the deepest etch point and the highest accumulation point in the current row of data, determine the left and right analysis ranges; Linear fitting was performed on the two data segments representing the substrate background and etching transition trend, respectively, within the analysis range on the left and the analysis range on the right. The etching boundary points are identified by detecting the points where the sign of the difference between the two fitted lines changes.
[0010] Furthermore, identifying the left boundary point and the right boundary point of the stacked region in the current row of data includes: Using the etched region boundary points already identified in the current row of data as the initial boundary, identify the point with the highest accumulation. Nonlinear curve fitting was performed on the data segments to the left and right of the highest point of accumulation, respectively. Calculate the intersection points of the nonlinear fitting curves on each side and the baseline height line; Based on the validity verification results of the intersection points, the final boundary points of the accumulation area are identified.
[0011] Furthermore, the method also includes: Acquire cantilever beam deflection signal data corresponding to the nanofabrication region; Based on the cantilever beam deflection signal data, the real-time process parameters corresponding to the boundary points of each identified feature region are calculated.
[0012] The calculation of real-time process parameters corresponding to the boundary points of each feature region based on the cantilever beam deflection signal data includes: Based on the row coordinates or timestamps of the feature boundary points in the AFM topography data matrix, the corresponding signal values are extracted from the synchronously acquired cantilever beam deflection signal data. According to the formula Calculate the normal force ,in, The probe spring constant is This is the sensitivity coefficient. This represents the deflection signal value of the cantilever beam. The calculated positive pressure is used as a real-time process parameter.
[0013] Furthermore, after performing a line-by-line recognition operation on the multi-row topography data within the feature region along the feature center line to identify the feature region boundary points in each row of data, the process further includes: Check whether the positions of the same type of feature boundary points identified between adjacent rows within the same sub-region are continuous, or whether the changes are within a reasonable threshold. If the check fails, the current row's identification result will be marked as abnormal or removed.
[0014] A second aspect of this application proposes a feature region recognition system based on atomic force microscopy morphology, comprising: The acquisition module is used to acquire the AFM morphology data matrix of the nanofabrication region, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; A partitioning module is used to divide the AFM morphology data matrix into multiple sub-regions corresponding to each prepared structure based on the number of prepared structures. The positioning module is used to perform a region positioning step for each sub-region to determine the start and end range of the characteristic regions representing nanostructures in each sub-region in the vertical direction. The fitting module is used to perform a centerline fitting step within the start and end range of the feature region, and obtain a feature centerline characterizing the extension direction of the nanostructure by fitting feature points in multiple rows of data. The extraction module is used to perform a line-by-line recognition operation on multiple rows of morphological data within the feature region along the feature center line, so as to identify the feature region boundary points in each row of data.
[0015] A third aspect of this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method described in the first aspect.
[0016] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects: This application proposes a method and system for identifying feature regions based on atomic force microscopy (AFM) morphology. The method includes: acquiring an AFM morphology data matrix of a nanofabricated region, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; dividing the AFM morphology data matrix into multiple sub-regions corresponding to each fabricated structure according to the number of fabricated structures; performing a region localization step on each sub-region to determine the start and end range of the feature region representing the nanostructure in the vertical direction within each sub-region; performing a centerline fitting step within the start and end range of the feature region, obtaining a feature centerline representing the extension direction of the nanostructure by fitting feature points in multiple rows of data; and performing a row-by-row identification operation along the feature centerline on multiple rows of morphology data within the feature region to identify the boundary points of the feature region in each row of data. The technical solution proposed in this application achieves batch, automatic, and high-precision identification of multiple nanostructure feature regions in complex AFM morphologies.
