Method, device, equipment and storage medium for detecting center of calibration plate

Abstract

The present application relates to a method, apparatus, device and storage medium for detecting the center of a calibration plate. In the present application, a perspective transformation matrix is ​​determined by creating a virtual standard square, and the perspective transformation matrix is ​​used to map the original calibration plate image into a square calibration plate image. In this way, the deformation problem of the calibration plate during tilted shooting can be effectively solved, so that the center detection is not affected by the tilt of the viewing angle; and, the present application merges the contours of each target circle in the square calibration plate image through a clustering algorithm, and determines the merged center coordinates through confidence, which can effectively suppress noise interference and improve the accuracy of center detection.

Description

Technical Field

[0001] The present application relates to the field of camera calibration technology, and in particular to a method, device, equipment and storage medium for detecting the center of a calibration plate. Background Art

[0002] The ‌Circular Calibration Plate‌ is a calibration tool used in machine vision, photogrammetry, and 3D reconstruction, consisting of a regularly arranged pattern of dots. Circular calibration plates are widely used for camera intrinsic and extrinsic calibration. The core principle is to detect circular feature points in the image and use geometric relationships to calculate the camera's distortion parameters and attitude information. However, when there is a large inclination angle between the calibration plate and the camera, the circle in the image will be deformed into an ellipse due to perspective projection, affecting the calibration accuracy. Existing technical solutions usually use an ellipse fitting method to restore the center position of the circle. However, when the inclination angle is large, the minor axis of the ellipse will be significantly reduced, resulting in unstable numerical calculations, error amplification, and even inability to accurately fit. Secondly, this method is highly dependent on edge detection and is easily affected by factors such as noise, illumination changes, and reflections, resulting in unstable ellipse fitting. In addition, due to the influence of perspective distortion, the major-minor axis ratio cannot be directly used to restore the original circle radius, requiring additional geometric calculations and optimization, which increases uncertainty. Furthermore, the computational complexity of ellipse fitting is high, especially in multi-point calibration tasks, which may affect real-time performance.

[0003] Therefore, how to improve the accuracy of calibration plate center extraction in tilted scenes is a problem that those skilled in the art need to solve. Summary of the Invention

[0004] The present application provides a method, apparatus, device and storage medium for detecting the center of a calibration plate, so as to improve the accuracy of extracting the center of the calibration plate in an inclined scene.

[0005] In a first aspect, the present application provides a method for detecting the center of a calibration plate, comprising:

[0006] Get the original calibration plate image;

[0007] Identifying quadrilateral corner points from the original calibration plate image;

[0008] Creating a virtual standard square, and determining a perspective transformation matrix according to virtual corner points of the virtual standard square and corner points of the quadrilateral;

[0009] Mapping the original calibration plate image into a square calibration plate image using the perspective transformation matrix;

[0010] Detecting the target circle outline in the square calibration plate image;

[0011] Divide each target circle contour into at least one cluster subset, and generate a merged center coordinate based on the center coordinate and confidence level of each target circle contour in each cluster subset;

[0012] The actual circle center coordinates corresponding to the original calibration plate image are determined using the perspective transformation matrix and the merged center coordinates.

[0013] Optionally, identifying quadrilateral corner points from the original calibration plate image includes:

[0014] Binarizing the original calibration plate image according to a first target threshold to generate a processed first calibration plate binary image;

[0015] Detecting the quadrilateral outline of the first calibration plate binary image;

[0016] Filtering quadrilateral corner points according to the cosine values ​​of the vertices of the quadrilateral outline;

[0017] If the number of the screened quadrilateral corner points is less than four, the first target threshold is adjusted, and the step of binarizing the original calibration plate image is continued based on the adjusted first target threshold.

[0018] Optionally, after identifying quadrilateral corner points from the original calibration plate image, the method further includes:

[0019] Taking each quadrilateral corner point as the center, determine the rectangular region of interest of each quadrilateral corner point;

[0020] Perform binarization processing on each rectangular region of interest to generate a rectangular region binary image;

[0021] Convert the binary images of each rectangular area into corresponding two-dimensional point cloud data;

[0022] By using a clustering algorithm, a target point cloud data subset that is closest to the quadrilateral corner point in the two-dimensional point cloud data is determined, and a fitting edge segment of the target point cloud data subset is determined;

[0023] If the angle between the fitted edge segments of the target rectangular region of interest is detected to be less than the angle threshold, the quadrilateral corner points corresponding to the target rectangular region of interest are used as triangular marker corner points;

[0024] The corner point input order is determined based on the triangle marked corner points; wherein the corner point input order is used to determine the input order of each quadrilateral corner point when determining the perspective transformation matrix.

[0025] Optionally, detecting the target circle contour in the square calibration plate image includes:

[0026] Binarizing the square calibration plate image according to a second target threshold to generate a processed second calibration plate binary image; wherein the value of the second target threshold is different in each iteration process;

[0027] Detecting an initial circle contour in the binary image of the second calibration plate;

[0028] Calculating the area and second geometric characteristic parameters of each initial circular contour; the second geometric characteristic parameters include at least one of roundness, inertia ratio, and convexity;

[0029] If the area of ​​the initial circular contour is within the valid area range and the second geometric feature parameter is within the valid parameter range, the initial circular contour is used as the target circular contour.

[0030] Optionally, the process of determining the effective area range includes:

[0031] Performing edge detection on the square calibration plate image, extracting edge pixels, and determining target line segments based on the edge pixels;

[0032] Searching for closed target line segments according to the first geometric characteristic parameters of each target line segment;

[0033] If so, perform circle fitting on the closed target line segment to generate a fitted target circle;

[0034] An area index is determined according to the area of ​​each target circle, and an effective area range is determined according to the area index.

[0035] Optionally, each target circle contour is divided into at least one cluster subset, including:

[0036] Calculating the confidence level of each target circle contour according to the second geometric characteristic parameter of the target circle contour;

[0037] Calculate the center coordinates of each target circle contour according to the area and pixel value of the target circle contour;

[0038] According to the contour radius, center coordinates and center distance threshold of each target circle contour in each iteration process, each target circle contour is divided into at least one cluster subset.

