Method suitable for field curvature automatic detection of microscope objective

Abstract

The invention relates to the technical field of measurement and control, and discloses a method for automatically detecting field curvature of a microscope objective, which comprises the following steps of: reading a microscopic image with a standard pattern, calculating an acute angle between a straight line segment and a horizontal line in the microscopic image, adjusting a rotation angle of a microscope corresponding to a mobile platform, and calculating resolution distribution of the microscopic image, calculating the camera visual center of the microscopic image, determining the pattern center of the microscopic image, calculating the pixel coordinate difference between the camera visual angle center and the pattern center, adjusting the displacement of the mobile platform, positioning the region of interest of the microscopic image, and determining the sampling point of the region of interest; determining a local meridian direction and a local sagittal direction of the sampling point to construct a local dynamic coordinate system of the sampling point, and analyzing a meridian magnification and a sagittal magnification of the sampling point; and fitting the three-dimensional imaging curved surface of the microscopic image, and determining the field curvature of the objective lens corresponding to the microscope. According to the invention, the accuracy and efficiency of field curvature automatic detection of the microscope objective can be improved.

Description

Technical Field

[0001] The invention relates to a method suitable for automatic detection of field curvature of a microscope objective lens, belonging to the technical field of measurement and control. Background Art

[0002] Microscope field curvature refers to the phenomenon where the image plane of a microscope objective is not an ideal flat plane perpendicular to the optical axis, but rather a curved surface. Testing microscope field curvature provides an intuitive understanding of the uniformity and clarity of the image within the objective's field of view, allowing us to screen out substandard products and ensure the stable performance of the manufactured objective.

[0003] The traditional method for detecting field curvature of a microscope objective is the resolution plate method. This method uses a pattern with a known spatial frequency (such as a resolution plate, grid, or mesh) as a target and determines the field curvature by observing its image clarity at different image plane positions. This method still has a certain subjective judgment component when reading and recording the clearest position or highest resolution, making it difficult to quantify. It can only qualitatively determine whether field curvature exists and the approximate direction and degree of curvature, resulting in low accuracy and poor results in detecting field curvature of a microscope objective. Summary of the Invention

[0004] The present invention provides a method for automatic detection of field curvature of a microscope objective lens, the main purpose of which is to improve the accuracy and efficiency of automatic detection of field curvature of a microscope objective lens.

[0005] To achieve the above objectives, the present invention provides a method for automatically detecting field curvature of a microscope objective lens, comprising:

[0006] reading a microscopic image with a standard pattern, the microscopic image being read through a microscope, calculating an acute angle between a straight line segment and a horizontal line in the microscopic image, and when the acute angle is not zero, adjusting a rotation angle of a corresponding movable platform of the microscope, and returning to the step of reading the microscopic image with the standard pattern;

[0007] When the acute angle is zero, calculating the resolution distribution of the microscopic image to calculate the camera visual center of the microscopic image, extracting morphological features of the microscopic image to determine the pattern center of the microscopic image, calculating the pixel coordinate difference between the camera visual angle center and the pattern center, and when the pixel coordinate difference is not zero, adjusting the displacement of the mobile platform and returning to the step of extracting the morphological features of the microscopic image;

[0008] When the pixel coordinate difference is zero, locating the region of interest of the microscopic image and determining a sampling point of the region of interest;

[0009] Determining the local meridian direction and the local sagittal direction of the sampling point to construct a local dynamic coordinate system of the sampling point, and analyzing the meridian magnification and sagittal magnification of the sampling point based on the sampling point, the standard pattern, and the local dynamic coordinate system;

[0010] According to the meridional magnification, the sagittal magnification and the resolution distribution, the three-dimensional imaging surface of the microscopic image is fitted, the full field curvature distribution map of the three-dimensional imaging surface is analyzed, and based on the full field curvature distribution map, the field curvature of the corresponding objective lens of the microscope is determined.

[0011] Optionally, calculating the acute angle between the straight line segment and the horizontal line in the microscopic image includes:

[0012] performing binarization processing on the microscopic image to obtain a binarized image;

[0013] Performing edge detection on the binary image to obtain image edges;

[0014] Based on the image edges, detecting straight line segments of the binary image, and determining a target straight line segment among the straight line segments;

[0015] Determine the pixel coordinates of the starting point and the ending point of the target straight line segment;

[0016] The acute angle between the target straight line segment and the horizontal line is calculated according to the starting pixel coordinates and the ending pixel coordinates.

[0017] Optionally, calculating the resolution distribution of the microscopic image includes:

[0018] dividing the microscopic image into multiple image regions on an even basis;

[0019] Calculating horizontal gradients and vertical gradients of pixels within the multi-image area;

[0020] determining a gradient magnitude of the pixel according to the horizontal gradient and the vertical gradient;

[0021] Calculating the gradient magnitude mean and the gradient magnitude mean variance of the multiple image regions;

[0022] The resolution of the multi-image region is determined according to the gradient amplitude mean and the gradient amplitude mean variance, so as to determine the resolution distribution of the microscopic image.

[0023] Optionally, extracting morphological features of the microscopic image includes:

[0024] performing image enhancement on the microscopic image to obtain an enhanced microscopic image;

[0025] performing threshold segmentation on the enhanced microscopic image to obtain a segmented microscopic image;

[0026] performing contour extraction on the segmented microscopic image to obtain a pattern contour;

[0027] extracting shape features and spatial distribution features of the segmented microscopic image;

[0028] The morphological features of the microscopic image are determined according to the pattern outline, the shape features, and the spatial distribution features.

[0029] Optionally, determining the sampling points of the region of interest includes:

[0030] performing contrast enhancement on the region of interest to obtain an enhanced region of interest;

[0031] Binarizing the enhanced region of interest to obtain a binarized region of interest;

[0032] Performing morphological operations on the binarized region of interest to obtain an outline of a pattern of interest;

[0033] Calculating the total length of the contour of the pattern of interest and detecting contour features of the contour of the pattern of interest;

[0034] Sampling points on the contour of the pattern of interest are determined according to the total contour length and the contour features.

[0035] Optionally, determining the local meridian direction and the local sagittal direction of the sampling point includes:

[0036] determining the sampling point coordinates of the sampling point according to the microscopic image corresponding to the sampling point;

[0037] determining the pattern center coordinates of the microscopic image corresponding to the pattern center;

[0038] Calculating the meridian direction vector of the sampling point according to the sampling point coordinates and the pattern center coordinates;

[0039] Determining the local meridian direction of the sampling point based on the meridian direction vector;

[0040] Determine a 90° rotation matrix of the meridian direction vector, and calculate the sagittal direction vector of the sampling point based on the 90° rotation matrix and the meridian direction vector;

[0041] The local sagittal direction of the sampling point is determined according to the sagittal direction vector.