[0017] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0018] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a method for identifying feature regions based on atomic force microscopy morphology according to an embodiment of this application; Figure 2 This is a detailed flowchart of a method for identifying feature regions based on atomic force microscopy morphology according to an embodiment of this application; Figure 3 This is a structural diagram of a feature region identification system based on atomic force microscopy morphology according to an embodiment of this application. Detailed Implementation
[0019] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0020] This application proposes a method and system for identifying feature regions based on atomic force microscopy (AFM) morphology. The method includes: acquiring an AFM morphology data matrix of a nanofabricated region, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; dividing the AFM morphology data matrix into multiple sub-regions corresponding to each fabricated structure according to the number of fabricated structures; performing a region localization step on each sub-region to determine the start and end range of the feature region representing the nanostructure in the vertical direction within each sub-region; performing a centerline fitting step within the start and end range of the feature region, obtaining a feature centerline representing the extension direction of the nanostructure by fitting feature points in multiple rows of data; and performing a row-by-row identification operation along the feature centerline on multiple rows of morphology data within the feature region to identify the boundary points of the feature region in each row of data. The technical solution proposed in this application achieves batch, automatic, and high-precision identification of multiple nanostructure feature regions in complex AFM morphologies.
[0021] The following description, with reference to the accompanying drawings, illustrates a method and system for identifying feature regions based on atomic force microscopy morphology, according to an embodiment of this application.
[0022] Example 1 Figure 1 This is a flowchart illustrating a method for identifying feature regions based on atomic force microscopy morphology according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes: Step 1: Obtain the AFM morphology data matrix of the nanofabrication region, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; Step 2: Divide the AFM morphology data matrix into multiple sub-regions corresponding to each prepared structure according to the number of prepared structures; Step 3: Perform a region localization step for each sub-region to determine the start and end range of the characteristic regions representing the nanostructure in the vertical direction within each sub-region; In this embodiment of the disclosure, the area positioning step includes: Starting from the middle row of the sub-region, perform row scanning upwards and downwards respectively; While scanning each row, calculate the difference between the maximum and minimum values of the current row's topography data; The difference is compared with a preset threshold, which is determined based on the morphological statistics of the non-etched region. Based on the comparison results, the starting and ending row positions of the feature region are determined and marked.
[0023] Step 4: Perform a centerline fitting step within the start and end range of the feature region. By fitting the feature points in multiple rows of data, a feature centerline characterizing the extension direction of the nanostructure is obtained. In this embodiment of the disclosure, the centerline fitting step includes: Within the start and end range of the feature region, predetermined feature points are extracted from each row of morphological data. The predetermined feature points include the minimum or maximum value points in the current row of data. A first linear fit is performed on the coordinates of the predetermined feature points extracted from each row; Remove outlier feature points whose deviation from the first fitting result is greater than a preset range; A second linear fit is performed using the coordinates of the proposed feature points, and the straight line obtained from the second fit is taken as the feature center line.
[0024] Step 5: Perform a line-by-line recognition operation on the multiple rows of topographic data within the feature region along the feature center line to identify the feature region boundary points in each row of data.
[0025] In this embodiment of the disclosure, the step of performing line-by-line recognition of multiple rows of topographic data within the feature region along the feature center line includes: For each row of morphological data, determine the dominant morphological features of each row; If it is determined that etching is dominant, then the left boundary point and the right boundary point of the etching region in the current row of data are identified; If the data is determined to be dominated by stacking, then the left and right boundary points of the stacking region in the current row of data are identified.
[0026] The identification of the left and right boundary points of the etched region in the current row of data includes: Based on the deepest etch point and the highest accumulation point in the current row of data, determine the left and right analysis ranges; Linear fitting was performed on the two data segments representing the substrate background and etching transition trend, respectively, within the analysis range on the left and the analysis range on the right. The etching boundary points are identified by detecting the points where the sign of the difference between the two fitted lines changes.
[0027] The process of identifying the left and right boundary points of the stacked region in the current row of data includes: Using the etched region boundary points already identified in the current row of data as the initial boundary, identify the point with the highest accumulation. Nonlinear curve fitting was performed on the data segments to the left and right of the highest point of accumulation, respectively. Calculate the intersection points of the nonlinear fitting curves on each side and the baseline height line; Based on the validity verification results of the intersection points, the final boundary points of the accumulation area are identified.
[0028] In this embodiment of the disclosure, the method further includes: Acquire cantilever beam deflection signal data corresponding to the nanofabrication region; Based on the cantilever beam deflection signal data, the real-time process parameters corresponding to the boundary points of each identified feature region are calculated.