[0039] Optionally, before determining the actual center coordinates corresponding to the original calibration plate image, the method further includes:

[0040] Determine whether the number of cluster subsets is the same as the number of standard circles;

[0041] If yes, continue to perform the step of using the perspective transformation matrix and the merged center coordinates to determine the actual circle center coordinates corresponding to the original calibration plate image;

[0042] If not, the effective parameter range is adjusted, and the target circle contour is re-determined based on the effective area range and the adjusted effective parameter range.

[0043] In a second aspect, the present application provides a device for detecting the center of a calibration plate, comprising:

[0044] An acquisition module is used to obtain the original calibration plate image;

[0045] A recognition module, configured to recognize quadrilateral corner points from the original calibration plate image;

[0046] a matrix determination module, configured to create a virtual standard square and determine a perspective transformation matrix based on virtual corner points of the virtual standard square and corner points of the quadrilateral;

[0047] A mapping module, configured to map the original calibration plate image into a square calibration plate image using the perspective transformation matrix;

[0048] A detection module, configured to detect the target circle outline in the square calibration plate image;

[0049] a merging module, configured to divide each target circle contour into at least one cluster subset, and generate a merged center coordinate based on the center coordinates and confidence levels of each target circle contour in each cluster subset;

[0050] A coordinate determination module is used to determine the actual circle center coordinates corresponding to the original calibration plate image using the perspective transformation matrix and the merged center coordinates.

[0051] In a third aspect, the present application provides an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0052] Memory for storing computer programs;

[0053] The processor is configured to implement the steps of the circle center detection method when executing the program stored in the memory.

[0054] In a fourth aspect, the present application further provides a computer storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to execute the steps of the above-mentioned center of circle detection method of the present application.

[0055] The above-mentioned technical solution provided by the embodiments of the present application has the following advantages over the prior art: the present application provides a method, device, equipment and storage medium for detecting the center of a calibration plate. In the present application, a perspective transformation matrix is ​​determined by creating a virtual standard square, and the perspective transformation matrix is ​​used to map the original calibration plate image into a square calibration plate image. In this way, the deformation problem of the calibration plate during tilted shooting can be effectively solved, so that the center detection is not affected by the tilt of the viewing angle; and, the present application merges the contours of each target circle in the square calibration plate image through a clustering algorithm, and determines the merged center coordinates through confidence, which can effectively suppress noise interference and improve the accuracy of center detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0058] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings. These exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements. Unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0059] Figure 1 A schematic flow chart of a method for detecting the center of a calibration plate provided in an embodiment of the present application;

[0060] Figure 2 A schematic diagram of an original calibration plate image provided in an embodiment of the present application;

[0061] Figure 3 A square calibration plate image provided in an embodiment of the present application;

[0062] Figure 4 Another square calibration plate image provided in an embodiment of the present application;

[0063] Figure 5 A schematic structural diagram of a center detection device for a calibration plate provided in an embodiment of the present application;

[0064] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] Existing solutions typically use ellipse fitting, detecting and fitting the edges of an ellipse to calculate its center coordinates, major and minor axes, and rotation angles. This method then infers the original center position and camera calibration parameters to compensate for perspective distortion. However, this approach suffers from poor robustness at high inclination angles and struggles to meet high-precision calibration requirements.

[0066] Therefore, in embodiments of the present application, a method, apparatus, device, and storage medium for detecting the center of a calibration plate are provided. This solution can address the problem of insufficient circle detection accuracy in complex scenes due to image tilt. By combining perspective transformation, edge detection, and local feature analysis, this solution significantly improves the accuracy and robustness of circle detection, thereby improving the accuracy of calibration plate center extraction, and is particularly suitable for tilted scenes.

[0067] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0068] The disclosure below provides many different embodiments or examples for implementing different configurations of the present invention. To simplify the disclosure of the present invention, the components and configurations of specific examples are described below. Of course, these are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numerals and / or letters in different examples. Such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or configurations discussed.

[0069] See also Figure 1 , is a flow chart of a method for detecting the center of a calibration plate provided in an embodiment of the present application, the method comprising the following steps:

[0070] S101, obtaining the original calibration plate image;

[0071] In this application, the original calibration plate image is an image of a circular calibration plate captured by an image capture device. When the circular calibration plate is at a large inclination angle relative to the image capture device, the circle in the original calibration plate image can be deformed into an ellipse due to perspective projection, affecting calibration accuracy. Therefore, in this application, a subsequent process can be used to eliminate perspective distortion caused by the shooting angle and improve the accuracy of circle center detection.

[0072] S102, identifying quadrilateral corner points from the original calibration plate image;

[0073] In this embodiment, the input original calibration plate image is first subjected to threshold binarization processing to generate a calibration plate binary image, thereby providing a high-quality image basis for subsequent contour extraction. Then, an edge detection algorithm is used to identify the edge contour of the original calibration plate image, and then the four corner points of the quadrilateral in the original calibration plate image are determined based on the edge contour. In this embodiment, the corner points in the original calibration plate image are referred to as quadrilateral corner points. Figure 2 , which is a schematic diagram of an original calibration plate image provided in an embodiment of the present application, see Figure 2 , Figure 2 The positions 1, 2, 3, and 4 in the image are the corner points of the quadrilateral in the original calibration plate image.

[0074] S103, creating a virtual standard square, and determining a perspective transformation matrix according to virtual corner points of the virtual standard square and corner points of the quadrilateral;

[0075] S104, using a perspective transformation matrix, mapping the original calibration plate image into a square calibration plate image;

[0076] In this application, in order to solve the deformation problem of the circular calibration plate when shooting at an angle, an ideal square can be defined in the target space as a virtual projection plane based on the characteristic that the outer frame of the calibration plate is a square. The coordinates of the four corner points of the square are set according to the standard square geometry. This application refers to this square as a virtual standard square. Then, a correspondence is established between the real quadrilateral corner points detected in the original calibration plate image and the virtual corner points of the virtual standard square, and the perspective transformation matrix between the two is calculated. By applying this perspective transformation matrix, the original calibration plate image can be perspective corrected so that the tilted original calibration plate image actually shot can be accurately mapped to the preset square projection plane, thereby eliminating the perspective distortion caused by the shooting angle and generating a square calibration plate image that conforms to the standard coordinate system.