[0042] Optionally, analyzing the meridian magnification and sagittal magnification of the sampling point according to the sampling point, the standard pattern, and the local dynamic coordinate system includes:

[0043] Mapping the standard points corresponding to the sampling points on the standard pattern, and determining the actual coordinates of the standard points;

[0044] Determining the meridian direction basis vector and the sagittal direction basis vector of the local dynamic coordinate system;

[0045] Calculating the sampling point meridian distance and the sampling point sagittal distance between the sampling points according to the sampling point coordinates corresponding to the sampling points, the meridian direction basis vector, and the sagittal direction basis vector;

[0046] Calculating the standard point meridian direction distance and the standard point sagittal direction distance between the standard points based on the actual coordinates, the meridian direction basis vector and the sagittal direction basis vector;

[0047] Determining the meridian magnification of the sampling point according to the meridian distance of the sampling point and the meridian distance of the standard point;

[0048] The sagittal magnification of the sampling point is determined according to the sagittal distance of the sampling point and the sagittal distance of the standard point.

[0049] Optionally, fitting the three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification, and the resolution distribution includes:

[0050] Normalizing the meridional magnification and the sagittal magnification to obtain a normalized meridional magnification and a normalized sagittal magnification;

[0051] Based on the resolution distribution, identifying the center point resolution and the edge point resolution of the microscopic image;

[0052] Calculating a resolution variation coefficient of the microscopic image according to the center point resolution and the edge point resolution;

[0053] determining a surface fitting algorithm for the microscopic image to construct an initial imaging surface of the microscopic image;

[0054] Determining a surface optimization objective function of the surface fitting algorithm according to the normalized meridional magnification, the normalized sagittal magnification, and the resolution variation coefficient;

[0055] Defining constraints of the surface fitting algorithm;

[0056] Iteratively optimizing the initial imaging surface according to the constraint conditions and the surface optimization objective function to obtain an optimized imaging surface;

[0057] The surface error of the optimized imaging surface is calculated, and when the surface error is less than a preset surface error threshold, the optimized imaging surface is used as the three-dimensional imaging surface of the microscopic image.

[0058] Optionally, normalizing the meridian magnification to obtain a normalized meridian magnification includes:

[0059] calculating a meridional magnification mean and a meridional magnification variance of the meridional magnification,

[0060] The meridional magnification is normalized based on the meridional magnification mean and the meridional magnification standard deviation to obtain a normalized meridional magnification.

[0061] Optionally, analyzing the full-field curvature distribution map of the three-dimensional imaging surface includes:

[0062] determining a surface position function of the three-dimensional imaging surface;

[0063] calculating first and second derivatives of the surface position function;

[0064] constructing a characteristic matrix of the three-dimensional imaging surface based on the first-order derivative and the second-order derivative;

[0065] Calculating the eigenvalues ​​of the characteristic matrix, and calculating the principal curvature of the three-dimensional imaging surface based on the eigenvalues;

[0066] Calculating the mean curvature and Gaussian curvature of the three-dimensional imaging surface according to the principal curvature;

[0067] A full-field curvature distribution map of the three-dimensional imaging surface is determined according to the average curvature, the Gaussian curvature, and the principal curvature.

[0068] Compared with the problems described in the background technology, the embodiment of the present invention can realize accurate automatic rotation calibration of the microscope platform by calculating the acute angles of the straight line segments and the horizontal lines in the microscopic image, thereby improving image quality and measurement accuracy; optionally, the embodiment of the present invention can correct the initial geometric deviation by adjusting the displacement of the mobile platform when the pixel coordinate difference is not zero, and returning to the step of extracting the morphological features of the microscopic image, so that subsequent image acquisition and analysis are based on an accurate and unified spatial reference system, thereby significantly improving the accuracy, stability and automation level of microscopic image analysis; the embodiment of the present invention can significantly reduce the amount of data to be processed by determining the sampling points of the region of interest, thereby improving the analysis accuracy and representativeness; the embodiment of the present invention can significantly reduce the amount of data to be processed by determining the sampling points of the region of interest, thereby improving the analysis accuracy and representativeness; The sample point, the standard pattern, and the local dynamic coordinate system are analyzed, and the meridional magnification and sagittal magnification of the sample point are analyzed to directly calculate the distortion parameters in the local coordinate system, avoiding complex nonlinear transformations in the global coordinate system. The embodiment of the present invention fits the three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification, and the resolution distribution, which can be used to more accurately understand the mapping relationship from the object plane to the image plane, and globally and quantitatively describe the aberration distribution and distortion degree within the entire field of view. Finally, the embodiment of the present invention determines the field curvature of the corresponding objective lens of the microscope based on the full field curvature distribution map, which can accurately measure the size of the field curvature generated by the objective lens, objectively compare the field curvature performance of different objective lens samples, and realize the control of the objective lens production quality. Therefore, the method for automated detection of field curvature of microscope objective lenses provided by the embodiment of the present invention can improve the accuracy and efficiency of automated detection of field curvature of microscope objective lenses. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A schematic flow chart of a method for automatically detecting field curvature of a microscope objective lens provided in one embodiment of the present invention;

[0070] Figure 2 An example diagram of a standard pattern for implementing the method for automated detection of field curvature of a microscope objective lens provided in one embodiment of the present invention;

[0071] Figure 3 A diagram showing the actual effect of a standard pattern for implementing the method for automated detection of field curvature of a microscope objective lens provided in one embodiment of the present invention;

[0072] Figure 4 A schematic diagram of the center point resolution of the method for automatically detecting field curvature of a microscope objective lens provided in one embodiment of the present invention;

[0073] Figure 5 A schematic diagram of edge point resolution for implementing the method for automatic detection of field curvature of a microscope objective lens provided in one embodiment of the present invention;

[0074] Figure 6 A schematic diagram of modules for implementing the method for automatic detection of field curvature of a microscope objective lens provided in one embodiment of the present invention.

[0075] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0076] 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.

[0077] The present embodiment provides a method for automated detection of field curvature of a microscope objective lens. The method for automated detection of field curvature of a microscope objective lens can be executed by at least one of the following electronic devices, including a server and a terminal, that can be configured to execute the method provided in the present embodiment. In other words, the method for automated detection of field curvature of a microscope objective lens can be executed by software or hardware installed on a terminal device or a server device. The server device includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0078] Example 1:

[0079] Reference Figure 1 FIG. 1 is a flow chart of a method for automatically detecting field curvature of a microscope objective lens according to an embodiment of the present invention. In this embodiment, the method for automatically detecting field curvature of a microscope objective lens includes:

[0080] S1. Read a microscopic image with a standard pattern, where the microscopic image is read through a microscope, calculate the acute angle between a straight line segment and a horizontal line in the microscopic image, and when the acute angle is not zero, adjust the rotation angle of the corresponding moving platform of the microscope, and return to the step of reading the microscopic image with the standard pattern.