[0029] The calculation of real-time process parameters corresponding to the boundary points of each feature region based on the cantilever beam deflection signal data includes: Based on the row coordinates or timestamps of the feature boundary points in the AFM topography data matrix, the corresponding signal values are extracted from the synchronously acquired cantilever beam deflection signal data. According to the formula Calculate the normal force ,in, The probe spring constant is This is the sensitivity coefficient. This represents the deflection signal value of the cantilever beam. The calculated positive pressure is used as a real-time process parameter.
[0030] In this embodiment of the disclosure, after performing a line-by-line recognition operation on multiple rows of topographic data within the feature region along the feature center line to identify the feature region boundary points in each row of data, the method further includes: Check whether the positions of the same type of feature boundary points identified between adjacent rows within the same sub-region are continuous, or whether the changes are within a reasonable threshold. If the check fails, the current row's identification result will be marked as abnormal or removed.
[0031] like Figure 2 As shown, the detailed process of the feature region identification method based on atomic force microscopy morphology is as follows: I. Preliminary Preparations and Data Input Before starting the recognition process, data collection and preparation must be completed, and the core input data must be clearly defined: Collect core data, including a complete AFM morphology data matrix of the nanofabrication region (with dimensions of m×n, comprehensively characterizing the surface morphology), DFL (cantilever beam deflection) signal data related to the fabrication parameters (used for subsequent calculation of normal pressure), and the number of nanostructures contained in the AFM morphology (i.e., the number of fabrications N).
[0032] Perform preliminary validation on the input data to ensure that the data format is standardized and there are no obvious missing or disordered data, laying the foundation for subsequent processing.
[0033] II. Data Division The input data is divided according to the number of fabrications N to enable independent analysis of individual nanostructures: The AFM topography data matrix is divided into N sub-matrices (denoted as AFM). i (i=1,2,…,N), each submatrix corresponds to the morphology data of a nanostructure fabrication.
[0034] Synchronously divide the DFL signal into N segments (denoted as DFL). i This ensures that each signal segment is matched one-to-one with its corresponding submatrix (i.e., a single nanostructure), guaranteeing the accuracy of the correlation between subsequent process parameters and feature regions.
[0035] III. Vertical positioning of feature regions (region positioning steps) For each submatrix AFM i To accurately determine the start and end range of the feature region in the vertical direction (row direction), the specific process is as follows: Starting from the middle row of the submatrix (j=[m / 2]), scan row by row upwards (row number decreasing by j) and downwards (row number increasing by j).
[0036] During the scanning process, the difference between the maximum and minimum values of each row of topographic data is calculated, and this difference is compared with a preset threshold (the preset threshold is determined in advance based on the topographic statistics of the non-etched area).
[0037] When scanning upwards, if the difference in a certain row is less than a preset threshold, then that row is marked as the starting row position y of the feature region. begin Similarly, when scanning downwards, mark the end row position y of the feature region. end .
[0038] By following the steps above, the vertical row range (y) of the feature region is determined. begin To y end This completes the initial localization of the feature region.
[0039] IV. Feature Centerline Fitting (Centerline Fitting Steps) Based on the located feature regions, a feature center line representing the extension direction of the nanostructure is obtained through fitting, which improves the accuracy of subsequent line-by-line identification: In the row range (y) of the feature region begin To y end Within each row of topographic data, a predetermined feature point is extracted. If the region is dominated by etching, the minimum point (corresponding to the candidate point at the deepest etching point) is extracted. If the region is dominated by accumulation, the maximum point (corresponding to the candidate point at the highest accumulation point) is extracted.
[0040] The coordinates of all extracted predetermined feature points are subjected to a first linear fit to obtain a preliminary fitted line.
[0041] Detect and remove outlier feature points that deviate from the first fitting result by more than a preset range (these points are often affected by noise, which affects the fitting accuracy).
[0042] A second linear fitting is performed using the coordinates of the feature points after removing outliers, and the resulting straight line is determined as the feature center line, which accurately characterizes the extension direction of the nanostructure.
[0043] V. Line-by-line feature region boundary recognition (line-by-line recognition operation) Along the feature center line, a detailed analysis is performed on each row of morphological data within the feature region to identify the boundary points of various feature regions. The specific process is as follows: Determine the dominant morphological features of the current row line by line, that is, distinguish whether it is dominated by etching or by accumulation.