[0077] S105, detecting the target circle outline in the square calibration plate image;

[0078] In this embodiment, after acquiring the square calibration plate image, it is necessary to perform binarization processing on the square calibration plate image to generate a second calibration plate binary image. Then, contour detection is performed on the second calibration plate binary image using a contour detection method. Because the circular calibration plate image in this application includes multiple circles, the contours detected from the square calibration plate image in this application are circular contours. In this application, the circular contours detected from the square calibration plate image are referred to as target circular contours.

[0079] S106, dividing each target circle contour into at least one cluster subset, and generating a merged center coordinate based on the center coordinate and confidence level of each target circle contour in each cluster subset;

[0080] It should be noted that when the present application detects the target circle contour from the square calibration plate image, multiple target circle contours may be detected due to reasons such as lighting and imaging conditions. Therefore, in the present application, it is necessary to divide each target circle contour into at least one cluster subset; in each cluster subset, multiple target circle contours with similar center coordinate distances are included, and then according to the center coordinates and confidence of each target circle contour in each cluster subset, an updated merged center coordinate can be generated. Among them, the confidence of the target circle contour is used to quantify the probability index that the detected target circle contour is close to the ideal circle. When calculating the confidence, geometric parameters such as roundness, inertia ratio, and convexity can be integrated to evaluate the geometric reliability of the contour. Therefore, the present application generates a merged center coordinate by using the center coordinates and confidence of each target circle contour in the cluster subset, which can effectively suppress noise interference and improve the accuracy of circle center detection.

[0081] S107 , using the perspective transformation matrix and the merged center coordinates, determine the actual circle center coordinates corresponding to the original calibration plate image.

[0082] In this application, the merged center coordinates detected by the above steps are located in the image space after perspective transformation. Therefore, in order to obtain the true center position of the circle in the original calibration plate image, the merged center coordinates need to be mapped back to the original image coordinate system through the inverse transformation of the perspective transformation matrix. Specifically, for each detected merged center coordinate , apply the inverse perspective transformation matrix Perform back projection calculation: ;in This is the actual circle center coordinate corresponding to the original calibration plate image. By this method, the present application can ensure that all geometric measurement results can ultimately accurately correspond to the position of the original shooting scene.

[0083] In summary, in this application, a virtual standard square is created to determine the perspective transformation matrix, and the perspective transformation matrix is ​​used to map the original calibration plate image into a square calibration plate image. In this way, the deformation problem of the calibration plate during tilted shooting can be effectively solved, so that the center detection is not affected by the tilt of the viewing angle; and, in this application, the contours of each target circle are merged in the square calibration plate image through a clustering algorithm, and the merged center coordinates are determined by the confidence level, which can effectively suppress noise interference and improve the accuracy of center detection.

[0084] In another embodiment provided in the present application, the process of identifying quadrilateral corner points from the original calibration plate image specifically includes:

[0085] The original calibration plate image is binarized according to a first target threshold to generate a processed first calibration plate binarized image; the quadrilateral outline of the first calibration plate binarized image is detected; the quadrilateral corner points are screened according to the cosine value of each vertex of the quadrilateral outline; if the number of screened quadrilateral corner points is less than four, the first target threshold is adjusted, and the step of binarizing the original calibration plate image is continued based on the adjusted first target threshold.

[0086] Specifically, when identifying quadrilateral corner points from the original calibration plate image, the application first needs to perform threshold binarization processing on the input calibration plate image to generate a first calibration plate binarization image; all contours in the first calibration plate binarization image are detected through an edge detection algorithm, and a hierarchical relationship tree is established. The parent-child nesting characteristics of the contours are used to screen out candidate contour groups that may contain calibration plate features, effectively eliminating irrelevant interfering contours. Then, the number of convex hulls of each candidate contour in the candidate contour group is calculated to quickly determine the basic shape features of each candidate contour. If the number of vertices of the candidate contour convex hull is less than 4, it is directly excluded to ensure that the object to be subsequently processed has basic quadrilateral features. If the number of vertices of the candidate contour is 4, the candidate contour is determined to be a quadrilateral contour.

[0087] It should be noted that, since each angle in the original calibration image is ninety degrees, after the present application determines the contours of each quadrilateral with four convex hulls, the four quadrilateral corner points can be screened out according to the vertex angles of each convex hull. In the present application, it is first necessary to calculate the cosine value of the angle between the adjacent edge vectors at each vertex of the convex hull, and quantify the sharpness of the corner point by the absolute value of the cosine value, so as to provide an objective geometric feature basis for the subsequent corner point screening. Since the cosine value corresponding to the ninety-degree angle is 0, when screening the quadrilateral corner points, the present application can sort all the convex hull vertices in descending order according to the cosine value of the angle, and select the first four corner points with the cosine value closest to 0. These four corner points are the four sharpest corner points. The present application refers to the four sharpest corner points as quadrilateral corner points. In this way, it can be ensured that the screened corner points have the best geometric features and stability.

[0088] Furthermore, if the present application fails to find the four sharpest corner points in the initial detection, it can automatically enter the threshold optimization stage, that is: when the present application binarizes the original calibration plate image according to the first target threshold, the first target threshold can be set to the minimum threshold. If the four sharpest corner points cannot be found, the first target threshold is gradually adjusted according to the step size, and the next round of iteration process is re-executed, and the above process of determining the corner points of the quadrilateral is continued until the four sharpest corner points are found, or the first target threshold is adjusted to the maximum threshold, and the threshold optimization stage is exited.