[0081] By reading a microscopic image with a standard pattern and analyzing the appearance of these features in the image, the embodiments of the present invention can accurately calculate the microscope's magnification, pixel size (pixel equivalent), and correct lens distortion (such as radial distortion and tangential distortion), thereby ensuring the accuracy of subsequent measurements (such as length, area, and spacing). The standard pattern refers to a pattern that is specially designed and manufactured with known precise geometric dimensions and / or specific features and is used to calibrate, test, and verify the performance of a microscope system. The microscopic image refers to a magnified two-dimensional image of a micro standard pattern captured by the microscope system.

[0082] For further information on the standard patterns, see Figure 2As shown in FIG. 1 , an example diagram of a standard pattern for realizing the method for automatically detecting field curvature of a microscope objective lens provided by an embodiment of the present invention is shown. Figure 2 In the process, the precise geometric size and two-dimensional position of the standard pattern can be accurately identified.

[0083] Embodiments of the present invention can achieve precise automatic rotation calibration of a microscope platform by calculating the acute angle between a straight line segment and a horizontal line in the microscopic image, thereby improving image quality and measurement accuracy. The straight line segment refers to a detectable, straight line segment in the microscopic image. The horizontal line refers to a horizontal reference line inherent in the image. The acute angle refers to the inclination angle of a straight line segment (from a standard pattern) in the microscopic image relative to the reference horizontal line.

[0084] As an embodiment of the present invention, the calculating the acute angle between the straight line segment and the horizontal line in the microscopic image includes:

[0085] performing binarization processing on the microscopic image to obtain a binarized image;

[0086] Performing edge detection on the binary image to obtain image edges;

[0087] Based on the image edges, detecting straight line segments of the binary image, and determining a target straight line segment among the straight line segments;

[0088] Determine the pixel coordinates of the starting point and the ending point of the target straight line segment;

[0089] According to the starting pixel coordinates and the ending pixel coordinates, the acute angle between the target straight line segment and the horizontal line is calculated using the following formula:

[0090]

[0091] Wherein, θ represents the acute angle, arctan represents the inverse tangent function, q2 represents the ordinate of the end pixel coordinate, q1 represents the ordinate of the start pixel coordinate, p2 represents the abscissa of the end pixel coordinate, and p1 represents the abscissa of the start pixel coordinate.

[0092] The binary image refers to an image obtained through binarization processing, in which the grayscale values ​​of all pixels are divided into only two types: 0 (black) or 255 (white). The image edge refers to an area in the image where the brightness (grayscale) changes dramatically. The target straight line segment refers to a straight line segment with a nearly horizontal direction that is screened out from all detected straight line segments. The starting pixel coordinates refer to the position of the starting point of the target straight line segment in the image, expressed in pixel coordinates. The end pixel coordinates refer to the position of the end point of the target straight line segment in the image, expressed in pixel coordinates.

[0093] Optionally, the image edge can be obtained by an edge detection algorithm, such as a gradient operator method, a Canny edge detection algorithm, a second-order derivative method, and the like.

[0094] Optionally, the straight line segments of the binary image may be detected by a straight line detection algorithm, such as Hough transform, least squares method, etc.

[0095] Optionally, the starting pixel coordinates and the ending pixel coordinates of the target straight line segment can be determined by constructing a two-dimensional coordinate system of the binary image with the upper left vertex of the binary image as the origin, and determining the starting pixel coordinates and the ending pixel coordinates of the target straight line segment based on the two-dimensional coordinate system.

[0096] In this embodiment of the present invention, when the acute angle is not zero, the rotation angle of the corresponding movable platform of the microscope is adjusted, and the process of returning to the step of reading the microscopic image with the standard pattern is performed. This allows for a closed-loop control process of "detection-calculation-adjustment-redetection," thereby accurately calibrating the platform back to the correct orientation. The rotation angle refers to the specific angle by which the platform must be rotated to calibrate the microscope platform.

[0097] For example, when the acute angle is 35° and indicates that the straight line deviates from the horizontal clockwise, the rotation angle of the mobile platform is -35°, and the mobile platform needs to be rotated 35° counterclockwise. When the acute angle is 35° and the straight line deviates from the horizontal counterclockwise, the rotation angle of the mobile platform is 35°, and the mobile platform needs to be rotated 35° clockwise.

[0098] S2. When the acute angle is zero, calculate the resolution distribution of the microscopic image to calculate the camera visual center of the microscopic image, extract the morphological features of the microscopic image to determine the pattern center of the microscopic image, calculate the pixel coordinate difference between the camera viewing angle center and the pattern center, and when the pixel coordinate difference is not zero, adjust the displacement of the mobile platform and return to the step of extracting the morphological features of the microscopic image.

[0099] In embodiments of the present invention, by calculating the resolution distribution of the microscopic image when the acute angle is zero, the position of the visual center can be more accurately determined, thereby providing a reference for subsequent displacement calibration. The resolution distribution refers to the spatial distribution of detail richness and clarity in different areas of the image.

[0100] As an embodiment of the present invention, the calculating the resolution distribution of the microscopic image includes:

[0101] dividing the microscopic image into multiple image regions on an even basis;

[0102] Calculating horizontal gradients and vertical gradients of pixels within the multi-image area;

[0103] determining a gradient magnitude of the pixel according to the horizontal gradient and the vertical gradient;

[0104] Calculating the gradient magnitude mean and the gradient magnitude mean variance of the multiple image regions;

[0105] The resolution of the multi-image region is determined according to the gradient amplitude mean and the gradient amplitude mean variance, so as to determine the resolution distribution of the microscopic image.

[0106] Among them, the multi-image area refers to a plurality of local image blocks into which the original microscopic image is evenly divided. The horizontal gradient refers to the rate of change of pixel brightness of the image in the horizontal direction (i.e., the x-axis direction). The vertical gradient refers to the rate of change of pixel brightness of the image in the vertical direction (i.e., the y-axis direction). The gradient amplitude refers to the combined intensity of the gradient components of the pixels in the image in the horizontal and vertical directions. The gradient amplitude mean refers to the average value of the gradient amplitudes of all pixels in the local image block in the image. The gradient amplitude mean variance refers to the degree of discreteness between the gradient amplitude means of each image area. The resolution refers to a measure of the edge clarity of the local image block of the image.

[0107] Optionally, the multiple image regions may be divided by a grid division method, such as constructing a 10×10 grid of the microscopic image, and dividing the microscopic image into 100 image regions on average according to the 10×10 grid, thereby obtaining multiple image regions.