[0044] If it is determined to be etching-dominant: Based on the deepest point of etching (determined by the feature point associated with the feature center line) and the highest point of accumulation in the current row of data, the left and right analysis ranges are defined.
[0045] Within the analysis ranges on the left and right sides, linear fitting is performed on two data segments representing different morphological trends.
[0046] By detecting the intersection of two fitted lines, the left and right boundary points of the etched region in the current row are determined; at the same time, the width at half the etch depth is calculated to identify the half-width region.
[0047] If it is determined that accumulation is the dominant factor: Using the boundary points of the etched area already identified in the current row as the initial boundary, locate the highest point of the accumulation.
[0048] Nonlinear curve fitting was performed on the data segments to the left and right of the highest point of accumulation, respectively.
[0049] Calculate the intersection points of the nonlinear fitting curves on each side with the baseline height line, and determine the final left and right boundary points of the accumulation area based on the validity of the intersection points (such as whether they are within a reasonable data range).
[0050] VI. Correlation of Process Parameters The identified feature regions are correlated with real-time process parameters to provide data support for subsequent process optimization. Based on the row coordinates or corresponding timestamps of the feature region boundary points in the AFM topography data matrix, from the synchronously acquired DFL signal segments (DFL... i Extract the corresponding signal value from ).
[0051] According to the formula Calculate the real-time positive pressure.
[0052] The calculated positive pressure is used as the real-time process parameter corresponding to the feature region, and the correspondence between the feature region and the process parameter is established.
[0053] VII. Result Verification and Anomaly Handling The recognition results are rigorously verified to ensure data reliability. The specific steps are as follows: Check whether the positions of boundary points of the same type of feature (such as etched left boundary points, stacked right boundary points, etc.) are continuous between adjacent rows in the same sub-region, or whether the position changes are within a preset reasonable threshold.
[0054] If the check passes, the recognition result of the feature region is confirmed to be valid. Record the feature location information (including the boundary of the etched area, the location of the accumulation area, the half-width area, the coordinates of the deepest etched area, etc.) and the corresponding real-time process parameters.
[0055] If the check fails (e.g., abrupt changes in the location of boundary points or exceeding a reasonable range), the identification result of the current row will be marked as abnormal or directly removed to avoid abnormal data affecting the overall evaluation result.
[0056] VIII. Iterative Processing and Global Output Repeat steps three through seven above to identify the feature regions of all N sub-matrices (i.e., all nanofabricated structures) one by one to ensure that no one is missed.
[0057] After all submatrices have been processed, all valid recognition results are integrated to form the feature region information of the global AFM morphology, completing the entire recognition process and outputting the final result.
[0058] In summary, the feature region identification method based on atomic force microscopy morphology proposed in this embodiment achieves batch, automatic, and high-precision identification of multiple nanostructure feature regions in complex AFM morphologies.
[0059] Example 2 Figure 3 This is a structural diagram of a feature region recognition system based on atomic force microscopy morphology according to an embodiment of this application, as shown below. Figure 3 As shown, the system includes: The acquisition module 100 is used to acquire the AFM morphology data matrix of the nano-prepared region, wherein the AFM morphology data matrix contains AFM morphology data of multiple prepared structures. The partitioning module 200 is used to divide the AFM morphology data matrix into multiple sub-regions corresponding to each prepared structure according to the number of prepared structures. The positioning module 300 is used to perform a region positioning step for each sub-region to determine the start and end range of the characteristic region representing the nanostructure in the vertical direction in each sub-region. The fitting module 400 is used to perform a centerline fitting step within the start and end range of the feature region, and obtain a feature centerline characterizing the extension direction of the nanostructure by fitting feature points in multiple rows of data. The extraction module 500 is used to perform a line-by-line recognition operation on multiple rows of topographic data within the feature region along the feature center line, so as to identify the feature region boundary points in each row of data.