[0089] In summary, in the present application, the original calibration plate image can be binarized according to the first target threshold, and the quadrilateral corner points can be accurately found from the original calibration plate image through the edge detection algorithm and cosine value comparison method; and, the present application dynamically optimizes the image binarization effect by dynamically adjusting the first target threshold to ensure that the circular marking points of the calibration plate form an optimal contrast with the background. This method solves the limitations of the traditional fixed threshold method in complex scenes and improves the recognition accuracy of quadrilateral corner points.

[0090] In another embodiment provided by the present application, after identifying the quadrilateral corner points from the original calibration plate image, it also includes: determining the rectangular region of interest of each quadrilateral corner point with each quadrilateral corner point as the center; binarizing each rectangular region of interest to generate a rectangular region binarized image; converting each rectangular region binarized image into corresponding two-dimensional point cloud data; determining the target point cloud data subset closest to the quadrilateral corner point in the two-dimensional point cloud data through a clustering algorithm, and determining the fitting edge segments of the target point cloud data subset; if the angle between the fitting edge segments of the target rectangular region of interest is detected to be less than the angle threshold, the quadrilateral corner point corresponding to the target rectangular region of interest is used as a triangle mark corner point; determining the corner point input order according to the triangle mark corner point; wherein, the corner point input order is used to determine the input order of each quadrilateral corner point when determining the perspective transformation matrix.

[0091] Accordingly, the present application determines the perspective transformation matrix based on the virtual corner points of the virtual standard square and the corner points of the quadrilateral, including: using the corner point input order, the virtual corner points of the virtual standard square and the quadrilateral corner points to determine the perspective transformation matrix; wherein, the corner point input order is used to determine the input order of each quadrilateral corner point, which is consistent with the input order of each virtual corner point.

[0092] It should be noted that when determining the perspective transformation matrix based on the virtual corner points of the virtual standard square and the quadrilateral corner points, the input order of each corner point needs to be clarified. If the input order of the virtual corner points is inconsistent with the input order of the quadrilateral corner points, the final calculated perspective transformation matrix will be wrong. The wrong perspective transformation matrix will map the original calibration plate image to an incorrect square calibration plate image.

[0093] See also Figure 3 , is a square calibration plate image provided in an embodiment of the present application, which is an image generated by perspective-correcting the original calibration plate image through a correct perspective transformation matrix; see Figure 4, is another square calibration plate image provided in an embodiment of the present application. This square calibration plate image is generated by perspective-correcting the original calibration plate image using an incorrect perspective transformation matrix. It can be seen that when calculating and determining the perspective transformation matrix, if the input order of the corner points of the various quadrilaterals is incorrect, an incorrect perspective transformation matrix will be calculated, and then an incorrect square calibration plate image will be generated. Therefore, in the present application, the input order of the corner points of the various quadrilaterals can be determined by searching for the corner points of the triangular markers, so that the perspective distortion caused by the shooting angle can be eliminated through the correct perspective transformation matrix, and a square calibration plate image that conforms to the standard coordinate system can be generated.

[0094] See also Figure 2 , Figure 2 The quadrilateral corner point 1 is different from the quadrilateral corner point 2, quadrilateral corner point 3, and quadrilateral corner point 4. The quadrilateral corner point 1 is a triangular area, so the quadrilateral corner point 1 is a special marked point, that is, a triangular marked corner point. In this embodiment, in order to identify the triangular marked corner points, the coordinates of the four quadrilateral corner points can be obtained. With each quadrilateral corner point as the center, a fixed-size rectangular region of interest (ROI) is established. The size of each rectangular ROI can be reasonably set according to the actual size of the calibration plate to ensure that it can completely contain the possible triangular marked area. The triangular marked area is the area containing the triangular marked corner points. See Figure 2 , Figure 2 The black triangular area at the corner point 1 of the quadrilateral in the figure is the triangular marker area. The image of each rectangular region of interest is then binarized to separate the triangular marker from the background, generating a binary image of the rectangular region corresponding to each rectangular region of interest. Each binary image of the rectangular region is then converted into a two-dimensional point cloud data structure, where the XY coordinates of the point cloud correspond to the UV coordinates of the image pixel, and the Z coordinate is uniformly set to 0, forming a planar point cloud set.

[0095] Furthermore, this application also needs to perform a Euclidean clustering algorithm on the two-dimensional point cloud data of each rectangular area of ​​interest, divide the point cloud into several subsets according to spatial distance, and then calculate the Euclidean distance from the centroid of each subset to the corresponding quadrilateral corner point. After sorting the distance from small to large, the target point cloud data subset with the closest distance is selected as the candidate triangle marking area.

[0096] After determining the target point cloud data subset for each rectangular ROI, edge extraction is performed on the target point cloud data subset. A boundary tracing algorithm is used to obtain the outer contour edge points of the point cloud. These edge points are used for subsequent geometric feature analysis to eliminate interference from internal noise points. The RANSAC algorithm (RANdom SAmple Consensus) is then used to perform straight line fitting on the edge point cloud. After each fitting, the fitted point cloud is removed. This process is repeated until all major fitted edge segments are extracted. The parametric equation and endpoint coordinates of each segment are recorded. The angle between adjacent fitted edge segments of each rectangular ROI is calculated. If a sharp angle (e.g., less than 80 degrees) is detected, the current rectangular ROI is determined to be the target rectangular ROI. The corner points of the quadrilateral corresponding to the target rectangular ROI are the corner points of the triangle region, also known as the triangle marker corner points.