[0108] Optionally, the horizontal gradient and the vertical gradient of the pixels in the multi-image area may be calculated by a gradient operator method, such as a Sobel operator, a Prewitt operator, a Roberts operator, and the like.

[0109] Optionally, the gradient amplitude mean and the gradient amplitude mean variance of the multiple image regions can be calculated using the following formula:

[0110]

[0111] in, represents the mean value of gradient amplitude, D represents the mean variance of gradient amplitude, N represents the number of pixels in the multi-image area, F i Represents the gradient magnitude of the i-th pixel in the multi-image region.

[0112] In the embodiment of the present invention, by calculating the camera visual center of the microscopic image, the visual center can be used as a reference point to ensure that the center of the standard pattern is accurately placed in the optimal imaging area and reduce the influence of edge distortion. The camera visual center refers to the optimal imaging point in the microscope system.

[0113] Optionally, the camera visual center of the microscopic image can be determined by identifying a resolution peak in the resolution distribution, determining a peak position of the resolution peak, and determining the camera visual center of the microscopic image based on the peak position.

[0114] The embodiment of the present invention can provide a data basis for subsequent analysis of the pattern center of the microscopic image by extracting the morphological features of the microscopic image, wherein the morphological features refer to visual attributes that describe the shape, size, structure and spatial relationship of the standard pattern.

[0115] As an embodiment of the present invention, extracting the morphological features of the microscopic image includes:

[0116] performing image enhancement on the microscopic image to obtain an enhanced microscopic image;

[0117] performing threshold segmentation on the enhanced microscopic image to obtain a segmented microscopic image;

[0118] performing contour extraction on the segmented microscopic image to obtain a pattern contour;

[0119] extracting shape features and spatial distribution features of the segmented microscopic image;

[0120] The morphological features of the microscopic image are determined according to the pattern outline, the shape features, and the spatial distribution features.

[0121] Among them, the enhanced microscopic image refers to the image obtained after image enhancement processing on the basis of the original microscopic image. The segmented microscopic image refers to the process of dividing the different structures therein into independent areas by analyzing and processing the enhanced microscopic image during the image processing. The pattern contour refers to the boundary lines of each area obtained by edge detection in image processing. These boundary lines define the shape and position of different patterns in the image. The shape feature refers to the quantitative attribute used to describe the geometric shape of the pattern in the image. The spatial distribution feature refers to the quantitative attribute used to describe the arrangement, aggregation or dispersion of the pattern in the two-dimensional space of the image.

[0122] Optionally, the enhanced microscopic image may be enhanced by histogram equalization, such as global histogram equalization, local histogram equalization, contrast-limited adaptive histogram equalization, etc.

[0123] Optionally, the shape features and spatial distribution features can be extracted by shape feature extraction algorithms (such as boundary tracking algorithm, Moore-Neighbor algorithm, Bresenham algorithm) and spatial autocorrelation analysis algorithms (such as Moran's I algorithm, Geary's C algorithm, etc.).

[0124] By determining the center of the pattern in the microscopic image, the embodiment of the present invention can automatically focus, track, and move the stage based on this center point, ensuring that subsequent operations are always aligned with the target. The pattern center refers to the specific position of the standard pattern center in the microscopic image.

[0125] Exemplarily, the microscopic image corresponds to a standard pattern and has four circles for determining the center of the standard pattern. The center positions of the four circles in the microscopic image are determined based on the morphological features of the microscopic image, and the pattern center of the microscopic image is determined based on the center positions.

[0126] In this embodiment of the present invention, by calculating the pixel coordinate difference between the camera's viewing angle center and the pattern's center, the position of the standard pattern can be corrected, eliminating errors caused by misalignment between the pattern's center and the actual viewing angle center, thereby reducing errors in subsequent field curvature calculations. The pixel coordinate difference refers to the difference between the pixel coordinates of the camera's viewing angle center and the pattern's center.

[0127] The embodiment of the present invention can correct the initial geometric deviation by adjusting the displacement of the mobile platform when the pixel coordinate difference is not zero and returning to the step of extracting the morphological features of the microscopic image, so that subsequent image acquisition and analysis are established on an accurate and unified spatial reference system, thereby significantly improving the accuracy, stability and automation level of microscopic image analysis.

[0128] For example, if the pixel coordinate difference is 5 for the horizontal axis and -5 for the vertical axis, the mobile platform is translated by 5 in the negative direction of the horizontal axis and by 5 in the positive direction of the vertical axis, and the pixel coordinate difference between the center of the camera viewing angle and the center of the pattern after the translation is re-detected.

[0129] S3. When the pixel coordinate difference is zero, locate the region of interest of the microscopic image and determine the sampling points of the region of interest.

[0130] By locating the region of interest (ROI) in the microscopic image when the pixel coordinate difference is zero, embodiments of the present invention prioritize analysis of areas with the highest imaging quality, reducing the interference of edge distortion on field curvature assessment. The ROI refers to a specific image region within a complete microscopic image that is specifically selected for focused attention, analysis, or processing based on analysis objectives or task requirements.

[0131] Optionally, the region of interest of the microscopic image may be determined by image processing techniques, such as edge detection algorithms, morphological operation algorithms, and the like.

[0132] For more information on the areas of interest, see Figure 3 As shown in FIG. 1 , an embodiment of the present invention provides a standard pattern for realizing the method for automatically detecting the field curvature of a microscope objective lens, wherein Figure 3 In the embodiment of the present invention, the region of interest of the microscopic image can be determined by image processing technology, and the feature of interest of the region of interest can be accurately identified.

[0133] The embodiment of the present invention can significantly reduce the amount of data to be processed and improve analysis accuracy and representativeness by determining the sampling points of the interest region, wherein the sampling points refer to one or more representative pixel positions selected within the defined interest region.

[0134] As an embodiment of the present invention, determining the sampling points of the region of interest includes:

[0135] performing contrast enhancement on the region of interest to obtain an enhanced region of interest;

[0136] Binarizing the enhanced region of interest to obtain a binarized region of interest;

[0137] Performing morphological operations on the binarized region of interest to obtain an outline of a pattern of interest;

[0138] Calculating the total length of the contour of the pattern of interest and detecting contour features of the contour of the pattern of interest;

[0139] Sampling points on the contour of the pattern of interest are determined according to the total contour length and the contour features.

[0140] Among them, the enhanced region of interest refers to the new image region obtained after a series of contrast enhancement or other preprocessing operations are performed on the selected region of interest in the original image. The binarized region of interest refers to the new image region obtained after a binarization operation is performed on the enhanced region of interest after contrast enhancement processing. The contour of the pattern of interest refers to the boundary line of the target pattern identified by performing edge detection or contour search operations on the binarized region of interest. The total contour length refers to the cumulative length of the target pattern contour (i.e., the sequence of pixel points on the boundary of the target object). The contour features refer to the various geometric properties and statistical characteristics of the contour (boundary) of the target pattern extracted from the image, such as corner points, key points, contour area, etc.