[0060] In this embodiment of the disclosure, the positioning module 300 is further configured to: Starting from the middle row of the sub-region, perform row scanning upwards and downwards respectively; While scanning each row, calculate the difference between the maximum and minimum values of the current row's topography data; The difference is compared with a preset threshold, which is determined based on the morphological statistics of the non-etched region. Based on the comparison results, the starting and ending row positions of the feature region are determined and marked.
[0061] In this embodiment of the disclosure, the fitting module 400 is further configured to: Within the start and end range of the feature region, predetermined feature points are extracted from each row of morphological data. The predetermined feature points include the minimum or maximum value points in the current row of data. A first linear fit is performed on the coordinates of the predetermined feature points extracted from each row; Remove outlier feature points whose deviation from the first fitting result is greater than a preset range; A second linear fit is performed using the coordinates of the proposed feature points, and the straight line obtained from the second fit is taken as the feature center line.
[0062] In this embodiment of the disclosure, the extraction module 500 is further configured to: For each row of morphological data, determine the dominant morphological features of each row; If it is determined that etching is dominant, then the left boundary point and the right boundary point of the etching region in the current row of data are identified; If the data is determined to be dominated by stacking, then the left and right boundary points of the stacking region in the current row of data are identified.
[0063] Furthermore, the extraction module 500 is also used for: Based on the deepest etch point and the highest accumulation point in the current row of data, determine the left and right analysis ranges; Linear fitting was performed on the two data segments representing the substrate background and etching transition trend, respectively, within the analysis range on the left and the analysis range on the right. The etching boundary points are identified by detecting the points where the sign of the difference between the two fitted lines changes.
[0064] Furthermore, the extraction module 500 is also used for: Using the etched region boundary points already identified in the current row of data as the initial boundary, identify the point with the highest accumulation. Nonlinear curve fitting was performed on the data segments to the left and right of the highest point of accumulation, respectively. Calculate the intersection points of the nonlinear fitting curves on each side and the baseline height line; Based on the validity verification results of the intersection points, the final boundary points of the accumulation area are identified.
[0065] In this embodiment of the disclosure, the extraction module 500 is further configured to: Acquire cantilever beam deflection signal data corresponding to the nanofabrication region; Based on the cantilever beam deflection signal data, the real-time process parameters corresponding to the boundary points of each identified feature region are calculated.
[0066] The calculation of real-time process parameters corresponding to the boundary points of each feature region based on the cantilever beam deflection signal data includes: Based on the row coordinates or timestamps of the feature boundary points in the AFM topography data matrix, the corresponding signal values are extracted from the synchronously acquired cantilever beam deflection signal data. According to the formula Calculate the normal force ,in, The probe spring constant is This is the sensitivity coefficient. This represents the deflection signal value of the cantilever beam. The calculated positive pressure is used as a real-time process parameter.
[0067] In this embodiment of the disclosure, the extraction module 500 is further configured to: Check whether the positions of the same type of feature boundary points identified between adjacent rows within the same sub-region are continuous, or whether the changes are within a reasonable threshold. If the check fails, the current row's identification result will be marked as abnormal or removed.
[0068] In summary, the feature region recognition system based on atomic force microscopy morphology proposed in this embodiment realizes batch, automatic, and high-precision recognition of multiple nanostructure feature regions in complex AFM morphologies.
[0069] Example 3 To implement the above embodiments, this disclosure also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in Embodiment 1.
[0070] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0071] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0072] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.
Claims
1. A method for identifying feature regions based on atomic force microscopy morphology, characterized in that, The method includes: AFM morphology data matrix of the nanofabrication region is obtained, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; The AFM morphology data matrix is divided into multiple sub-regions corresponding to each prepared structure based on the number of prepared structures. Perform a region localization step for each sub-region to determine the start and end range of the characteristic regions representing nanostructures in each sub-region in the vertical direction; A centerline fitting step is performed within the start and end range of the feature region. By fitting the feature points in multiple rows of data, a feature centerline characterizing the extension direction of the nanostructure is obtained. Along the feature center line, perform a line-by-line recognition operation on multiple rows of topographic data within the feature region to identify the feature region boundary points in each row of data.