[0097] After the present application determines the triangle marker corner points from the quadrilateral corner points, the order of inputting the corner points of each quadrilateral corner point can be determined according to a pre-set order. For example, the pre-set order is: triangle marker corner points, ordinary quadrilateral corner points, diagonal corner points of triangle marker corner points, and ordinary quadrilateral corner points. Therefore, after the present application determines the triangle marker corner points, the order of inputting the corner points of each quadrilateral can be determined. In this embodiment, the geometric center of the four corner points can be calculated, and a direction vector pointing from the center to the other corner points can be established. By comparing these direction vectors with the direction vectors of the identified triangle marker corner points, the position of the diagonal point that is 180 degrees antiparallel to the triangle marker corner point can be accurately found. See Figure 2 The direction vector from corner point 1 to corner point 3 is 180 degrees antiparallel to the direction vector from corner point 3 to corner point 1. Since corner point 1 is the triangle marked corner point, corner point 3 is the diagonal corner point of the triangle marked corner point. The final corner point input order is: corner point 1 - corner point 2 - corner point 3 - corner point 4.

[0098] In summary, it can be seen that after the present application binarizes the quadrilateral corner points and converts them into two-dimensional point cloud data, it can use clustering algorithms, edge extraction, straight line fitting, etc. to find the triangular marked corner points from each quadrilateral corner point, and then determine the input order of each quadrilateral corner point based on the triangular marked corner points. This method can ensure the consistency of the corner point order, provide standardized input data for subsequent perspective transformation and other processing, and ensure that an accurate perspective transformation matrix is ​​calculated.

[0099] In another embodiment of the present application, detecting target circle contours in a square calibration plate image and dividing each target circle contour into at least one cluster subset specifically includes the following steps:

[0100] Binarizing the square calibration plate image according to a second target threshold to generate a processed second calibration plate binary image; wherein the value of the second target threshold is different in each iteration; detecting initial circular contours in the second calibration plate binary image; calculating the area and second geometric characteristic parameters of each initial circular contour; wherein the second geometric characteristic parameters include at least one of roundness, moment of inertia ratio, and convexity;

[0101] If the area of ​​the initial circular contour is within the valid area range and the second geometric feature parameter is within the valid parameter range, the initial circular contour is used as the target circular contour;

[0102] According to the second geometric feature parameter of the target circle contour, the confidence of each target circle contour is calculated; according to the area and pixel value of the target circle contour, the center coordinates of each target circle contour are calculated; according to the contour radius, center coordinates and center distance threshold of each target circle contour in each iteration process, each target circle contour is divided into at least one cluster subset.

[0103] In the present application, when detecting the target circle contour in the square calibration plate image, it is first necessary to perform a binarization process on the input square calibration plate image. Among them, when the present application performs a binarization process on the square calibration plate image, it is necessary to perform multiple iterative processes. In the first iterative process, the second target threshold is set as the minimum threshold for binarization. In each subsequent iterative process, the second target threshold is gradually increased according to the step size step_thresh until the end threshold is reached. In each iteration, the improved findBlobsImprove function is called to detect the blob area and its center point in the current binary image; wherein, the blob area is a connected pixel area, that is, the area within the target circle contour in this application, and the center point of the blob area is the center coordinate of the target circle contour. The detected blob center points will be stored in curCenters. If a valid result is detected (curCenters is not empty), these center point sets will be added to the main vector curCenters_vector.

[0104] Here, the specific execution process of the findBlodsImprove function is explained:

[0105] 1. First, determine the contours of each target circle from the binary image of the second calibration plate:

[0106] a. Perform contour detection on the input binary image of the second calibration plate to obtain the initial circular contours of all possible candidate areas.

[0107] b. Traverse each initial circle contour and calculate the area of ​​the contour ( ) geometric features, the following is the formula for calculating the area: ;

[0108] in It is a binary image in pixels The value at (0 or 1).

[0109] c. Calculate the roundness of the initial circle contour ( ) geometric features, the following is the formula for calculating roundness: ;

[0110] (Perimeter) is calculated by the following formula:

[0111] ;

[0112] in, is a point on the contour, is the Euclidean distance. The circularity of an ideal circle is 1, and the smaller the value, the more it deviates from a circle.

[0113] d. Calculate the inertia ratio of the initial circle profile ( ) geometric characteristics, the following is the formula for calculating the inertia ratio: ;

[0114] in, and are the eigenvalues ​​of the inertia matrix:

[0115] ;

[0116] The eigenvalue is calculated as:

[0117] ;

[0118] The closer the inertia ratio is to 1, the closer the outline is to a circle; the closer it is to 0, the more slender the outline is.

[0119] e. Calculate the convexity of the initial circle contour ( ) geometric characteristics, the following is the formula for calculating the inertia ratio:

[0120] ;

[0121] ;

[0122] in, is the coordinate of the i-th vertex of the convex hull , , (Closed at both ends). : The contour is completely convex (such as a circle, rectangle, or convex polygon).

[0123] f. After calculating the area, roundness, moment of inertia ratio, and convexity of each initial circular profile through the above process, it is necessary to compare the area of ​​each initial circular profile with the effective area range, and compare the roundness, moment of inertia ratio, and convexity of each initial circular profile with the corresponding effective parameter range. If the area of ​​the initial circular profile is within the effective area range, and the roundness, moment of inertia ratio, and convexity are all within the effective parameter range, then the initial circular profile is determined to be the target circular profile.

[0124] In another embodiment of the present application, the process of determining the effective area range includes: performing edge detection on the square calibration plate image to extract edge pixel points, and determining the target line segments based on the edge pixel points; searching for closed target line segments based on the first geometric feature parameters of each target line segment; if so, performing circle fitting on the closed target line segments to generate a fitted target circle; determining an area index based on the area of ​​each target circle, and determining the effective area range based on the area index.

[0125] This application determines the effective area range based on the area of ​​each circle in the square calibration plate image. That is: this application performs edge detection on the square calibration plate image after perspective transformation and extracts edge pixel points. The edge pixel points are connected into continuous target line segments, and the first geometric features of each target line segment, such as the starting point, end point, length, etc., are calculated; then each target line segment is checked to see if it is close to closed, that is: the distance between the starting point and the end point is less than the threshold; if the target line segment is close to closed, the target line segment becomes a closed target line segment, and the closed target line segment is directly fitted with a circle. Circle fitting uses the least squares method to fit the circle equation:

[0126]

[0127] Then calculate the area of ​​each target circle generated after fitting, and take the arithmetic mean of the areas of all circles as the characteristic area index After determining the area index, this application can establish a dynamic threshold system based on the area index, setting the effective area range as The shape constraint parameters are uniformly verified using a triple validation mechanism, where the qualified thresholds for roundness, moment of inertia, and convexity are all set to the interval [0.7, 0.99], and the three feature parameters have equal weights.