[0141] Optionally, the binarized region of interest may be obtained by a binarization algorithm, such as a global threshold method, a local threshold method, a fixed threshold method, and the like.

[0142] Optionally, the total length of the contour of the pattern of interest can be calculated by an accumulation method based on pixel distance, such as decomposing the contour of the pattern of interest into a pixel set, calculating the Euclidean distance between adjacent pixel points in the pixel set, and accumulating the Euclidean distance to obtain the total length of the contour.

[0143] S4. Determine the local meridian direction and the local sagittal direction of the sampling point to construct a local dynamic coordinate system of the sampling point, and analyze the meridian magnification and sagittal magnification of the sampling point based on the sampling point, the standard pattern, and the local dynamic coordinate system.

[0144] Embodiments of the present invention accurately describe the directional characteristics of a sampling point in space by determining the local meridional direction and local sagittal direction of the sampling point. The local meridional direction refers to the direction passing through the sampling point and parallel to the optical axis (or principal ray). The local sagittal direction refers to the direction passing through the sampling point and perpendicular to the local meridional direction.

[0145] As an embodiment of the present invention, determining the local meridian direction and the local sagittal direction of the sampling point includes:

[0146] determining the sampling point coordinates of the sampling point according to the microscopic image corresponding to the sampling point;

[0147] determining the pattern center coordinates of the microscopic image corresponding to the pattern center;

[0148] Calculating the meridian direction vector of the sampling point according to the sampling point coordinates and the pattern center coordinates;

[0149] Determining the local meridian direction of the sampling point based on the meridian direction vector;

[0150] Determine a 90° rotation matrix of the meridian direction vector, and calculate the sagittal direction vector of the sampling point based on the 90° rotation matrix and the meridian direction vector;

[0151] The local sagittal direction of the sampling point is determined according to the sagittal direction vector.

[0152] The sampling point coordinates refer to the pixel coordinates of the sampling point in the microscopic image. The pattern center coordinates refer to the coordinates of the center position of the standard pattern in the microscopic image. The meridian direction vector refers to the direction vector from the center of the standard pattern to the sampling point. The 90° rotation matrix refers to the linear transformation matrix used to rotate a vector 90 degrees about the origin. The sagittal direction vector refers to the vector at the sampling point that is perpendicular to the local meridian direction vector.

[0153] Optionally, the meridian direction vector of the sampling point can be calculated using the following formula:

[0154]

[0155] in, represents the meridian direction vector, p s The horizontal axis coordinate of the sampling point, p c The horizontal axis coordinate of the pattern center, q s The vertical coordinate of the sampling point, q c The vertical axis coordinate representing the center coordinate of the pattern.

[0156] Optionally, the sagittal direction vector of the sampling point can be calculated using the following formula:

[0157]

[0158] in, Represents the sagittal direction vector, R 90 represents a 90° rotation matrix, Represents the meridian direction vector.

[0159] By constructing a local dynamic coordinate system for the sampling point, the embodiments of the present invention can dynamically adjust as the sampling point changes, thereby meeting the needs of real-time analysis. The local dynamic coordinate system refers to a reference coordinate system with the sampling point as the origin, its X-axis pointing in the meridian direction, and its Y-axis pointing in the sagittal direction (the two axes are perpendicular to each other), which is dynamically adjusted according to the sampling point position.

[0160] By analyzing the meridional magnification and sagittal magnification of the sampling points based on the sampling points, the standard pattern, and the local dynamic coordinate system, embodiments of the present invention can directly calculate distortion parameters in the local coordinate system, avoiding complex nonlinear transformations in the global coordinate system. The meridional magnification refers to the magnification ratio of the image size relative to the object size in the meridional plane passing through the optical axis and the object point (or image point). The sagittal magnification can measure the magnification capability of the optical system in the sagittal direction perpendicular to the meridional plane.

[0161] As an embodiment of the present invention, analyzing the meridian magnification and sagittal magnification of the sampling point according to the sampling point, the standard pattern, and the local dynamic coordinate system includes:

[0162] Mapping the standard points corresponding to the sampling points on the standard pattern, and determining the actual coordinates of the standard points;

[0163] Determining the meridian direction basis vector and the sagittal direction basis vector of the local dynamic coordinate system;

[0164] The following formula is used to calculate the sampling point meridian distance and the sampling point sagittal distance between the sampling points according to the sampling point coordinates, the meridian direction basis vector, and the sagittal direction basis vector:

[0165] D Z =(p c1 -p c2 )e zp +(q c1 -q c2 )e zq

[0166] D S =(p c1 -p c2 )e sq +(q c1 -q c2 )e sp

[0167] Among them, D Z Indicates the meridian distance of the sampling point, D S Indicates the sagittal distance of the sampling point, p c1 Indicates the horizontal coordinate of the sampling point c1, p c2 Indicates the horizontal coordinate of the sampling point c2, e zp represents the horizontal axis component of the meridian basis vector in the local dynamic coordinate system, e zq Represents the longitudinal component of the meridian basis vector in the local dynamic coordinate system, q c1 Indicates the ordinate of the sampling point c1, q c2 Indicates the ordinate of the sampling point c2, e sq Represents the horizontal axis component of the sagittal direction basis vector in the local dynamic coordinate system, e sp The vertical axis component of the sagittal direction basis vector in the local dynamic coordinate system;

[0168] Calculating the standard point meridian direction distance and the standard point sagittal direction distance between the standard points based on the actual coordinates, the meridian direction basis vector and the sagittal direction basis vector;

[0169] Determining the meridian magnification of the sampling point according to the meridian distance of the sampling point and the meridian distance of the standard point;

[0170] The sagittal magnification of the sampling point is determined according to the sagittal distance of the sampling point and the sagittal distance of the standard point.

[0171] Among them, the standard point refers to the actual point on the standard pattern that corresponds to the sampling point currently being analyzed. The actual coordinates refer to the real position coordinates of the standard point in the standard coordinate system. The meridian direction basis vector refers to the unit vector along the meridian direction in the local dynamic coordinate system. The sagittal direction basis vector refers to the unit vector along the sagittal direction in the local dynamic coordinate system. The sampling point meridian direction distance refers to the distance component along the meridian direction between the sampling points in the local dynamic coordinate system. The sampling point sagittal direction distance refers to the distance component along the sagittal direction between the sampling points in the local dynamic coordinate system. The standard point meridian direction distance refers to the distance component along the meridian direction between the standard points in the local dynamic coordinate system. The standard point sagittal direction distance refers to the distance component along the sagittal direction between the standard points in the local dynamic coordinate system.