2. The method as described in claim 1, characterized in that, The area positioning steps include: Starting from the middle row of the sub-region, perform row scanning upwards and downwards respectively; While scanning each row, calculate the difference between the maximum and minimum values of the current row's topography data; The difference is compared with a preset threshold, which is determined based on the morphological statistics of the non-etched region. Based on the comparison results, the starting and ending row positions of the feature region are determined and marked.
3. The method as described in claim 2, characterized in that, The centerline fitting step includes: Within the start and end range of the feature region, predetermined feature points are extracted from each row of morphological data. The predetermined feature points include the minimum or maximum value points in the current row of data. A first linear fit is performed on the coordinates of the predetermined feature points extracted from each row; Remove outlier feature points whose deviation from the first fitting result is greater than a preset range; A second linear fit is performed using the coordinates of the proposed feature points, and the straight line obtained from the second fit is taken as the feature center line.
4. The method as described in claim 3, characterized in that, The step of performing line-by-line recognition on multiple rows of topographic data within the feature region along the feature center line includes: For each row of morphological data, determine the dominant morphological features of each row; If it is determined that etching is dominant, then the left boundary point and the right boundary point of the etching region in the current row of data are identified; If the data is determined to be dominated by stacking, then the left and right boundary points of the stacking region in the current row of data are identified.
5. The method as described in claim 4, characterized in that, The process of identifying the left and right boundary points of the etched region in the current row of data includes: Based on the deepest etch point and the highest accumulation point in the current row of data, determine the left and right analysis ranges; Linear fitting was performed on the two data segments representing the substrate background and etching transition trend, respectively, within the analysis range on the left and the analysis range on the right. The etching boundary points are identified by detecting the points where the sign of the difference between the two fitted lines changes.
6. The method as described in claim 5, characterized in that, The process of identifying the left and right boundary points of the stacked region in the current row of data includes: Using the etched region boundary points already identified in the current row of data as the initial boundary, identify the point with the highest accumulation. Nonlinear curve fitting was performed on the data segments to the left and right of the highest point of accumulation, respectively. Calculate the intersection points of the nonlinear fitting curves on each side and the baseline height line; Based on the validity verification results of the intersection points, the final boundary points of the accumulation area are identified.
7. The method as described in claim 6, characterized in that, The method further includes: Acquire cantilever beam deflection signal data corresponding to the nanofabrication region; Based on the cantilever beam deflection signal data, the real-time process parameters corresponding to the boundary points of each identified feature region are calculated. The calculation of real-time process parameters corresponding to the boundary points of each feature region based on the cantilever beam deflection signal data includes: Based on the row coordinates or timestamps of the feature boundary points in the AFM topography data matrix, the corresponding signal values are extracted from the synchronously acquired cantilever beam deflection signal data. According to the formula Calculate the normal force ,in, The probe spring constant is This is the sensitivity coefficient. This represents the deflection signal value of the cantilever beam. The calculated positive pressure is used as a real-time process parameter.
8. The method as described in claim 7, characterized in that, After performing a line-by-line recognition operation on multiple rows of topographic data within the feature region along the feature center line to identify the feature region boundary points in each row of data, the process further includes: Check whether the positions of the same type of feature boundary points identified between adjacent rows within the same sub-region are continuous, or whether the changes are within a reasonable threshold. If the check fails, the current row's identification result will be marked as abnormal or removed.
9. A feature region recognition system based on atomic force microscopy morphology, based on the feature region recognition method for atomic force microscopy morphology as described in any one of claims 1-8, characterized in that, The system includes: The acquisition module is used to acquire the AFM morphology data matrix of the nanofabrication region, wherein the AFM morphology data matrix contains AFM morphology data of multiple fabricated structures; A partitioning module is used to divide the AFM morphology data matrix into multiple sub-regions corresponding to each prepared structure based on the number of prepared structures. The positioning module is used to perform a region positioning step for each sub-region to determine the start and end range of the characteristic regions representing nanostructures in each sub-region in the vertical direction. The fitting module is used to perform a centerline fitting step within the start and end range of the feature region, and obtain a feature centerline characterizing the extension direction of the nanostructure by fitting feature points in multiple rows of data. The extraction module is used to perform a line-by-line recognition operation on multiple rows of morphological data within the feature region along the feature center line, so as to identify the feature region boundary points in each row of data.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-8.