[0128] For each target circle contour, the center coordinates and confidence level need to be calculated through the following process:

[0129] 1. Use the image moment method to calculate the center coordinates of the target circle contour:

[0130] ;

[0131] in, , is the area of ​​the contour.

[0132] 2. Comprehensive scoring confidence based on geometric features :

[0133] ;

[0134] g. Use the Welzl algorithm to calculate the radius of the minimum circumscribed circle of the contour ( ).

[0135] 2. Through the above process, the contour radius, center coordinates and confidence level of each target circle contour in the binary image of the second calibration plate can be obtained in each iteration. This application can use a clustering algorithm to merge contours with similar distances. For any two target contours i and j, an adjacency relationship is established when any of the following conditions are met:

[0136] ;

[0137] in, is the center coordinate of target contour i, is the center coordinate of target contour j, is the contour radius of target contour i, is the contour radius of target contour j, is the preset minimum center distance threshold, represents the Euclidean distance.

[0138] 3. This application targets each cluster subset. , it is necessary to calculate the center coordinates, confidence level and radius of each target circle contour after merging in each cluster subset.

[0139] The merge center coordinates are obtained by confidence weighting, specifically:

[0140] ;

[0141] in, is the merge center coordinate of the mth cluster subset, is the mth cluster subset, is the confidence of the i-th target circle contour, is the center coordinate of the i-th target circle contour.

[0142] The combined confidence is obtained by arithmetic averaging:

[0143] ;

[0144] in, is a subset The number of contours in , is the combined confidence of the mth cluster subset.

[0145] Merge Radius:

[0146] ;

[0147] in, is the radius of the i-th target circle contour, is the merging radius of the mth cluster subset.

[0148] In another embodiment of the present application, before determining the actual circle center coordinates corresponding to the original calibration plate image, the method further includes: determining whether the number of cluster subsets is the same as the number of standard circles;

[0149] If yes, continue to perform the step of using the perspective transformation matrix and the combined center coordinates to determine the actual circle center coordinates corresponding to the original calibration plate image;

[0150] If not, the effective parameter range is adjusted, and the target circle contour is re-determined based on the effective area range and the adjusted effective parameter range.

[0151] It should be noted that in the initial target circle contour detection process of this application, a strict valid parameter range can be set, such as setting the valid parameter range of convexity, roundness, and inertia ratio to 0.99, and performing contour screening through a strict valid parameter range; if after clustering, the number of contours that meet the requirements is consistent with the number of standard circles of the calibration plate, the merged center coordinates of the current clustering results are directly used to determine the circle center coordinates; if the number is insufficient, the valid parameter range is gradually relaxed, the target circle contour is re-searched, and subsequent processes are executed; when the number of contours exceeds the number of standard circles, secondary clustering is initiated through the above-mentioned clustering process, and contours with high geometric quality are always prioritized. Through this hierarchical processing mechanism, this application not only ensures the accuracy of circle center positioning (only high-confidence contours are integrated), but also ensures complete coverage of all target circles through dynamic threshold adjustment. The weighted center coordinates finally output can accurately reflect the true circle center distribution of the calibration plate.

[0152] In order to verify that this solution can more accurately identify the center coordinates of the circle, this application is verified in the following way:

[0153] 1. Use the method described in this solution and the traditional ellipse fitting method to extract the pixel coordinates of the circle centers in the image, establish a coordinate system with the upper left corner of the calibration plate as the origin, and determine the physical position (in millimeters) of each circle center in this coordinate system.

[0154] 2. Based on the correspondence between two-dimensional pixel coordinates and three-dimensional physical coordinates, the Zhang Zhengyou calibration method is used to calculate the camera parameters and external parameters obtained by the two methods respectively. The camera parameters include focal length, principal point coordinates and distortion coefficient, and the external parameters include rotation vector and translation vector.

[0155] 3. Back-project the circle center coordinates in the calibration plate coordinate system onto the image plane using the calculated internal and external parameters to obtain the theoretical pixel coordinates. The Euclidean distance between the theoretical pixel coordinates and the actual detected circle center coordinates is then calculated as the reprojection error. Comparison results show that this approach achieves significantly lower calibration errors than the standard ellipse fitting method, resulting in a more accurate center extraction than traditional methods.

[0156] In summary, this application achieves sub-pixel precision in circle center positioning by combining multi-threshold iterative detection with geometric feature screening. Its dynamic threshold adjustment mechanism can adapt to different lighting and imaging conditions. It uses a confidence-weighted fusion algorithm to effectively suppress noise interference, and cooperates with a hierarchical clustering strategy to ensure that all calibration points can still be accurately identified under complex backgrounds. This application uses perspective transformation correction technology to effectively solve the deformation problem of the calibration plate during tilted shooting, so that the center detection is not affected by the tilt of the viewing angle. Finally, the accuracy of the extracted center is verified by the reprojection error. The whole method significantly improves the calibration accuracy and robustness while maintaining the degree of automation. It is particularly suitable for application scenarios such as high-precision visual measurement and three-dimensional reconstruction.

[0157] See also Figure 5 , Figure 5 This is a schematic diagram of the structure of a calibration plate center detection device provided in an embodiment of the present application, which specifically includes:

[0158] An acquisition module 11 is used to acquire an original calibration plate image;

[0159] The recognition module 12 is used to recognize the quadrilateral corner points from the original calibration plate image;

[0160] a matrix determination module 13, configured to create a virtual standard square and determine a perspective transformation matrix based on virtual corner points of the virtual standard square and corner points of the quadrilateral;

[0161] A mapping module 14 is configured to map the original calibration plate image into a square calibration plate image using the perspective transformation matrix;

[0162] A detection module 15 is configured to detect the target circle outline in the square calibration plate image;

[0163] A merging module 16 is configured to divide each target circle contour into at least one cluster subset, and generate a merged center coordinate based on the center coordinate and confidence level of each target circle contour in each cluster subset;

[0164] The coordinate determination module 17 is configured to determine the actual circle center coordinates corresponding to the original calibration plate image using the perspective transformation matrix and the merged center coordinates.