[0172] Optionally, the meridian direction basis vectors and sagittal direction basis vectors of the local dynamic coordinate system can be determined by a geometric constraint method, such as a reference plane method, a symmetry axis method, and the like.

[0173] Optionally, the meridian magnification of the sampling point can be obtained by calculating the ratio of the meridian distance of the sampling point to the meridian distance of the standard point.

[0174] S5. Fit the three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification, and the resolution distribution, analyze the full-field curvature distribution map of the three-dimensional imaging surface, and determine the field curvature of the corresponding objective lens of the microscope based on the full-field curvature distribution map.

[0175] By fitting the three-dimensional imaging surface of the microscopic image based on the meridional magnification, the sagittal magnification, and the resolution distribution, embodiments of the present invention can be used to more accurately understand the mapping relationship from the object plane to the image plane, and to globally and quantitatively describe the aberration distribution and degree of distortion across the entire field of view. The three-dimensional imaging surface is a mathematical model that describes the spatial geometric relationship between points on an ideal object plane and their corresponding points on the image plane after imaging through a microscope.

[0176] As an embodiment of the present invention, fitting the three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification, and the resolution distribution includes:

[0177] Normalizing the meridional magnification and the sagittal magnification to obtain a normalized meridional magnification and a normalized sagittal magnification;

[0178] Based on the resolution distribution, identifying the center point resolution and the edge point resolution of the microscopic image;

[0179] Calculating a resolution variation coefficient of the microscopic image according to the center point resolution and the edge point resolution;

[0180] determining a surface fitting algorithm for the microscopic image to construct an initial imaging surface of the microscopic image;

[0181] Determining a surface optimization objective function of the surface fitting algorithm according to the normalized meridional magnification, the normalized sagittal magnification, and the resolution variation coefficient;

[0182] Defining constraints of the surface fitting algorithm;

[0183] Iteratively optimizing the initial imaging surface according to the constraint conditions and the surface optimization objective function to obtain an optimized imaging surface;

[0184] The surface error of the optimized imaging surface is calculated, and when the surface error is less than a preset surface error threshold, the optimized imaging surface is used as the three-dimensional imaging surface of the microscopic image.

[0185] The normalized meridional magnification refers to the result obtained by normalizing the meridional magnification. The normalized sagittal magnification refers to the result obtained by normalizing the sagittal magnification. The center point resolution refers to the imaging resolution of the central area of ​​the microscopic image. The edge point resolution refers to the imaging resolution of the edge area of ​​the microscopic image. The resolution variation coefficient refers to a relative indicator used to quantify the difference in imaging resolution between the central area and the edge area of ​​the microscopic image. The surface fitting algorithm refers to a mathematical algorithm that uses a mathematical surface (e.g., a polynomial surface, spline surface, radial basis function surface, etc.) to approximate a set of given data points. The initial imaging surface refers to the first output result of the surface fitting algorithm. The surface optimization objective function refers to a function used to guide how to adjust and improve the initial imaging surface to make it better conform to the actual observation data (i.e., the normalized meridional and sagittal magnification). The constraints refer to the restrictions imposed on the optimization process during the surface fitting and optimization process to ensure that the final optimized imaging surface has physical meaning, conforms to the actual situation, or meets specific requirements. The optimized imaging surface refers to a mathematical model, obtained through calculation and optimization, that accurately describes the spatial variation characteristics of a microscope. The surface error measures the degree of deviation between the optimized imaging surface obtained through the optimization process and the actual observed data. The preset surface error threshold serves as a reference standard for determining whether the generated optimized imaging surface meets acceptable quality standards.

[0186] For further information on the center point resolution and the edge point resolution, see Figure 4 and Figure 5 As shown in FIG. 1 , a schematic diagram of the center point resolution and a schematic diagram of the edge point center for realizing an automated detection method for field curvature of a microscope objective lens provided by an embodiment of the present invention are provided. Figure 4 In the diagram of center point resolution, the high resolution and center resolution value of the center point corresponding to the microscopic image can be determined. Figure 4 The change of the middle curve indicates that the central field of view is close to the diffraction limit and the aberration is small; Figure 5 In the embodiment, the low resolution and edge resolution values ​​of the corresponding edge points of the microscopic image can be determined. Figure 5 The change of the middle curve indicates that there are aberrations at the edge and geometric distortion at the edge of the field of view.

[0187] Optionally, normalizing the meridian magnification to obtain a normalized meridian magnification includes:

[0188] calculating a meridional magnification mean and a meridional magnification variance of the meridional magnification,

[0189] Based on the meridian magnification mean and the meridian magnification standard deviation, the meridian magnification is normalized using the following formula to obtain the normalized meridian magnification:

[0190]

[0191] in, represents the normalized meridian magnification of the jth sampling point, Q j represents the meridian magnification of the jth sampling point, represents the mean meridian magnification, b j Indicates the standard deviation of the meridian magnification of the j-th sampling point.

[0192] The meridian magnification mean refers to the arithmetic mean of all data points in multiple sets of calculated meridian magnification data. The meridian magnification variance refers to the average of the squares of the differences between all data points and their mean in multiple sets of calculated meridian magnification data.

[0193] Optionally, the surface fitting algorithm of the microscopic image can be determined by a deep learning model, such as a convolutional neural network (CNN), a Transformer, etc.

[0194] Optionally, the surface optimization objective function of the surface fitting algorithm can be determined by a Bayesian optimization algorithm.

[0195] By analyzing the full-field curvature distribution map of the three-dimensional imaging surface, embodiments of the present invention can intuitively demonstrate the degree of curvature of the imaging surface in different regions, thereby quantitatively assessing the actual impact of field curvature on image clarity and resolution. The full-field curvature distribution map is a chart showing the distribution of curvature values ​​at various locations on the three-dimensional imaging surface.

[0196] As an embodiment of the present invention, analyzing the full-field curvature distribution map of the three-dimensional imaging surface includes:

[0197] determining a surface position function of the three-dimensional imaging surface;

[0198] calculating first and second derivatives of the surface position function;

[0199] constructing a characteristic matrix of the three-dimensional imaging surface based on the first-order derivative and the second-order derivative;

[0200] Calculating the eigenvalues ​​of the characteristic matrix, and calculating the principal curvature of the three-dimensional imaging surface based on the eigenvalues;

[0201] Calculating the mean curvature and Gaussian curvature of the three-dimensional imaging surface according to the principal curvature;

[0202] A full-field curvature distribution map of the three-dimensional imaging surface is determined according to the average curvature, the Gaussian curvature, and the principal curvature.

