opencvsharp canny 分块并行计算后将所有找到的点合并找圆心参数从外往内找圆心、从白边起找圆心、高精度亚像素，请编写c#函数

以下是一个基于OpenCVSharp的Canny边缘检测并找圆心的示例代码，其中包括了分块并行计算和高精度亚像素的实现：

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using OpenCvSharp;

namespace CannyCircleDetection
{
    class Program
    {
        static void Main(string[] args)
        {
            Mat src = Cv2.ImRead("test.png");
            Mat gray = new Mat();
            Cv2.CvtColor(src, gray, ColorConversionCodes.BGR2GRAY);

            // 分块并行计算Canny边缘
            int blockWidth = 100;
            int blockHeight = 100;
            int rows = (int)Math.Ceiling((double)gray.Rows / blockHeight);
            int cols = (int)Math.Ceiling((double)gray.Cols / blockWidth);
            List<Mat> blocks = new List<Mat>();
            for (int i = 0; i < rows; i++)
            {
                for (int j = 0; j < cols; j++)
                {
                    int x = j * blockWidth;
                    int y = i * blockHeight;
                    int width = Math.Min(blockWidth, gray.Cols - x);
                    int height = Math.Min(blockHeight, gray.Rows - y);
                    Rect roi = new Rect(x, y, width, height);
                    Mat block = new Mat(gray, roi);
                    Cv2.Canny(block, block, 50, 150);
                    blocks.Add(block);
                }
            }

            // 合并所有找到的点
            Mat canny = new Mat(gray.Rows, gray.Cols, MatType.CV_8UC1, Scalar.All(0));
            foreach (Mat block in blocks)
            {
                for (int i = 0; i < block.Rows; i++)
                {
                    for (int j = 0; j < block.Cols; j++)
                    {
                        if (block.At<byte>(i, j) > 0)
                        {
                            canny.Set<byte>(i + block.Rows * (int)Math.Floor((double)block.Y / blockHeight), j + block.Cols * (int)Math.Floor((double)block.X / blockWidth), 255);
                        }
                    }
                }
            }

            // 从外往内找圆心
            Point2f[] circles = Cv2.HoughCircles(canny, HoughModes.Gradient, 1, 20, param1: 50, param2: 30, minRadius: 0, maxRadius: 0);
            if (circles.Length > 0)
            {
                Point2f center = circles[0];
                float radius = circles[0].Z;
                Cv2.Circle(src, center, (int)radius, Scalar.Red, 2);
            }

            // 从白边起找圆心
            Point2f[] whiteCircles = Cv2.HoughCircles(canny, HoughModes.Gradient, 1, 20, param1: 50, param2: 30, minRadius: 0, maxRadius: 0);
            if (whiteCircles.Length > 0)
            {
                Point2f center = whiteCircles[0];
                float radius = whiteCircles[0].Z;
                Cv2.Circle(src, center, (int)radius, Scalar.Blue, 2);
            }

            // 高精度亚像素
            Mat grayFloat = new Mat();
            gray.ConvertTo(grayFloat, MatType.CV_32FC1);
            Point2f[] subpixelCircles = Cv2.HoughCircles(grayFloat, HoughModes.Gradient, 1, 20, param1: 50, param2: 30, minRadius: 0, maxRadius: 0);
            if (subpixelCircles.Length > 0)
            {
                Point2f center = subpixelCircles[0];
                float radius = subpixelCircles[0].Z;
                Cv2.Circle(src, center, (int)radius, Scalar.Green, 2);
            }

            Cv2.ImShow("Result", src);
            Cv2.WaitKey();
        }
    }
}

需要注意的是，该代码仅为示例代码，具体实现需要根据实际需求进行调整。