[0165] As an optional embodiment, the identification module includes:

[0166] a first processing unit, configured to perform binarization processing on the original calibration plate image according to a first target threshold value to generate a processed first calibration plate binarized image;

[0167] A first detection unit, configured to detect a quadrilateral outline of the first calibration plate binary image;

[0168] A screening unit, configured to screen quadrilateral corner points according to the cosine values ​​of the vertices of the quadrilateral outline;

[0169] The adjusting unit is configured to adjust the first target threshold when the number of the screened quadrilateral corner points is less than four, so that the first processing unit continues to perform binarization processing on the original calibration plate image according to the adjusted first target threshold.

[0170] As an optional embodiment, the circle center detection device further includes:

[0171] A region determination module is used to determine a rectangular region of interest of each quadrilateral corner point with each quadrilateral corner point as the center;

[0172] A processing module is used to perform binarization processing on each rectangular region of interest to generate a binary image of the rectangular region;

[0173] A conversion module, used to convert the binary images of each rectangular area into corresponding two-dimensional point cloud data;

[0174] A clustering module is used to determine the target point cloud data subset closest to the quadrilateral corner point in the two-dimensional point cloud data through a clustering algorithm, and to determine the fitting edge segment of the target point cloud data subset;

[0175] A corner point determination module is configured to detect that the angle between the fitted edge segments of the target rectangular region of interest is less than an angle threshold, and then use the quadrilateral corner points corresponding to the target rectangular region of interest as triangular marker corner points;

[0176] The sequence determination module is used to determine the corner point input sequence according to the triangle marked corner points; wherein the corner point input sequence is used to determine the input sequence of each quadrilateral corner point when determining the perspective transformation matrix.

[0177] As an optional embodiment, the detection module includes:

[0178] a second processing unit, configured to perform binarization processing on the square calibration plate image according to a second target threshold value to generate a processed second calibration plate binarized image; wherein the value of the second target threshold value is different in each iteration process;

[0179] A second detection unit is used to detect the initial circle contour in the binary image of the second calibration plate;

[0180] a calculation unit, configured to calculate the area and a second geometric characteristic parameter of each initial circular contour; the second geometric characteristic parameter comprising at least one of roundness, moment of inertia ratio, and convexity;

[0181] The determining unit is configured to use the initial circular contour as the target circular contour if the area of ​​the initial circular contour is within the valid area range and the second geometric characteristic parameter is within the valid parameter range.

[0182] As an optional embodiment, the circle center detection device further includes:

[0183] A line segment determination module is used to perform edge detection on the square calibration plate image, extract edge pixels, and determine the target line segment based on the edge pixels;

[0184] A search module is used to search for a closed target line segment according to the first geometric feature parameter of each target line segment; if so, the fitting module is triggered;

[0185] The fitting module is used to perform circle fitting on the closed target line segment and generate the fitted target circle;

[0186] The range determination module is used to determine an area index according to the area of ​​each target circle, and determine an effective area range according to the area index.

[0187] As an optional embodiment, the merging module is specifically configured to:

[0188] According to the second geometric feature parameter of the target circle contour, the confidence of each target circle contour is calculated; according to the area and pixel value of the target circle contour, the center coordinates of each target circle contour are calculated; according to the contour radius, center coordinates and center distance threshold of each target circle contour in each iteration process, each target circle contour is divided into at least one cluster subset.

[0189] As an optional embodiment, the circle center detection device further includes:

[0190] A judgment module is used to judge whether the number of cluster subsets is the same as the number of standard circles;

[0191] If so, the coordinate determination module is triggered; if not, the effective parameter range is adjusted, and the detection module re-determines the target circle contour based on the effective area range and the adjusted effective parameter range.

[0192] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0193] See also Figure 6 , Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application includes a processor 21, a communication interface 22, a memory 23, and a communication bus 24, wherein the processor 21, the communication interface 22, and the memory 23 communicate with each other via the communication bus 24;

[0194] Memory 23, for storing computer programs;

[0195] The processor 21 is used to execute the program stored in the memory 23 to implement the steps of the circle center detection method described in any of the above method embodiments, which will not be repeated here.

[0196] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0197] The communication interface is used for communication between the above terminal and other devices.

[0198] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0199] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0200] In another exemplary embodiment, a computer storage medium is provided. When executed by a processor, the program instructions implement the steps of the circle center detection method described in any of the above method embodiments. The storage medium may include any medium capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0201] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment will not be described in detail here.