[0203] The surface position function refers to a function used to mathematically describe the position of a three-dimensional imaging surface. The first-order derivative refers to the local change of the surface in the parameter direction. The second-order derivative refers to the change in curvature of the surface in the parameter direction. The characteristic matrix refers to a matrix used to describe the geometric characteristics of the surface. The eigenvalue refers to a scalar that satisfies the surface position function. The principal curvature refers to the maximum and minimum curvatures of the three-dimensional imaging surface at that point. The mean curvature refers to the average measure of the curvature of the surface at that point. The Gaussian curvature refers to the product of the maximum and minimum principal curvatures at that point.

[0204] Optionally, the surface position function of the three-dimensional imaging surface can be determined by parameterized fitting, such as bilinear, bicubic B-spline, NURBS (non-uniform rational B-spline), etc.

[0205] Optionally, the feature matrix of the three-dimensional imaging surface can be constructed by curvature matrix estimation driven by deep learning, such as neural networks, graph neural networks, etc.

[0206] Optionally, the principal curvature of the three-dimensional imaging surface can be calculated by adaptive eigenvalue weighted fusion, such as identifying the weight factor of the eigenvalue, performing nonlinear weighting on the eigenvalue based on the weight factor to obtain a weighted eigenvalue, and calculating the principal curvature of the three-dimensional imaging surface based on the weighted eigenvalue.

[0207] The embodiment of the present invention determines the field curvature of the objective lens corresponding to the microscope based on the full field curvature distribution map, and can accurately measure the magnitude of the field curvature generated by the objective lens, objectively compare the field curvature performance of different objective lens samples, and achieve control of the objective lens production quality.

[0208] Compared with the problems described in the background technology, the embodiment of the present invention can realize accurate automatic rotation calibration of the microscope platform by calculating the acute angles of the straight line segments and the horizontal lines in the microscopic image, thereby improving image quality and measurement accuracy; optionally, the embodiment of the present invention can correct the initial geometric deviation by adjusting the displacement of the mobile platform when the pixel coordinate difference is not zero, and returning to the step of extracting the morphological features of the microscopic image, so that subsequent image acquisition and analysis are based on an accurate and unified spatial reference system, thereby significantly improving the accuracy, stability and automation level of microscopic image analysis; the embodiment of the present invention can significantly reduce the amount of data to be processed by determining the sampling points of the region of interest, thereby improving the analysis accuracy and representativeness; the embodiment of the present invention can significantly reduce the amount of data to be processed by determining the sampling points of the region of interest, thereby improving the analysis accuracy and representativeness; The sample point, the standard pattern, and the local dynamic coordinate system are analyzed, and the meridional magnification and sagittal magnification of the sample point are analyzed to directly calculate the distortion parameters in the local coordinate system, avoiding complex nonlinear transformations in the global coordinate system. The embodiment of the present invention fits the three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification, and the resolution distribution, which can be used to more accurately understand the mapping relationship from the object plane to the image plane, and globally and quantitatively describe the aberration distribution and distortion degree within the entire field of view. Finally, the embodiment of the present invention determines the field curvature of the corresponding objective lens of the microscope based on the full field curvature distribution map, which can accurately measure the size of the field curvature generated by the objective lens, objectively compare the field curvature performance of different objective lens samples, and realize the control of the objective lens production quality. Therefore, the method for automated detection of field curvature of microscope objective lenses provided by the embodiment of the present invention can improve the accuracy and efficiency of automated detection of field curvature of microscope objective lenses.

[0209] Example 2:

[0210] like Figure 6 FIG. 1 is a functional module diagram of a system suitable for automatic detection of field curvature of a microscope objective lens according to the present invention.

[0211] The system 600 for automated detection of field curvature of a microscope objective lens described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the system can include an image derotation module 601, an image correction module 602, a sampling point positioning module 603, a magnification calculation module 604, and a field curvature analysis module 605. A module, also referred to as a unit, is a series of computer program segments that can be executed by a processor in an electronic device and perform a fixed function. These modules are stored in the memory of the electronic device.

[0212] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0213] The image derotation module 601 is configured to read a microscopic image with a standard pattern, the microscopic image being read through a microscope, calculate an acute angle between a straight line segment and a horizontal line in the microscopic image, and when the acute angle is not zero, adjust the rotation angle of the corresponding moving platform of the microscope and return to the step of reading the microscopic image with the standard pattern;

[0214] The image correction module 602 is configured to calculate the resolution distribution of the microscopic image when the acute angle is zero, so as to calculate the camera visual center of the microscopic image, extract morphological features of the microscopic image to determine the pattern center of the microscopic image, calculate the pixel coordinate difference between the camera visual center and the pattern center, and when the pixel coordinate difference is not zero, adjust the displacement of the mobile platform and return to the step of extracting the morphological features of the microscopic image;

[0215] The sampling point positioning module 603 is configured to locate the region of interest of the microscopic image and determine the sampling point of the region of interest when the pixel coordinate difference is zero;

[0216] The magnification calculation module 604 is configured to determine the local meridian direction and the local sagittal direction of the sampling point to construct a local dynamic coordinate system of the sampling point, and analyze the meridian magnification and sagittal magnification of the sampling point based on the sampling point, the standard pattern, and the local dynamic coordinate system;

[0217] The field curvature analysis module 605 is used to fit the three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification and the resolution distribution, analyze the full field curvature distribution map of the three-dimensional imaging surface, and determine the field curvature of the corresponding objective lens of the microscope based on the full field curvature distribution map.

[0218] In detail, the modules in the system 600 for automatic detection of field curvature of a microscope objective lens according to the embodiment of the present invention are used in the same manner as described above. Figure 1The same technical means as the method for automatic detection of field curvature of a microscope objective described in and can produce the same technical effects are not described here.

[0219] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0220] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for automated detection of field curvature of a microscope objective lens, characterized in that: The method comprises: reading a microscopic image with a standard pattern, the microscopic image being read through a microscope, calculating an acute angle between a straight line segment and a horizontal line in the microscopic image, and when the acute angle is not zero, adjusting a rotation angle of a corresponding movable platform of the microscope, and returning to the step of reading the microscopic image with the standard pattern; When the acute angle is zero, calculating the resolution distribution of the microscopic image to calculate the camera visual center of the microscopic image, extracting morphological features of the microscopic image to determine the pattern center of the microscopic image, calculating the pixel coordinate difference between the camera visual angle center and the pattern center, and when the pixel coordinate difference is not zero, adjusting the displacement of the mobile platform and returning to the step of extracting the morphological features of the microscopic image; When the pixel coordinate difference is zero, locating the region of interest of the microscopic image and determining a sampling point of the region of interest; Determining the local meridian direction and the local sagittal direction of the sampling point to construct a local dynamic coordinate system of the sampling point, and analyzing the meridian magnification and sagittal magnification of the sampling point based on the sampling point, the standard pattern, and the local dynamic coordinate system; According to the meridional magnification, the sagittal magnification and the resolution distribution, the three-dimensional imaging surface of the microscopic image is fitted, the full field curvature distribution map of the three-dimensional imaging surface is analyzed, and based on the full field curvature distribution map, the field curvature of the corresponding objective lens of the microscope is determined.