[0202] It should be understood that the terms used herein are for the purpose of describing specific example embodiments only and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "comprise", "include", "contain" and "have" are inclusive and therefore specify the presence of stated features, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring them to be performed in the specific order described or illustrated, unless the order of execution is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0203] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is intended to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for detecting the center of a calibration plate, characterized in that: include: Get the original calibration plate image; Identifying quadrilateral corner points from the original calibration plate image; Creating a virtual standard square and determining a perspective transformation matrix based on the virtual corner points of the virtual standard square and the corner points of the quadrilateral; wherein, based on the square characteristic of the outer frame of the calibration plate, an ideal square is defined in the target space as a virtual projection plane, and the coordinates of the four corner points of the square are set according to the standard square geometry, and the square is referred to as the virtual standard square; Mapping the original calibration plate image into a square calibration plate image using the perspective transformation matrix; Detecting the target circle outline in the square calibration plate image; Divide each target circle contour into at least one cluster subset, and generate a merged center coordinate based on the center coordinate and confidence level of each target circle contour in each cluster subset; Determine the actual circle center coordinates corresponding to the original calibration plate image using the perspective transformation matrix and the merged center coordinates; After identifying the quadrilateral corner points from the original calibration plate image, the method further includes: Taking each quadrilateral corner point as the center, determine the rectangular region of interest of each quadrilateral corner point; Perform binarization processing on each rectangular region of interest to generate a rectangular region binary image; Convert the binary images of each rectangular area into corresponding two-dimensional point cloud data; By using a clustering algorithm, a target point cloud data subset that is closest to the quadrilateral corner point in the two-dimensional point cloud data is determined, and a fitting edge segment of the target point cloud data subset is determined; If the angle between the fitted edge segments of the target rectangular region of interest is detected to be less than the angle threshold, the quadrilateral corner points corresponding to the target rectangular region of interest are used as triangular marker corner points; Determining a corner point input order based on the triangular marked corner points; wherein the corner point input order is used to determine the input order of each quadrilateral corner point when determining the perspective transformation matrix; Before determining the actual center coordinates corresponding to the original calibration plate image, the method further includes: Determine whether the number of cluster subsets is the same as the number of standard circles; If yes, continue to perform the step of using the perspective transformation matrix and the merged center coordinates to determine the actual circle center coordinates corresponding to the original calibration plate image; If not, the effective parameter range is adjusted, and the target circle contour is re-determined based on the effective area range and the adjusted effective parameter range.

2. The circle center detection method according to claim 1, wherein: The identifying quadrilateral corner points from the original calibration plate image includes: Binarizing the original calibration plate image according to a first target threshold to generate a processed first calibration plate binary image; Detecting the quadrilateral outline of the first calibration plate binary image; Filtering quadrilateral corner points according to the cosine values ​​of the vertices of the quadrilateral outline; If the number of the screened quadrilateral corner points is less than four, the first target threshold is adjusted, and the step of binarizing the original calibration plate image is continued based on the adjusted first target threshold.

3. The circle center detection method according to claim 1 or 2, characterized in that: Detecting the target circle contour in the square calibration plate image includes: Binarizing the square calibration plate image according to a second target threshold to generate a processed second calibration plate binary image; wherein the value of the second target threshold is different in each iteration process; Detecting an initial circle contour in the binary image of the second calibration plate; Calculating the area and second geometric characteristic parameters of each initial circular contour; the second geometric characteristic parameters include at least one of roundness, inertia ratio, and convexity; If the area of ​​the initial circular contour is within the valid area range and the second geometric feature parameter is within the valid parameter range, the initial circular contour is used as the target circular contour.

4. The circle center detection method according to claim 3, characterized in that: The process of determining the effective area range includes: Performing edge detection on the square calibration plate image, extracting edge pixels, and determining target line segments based on the edge pixels; Searching for closed target line segments according to the first geometric characteristic parameters of each target line segment; If so, perform circle fitting on the closed target line segment to generate a fitted target circle; An area index is determined according to the area of ​​each target circle, and an effective area range is determined according to the area index.

5. The circle center detection method according to claim 3, characterized in that: Divide each target circle contour into at least one cluster subset, including: Calculating the confidence level of each target circle contour according to the second geometric characteristic parameter of the target circle contour; Calculate the center coordinates of each target circle contour according to the area and pixel value of the target circle contour; According to the contour radius, center coordinates and center distance threshold of each target circle contour in each iteration process, each target circle contour is divided into at least one cluster subset.

6. A device for detecting the center of a calibration plate, characterized in that: include: An acquisition module is used to obtain the original calibration plate image; A recognition module, configured to recognize quadrilateral corner points from the original calibration plate image; a matrix determination module, configured to create a virtual standard square and determine a perspective transformation matrix based on the virtual corner points of the virtual standard square and the corner points of the quadrilateral; wherein, based on the characteristic that the outer border of the calibration plate is a square, an ideal square is defined in the target space as a virtual projection plane, and the coordinates of the four corner points of the square are set according to the standard square geometry, and the square is referred to as the virtual standard square; A mapping module, configured to map the original calibration plate image into a square calibration plate image using the perspective transformation matrix; A detection module, configured to detect the target circle outline in the square calibration plate image; a merging module, configured to divide each target circle contour into at least one cluster subset, and generate a merged center coordinate based on the center coordinates and confidence levels of each target circle contour in each cluster subset; A coordinate determination module, configured to determine the actual circle center coordinates corresponding to the original calibration plate image using the perspective transformation matrix and the merged center coordinates; Wherein, the center detection device further includes: A region determination module is used to determine a rectangular region of interest of each quadrilateral corner point with each quadrilateral corner point as the center; A processing module is used to perform binarization processing on each rectangular region of interest to generate a binary image of the rectangular region; A conversion module, used to convert the binary images of each rectangular area into corresponding two-dimensional point cloud data; A clustering module is used to determine the target point cloud data subset closest to the quadrilateral corner point in the two-dimensional point cloud data through a clustering algorithm, and to determine the fitting edge segment of the target point cloud data subset; A corner point determination module is configured to detect that the angle between the fitted edge segments of the target rectangular region of interest is less than an angle threshold, and then use the quadrilateral corner points corresponding to the target rectangular region of interest as triangular marker corner points; An order determination module is used to determine the order of corner point input according to the triangular marked corner points; wherein the order of corner point input is used to determine the order of input of each quadrilateral corner point when determining the perspective transformation matrix; Wherein, the center detection device further includes: A judgment module is used to judge whether the number of cluster subsets is the same as the number of standard circles; If so, the coordinate determination module is triggered; if not, the effective parameter range is adjusted, and the detection module re-determines the target circle contour based on the effective area range and the adjusted effective parameter range.

7. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor is configured to implement the steps of the circle center detection method described in any one of claims 1 to 5 when executing a program stored in a memory.

8. A computer storage medium, characterized in that The computer storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the steps of the circle center detection method described in any one of claims 1 to 5 of the present application.

Images