2. The method for automatic detection of field curvature of a microscope objective lens according to claim 1, wherein: The calculating the acute angle between the straight line segment and the horizontal line in the microscopic image comprises: performing binarization processing on the microscopic image to obtain a binarized image; Performing edge detection on the binary image to obtain image edges; Based on the image edges, detecting straight line segments of the binary image, and determining a target straight line segment among the straight line segments; Determine the pixel coordinates of the starting point and the ending point of the target straight line segment; The acute angle between the target straight line segment and the horizontal line is calculated according to the starting pixel coordinates and the ending pixel coordinates.

3. The method for automated detection of field curvature of a microscope objective lens according to claim 1, wherein: Calculating the resolution distribution of the microscopic image includes: dividing the microscopic image into multiple image regions on an even basis; Calculating horizontal gradients and vertical gradients of pixels within the multi-image area; determining a gradient magnitude of the pixel according to the horizontal gradient and the vertical gradient; Calculating the gradient magnitude mean and the gradient magnitude mean variance of the multiple image regions; The resolution of the multi-image region is determined according to the gradient amplitude mean and the gradient amplitude mean variance, so as to determine the resolution distribution of the microscopic image.

4. The method for automated detection of field curvature of a microscope objective lens according to claim 1, wherein: The extracting of morphological features of the microscopic image comprises: performing image enhancement on the microscopic image to obtain an enhanced microscopic image; performing threshold segmentation on the enhanced microscopic image to obtain a segmented microscopic image; performing contour extraction on the segmented microscopic image to obtain a pattern contour; extracting shape features and spatial distribution features of the segmented microscopic image; The morphological features of the microscopic image are determined according to the pattern outline, the shape features, and the spatial distribution features.

5. The method for automated detection of field curvature of a microscope objective lens according to claim 1, wherein: The determining of the sampling points of the region of interest includes: performing contrast enhancement on the region of interest to obtain an enhanced region of interest; Binarizing the enhanced region of interest to obtain a binarized region of interest; Performing morphological operations on the binarized region of interest to obtain an outline of a pattern of interest; Calculating the total length of the contour of the pattern of interest and detecting contour features of the contour of the pattern of interest; Sampling points on the contour of the pattern of interest are determined according to the total contour length and the contour features.

6. The method for automated detection of field curvature of a microscope objective lens according to claim 1, wherein: Determining the local meridian direction and the local sagittal direction of the sampling point includes: determining the sampling point coordinates of the sampling point according to the microscopic image corresponding to the sampling point; determining the pattern center coordinates of the microscopic image corresponding to the pattern center; Calculating the meridian direction vector of the sampling point according to the sampling point coordinates and the pattern center coordinates; Determining the local meridian direction of the sampling point based on the meridian direction vector; Determine a 90° rotation matrix of the meridian direction vector, and calculate the sagittal direction vector of the sampling point based on the 90° rotation matrix and the meridian direction vector; The local sagittal direction of the sampling point is determined according to the sagittal direction vector.

7. The method for automated detection of field curvature of a microscope objective lens according to claim 1, wherein: Analyzing the meridian magnification and sagittal magnification of the sampling point according to the sampling point, the standard pattern, and the local dynamic coordinate system includes: Mapping a standard point corresponding to the sampling point on the standard pattern, and determining the actual coordinates of the standard point; Determining the meridian direction basis vector and the sagittal direction basis vector of the local dynamic coordinate system; Calculating the sampling point meridian distance and the sampling point sagittal distance between the sampling points according to the sampling point coordinates corresponding to the sampling points, the meridian direction basis vector, and the sagittal direction basis vector; Calculating the standard point meridian direction distance and the standard point sagittal direction distance between the standard points based on the actual coordinates, the meridian direction basis vector and the sagittal direction basis vector; Determining the meridian magnification of the sampling point according to the meridian distance of the sampling point and the meridian distance of the standard point; The sagittal magnification of the sampling point is determined according to the sagittal distance of the sampling point and the sagittal distance of the standard point.

8. The method for automated detection of field curvature of a microscope objective lens according to claim 1, wherein: The step of fitting a three-dimensional imaging surface of the microscopic image according to the meridional magnification, the sagittal magnification, and the resolution distribution includes: Normalizing the meridional magnification and the sagittal magnification to obtain a normalized meridional magnification and a normalized sagittal magnification; Based on the resolution distribution, identifying the center point resolution and the edge point resolution of the microscopic image; Calculating a resolution variation coefficient of the microscopic image according to the center point resolution and the edge point resolution; determining a surface fitting algorithm for the microscopic image to construct an initial imaging surface of the microscopic image; Determining a surface optimization objective function of the surface fitting algorithm according to the normalized meridional magnification, the normalized sagittal magnification, and the resolution variation coefficient; Defining constraints of the surface fitting algorithm; Iteratively optimizing the initial imaging surface according to the constraint conditions and the surface optimization objective function to obtain an optimized imaging surface; The surface error of the optimized imaging surface is calculated, and when the surface error is less than a preset surface error threshold, the optimized imaging surface is used as the three-dimensional imaging surface of the microscopic image.

9. The method for automated detection of field curvature of a microscope objective lens according to claim 8, wherein: Normalizing the meridian magnification to obtain a normalized meridian magnification includes: calculating a meridional magnification mean and a meridional magnification variance of the meridional magnification, The meridional magnification is normalized based on the meridional magnification mean and the meridional magnification standard deviation to obtain a normalized meridional magnification.

10. The method for automatic detection of field curvature of a microscope objective lens according to claim 1, wherein: The analyzing the full-field curvature distribution map of the three-dimensional imaging surface includes: determining a surface position function of the three-dimensional imaging surface; calculating first and second derivatives of the surface position function; constructing a characteristic matrix of the three-dimensional imaging surface based on the first-order derivative and the second-order derivative; Calculating the eigenvalues ​​of the characteristic matrix, and calculating the principal curvature of the three-dimensional imaging surface based on the eigenvalues; Calculating the mean curvature and Gaussian curvature of the three-dimensional imaging surface according to the principal curvature; A full-field curvature distribution map of the three-dimensional imaging surface is determined according to the average curvature, the Gaussian curvature, and the principal curvature.

Images