C# OpenCvSharp DNN Yolov8-OBB 旋转目标检测

目录

效果

模型信息

项目

代码

下载

C# OpenCvSharp DNN Yolov8-OBB 旋转目标检测

效果

模型信息

Model Properties
-------------------------
date：2024-02-26T08:38:44.171849
description：Ultralytics YOLOv8s-obb model trained on runs/DOTAv1.0-ms.yaml
author：Ultralytics
task：obb
license：AGPL-3.0 https://ultralytics.com/license
version：8.1.18
stride：32
batch：1
imgsz：[640, 640]
names：{0: 'plane', 1: 'ship', 2: 'storage tank', 3: 'baseball diamond', 4: 'tennis court', 5: 'basketball court', 6: 'ground track field', 7: 'harbor', 8: 'bridge', 9: 'large vehicle', 10: 'small vehicle', 11: 'helicopter', 12: 'roundabout', 13: 'soccer ball field', 14: 'swimming pool'}
---------------------------------------------------------------

Inputs
-------------------------
name：images
tensor：Float[1, 3, 640, 640]
---------------------------------------------------------------

Outputs
-------------------------
name：output0
tensor：Float[1, 20, 8400]
---------------------------------------------------------------

项目

代码

using OpenCvSharp;
using OpenCvSharp.Dnn;
using System;
using System.Collections.Generic;
using System.Drawing;
using System.IO;
using System.Linq;
using System.Windows.Forms;

namespace OpenCvSharp_DNN_Demo
{
    public partial class frmMain : Form
    {
        public frmMain()
        {
            InitializeComponent();
        }

        string fileFilter = "*.*|*.bmp;*.jpg;*.jpeg;*.tiff;*.tiff;*.png";
        string image_path = "";

        DateTime dt1 = DateTime.Now;
        DateTime dt2 = DateTime.Now;

        string modelpath;
        string classer_path;

        List class_names;

        Net opencv_net;
        Mat BN_image;

        Mat image;
        Mat result_image;

        string[] class_lables;

        private void button1_Click(object sender, EventArgs e)
        {
            OpenFileDialog ofd = new OpenFileDialog();
            ofd.Filter = fileFilter;
            if (ofd.ShowDialog() != DialogResult.OK) return;

            pictureBox1.Image = null;
            pictureBox2.Image = null;
            textBox1.Text = "";

            image_path = ofd.FileName;
            pictureBox1.Image = new Bitmap(image_path);
            image = new Mat(image_path);
        }

        private void Form1_Load(object sender, EventArgs e)
        {
            modelpath = "model/yolov8s-obb.onnx";
            classer_path = "model/lable.txt";

            opencv_net = CvDnn.ReadNetFromOnnx(modelpath);

            List str = new List();
            StreamReader sr = new StreamReader(classer_path);
            string line;
            while ((line = sr.ReadLine()) != null)
            {
                str.Add(line);
            }
            class_lables = str.ToArray();

            image_path = "test_img/1.png";
            pictureBox1.Image = new Bitmap(image_path);

        }

        private  void button2_Click(object sender, EventArgs e)
        {
            if (image_path == "")
            {
                return;
            }
            textBox1.Text = "检测中，请稍等……";
            pictureBox2.Image = null;
            button2.Enabled = false;
            Application.DoEvents();

            image = new Mat(image_path);

            //图片缩放
            image = new Mat(image_path);
            int max_image_length = image.Cols > image.Rows ? image.Cols : image.Rows;
            Mat max_image = Mat.Zeros(new OpenCvSharp.Size(max_image_length, max_image_length), MatType.CV_8UC3);
            Rect roi = new Rect(0, 0, image.Cols, image.Rows);
            image.CopyTo(new Mat(max_image, roi));

            float[] result_array;
            float factor = (float)(max_image_length / 640.0);

            // 将图片转为RGB通道
            Mat image_rgb = new Mat();
            Cv2.CvtColor(max_image, image_rgb, ColorConversionCodes.BGR2RGB);
            Mat resize_image = new Mat();
            Cv2.Resize(image_rgb, resize_image, new OpenCvSharp.Size(640, 640));

            BN_image = CvDnn.BlobFromImage(resize_image, 1 / 255.0, new OpenCvSharp.Size(640, 640), new Scalar(0, 0, 0), true, false);

            //配置图片输入数据
            opencv_net.SetInput(BN_image);

            //模型推理，读取推理结果
            Mat[] outs = new Mat[1] { new Mat() };
            string[] outBlobNames = opencv_net.GetUnconnectedOutLayersNames().ToArray();

            dt1 = DateTime.Now;

            opencv_net.Forward(outs, outBlobNames);

            dt2 = DateTime.Now;

            int num_proposal = outs[0].Size(1);
            int nout = outs[0].Size(2);

            if (outs[0].Dims > 2)
            {
                outs[0] = outs[0].Reshape(0, num_proposal);
            }

            Mat result_data = new Mat(20, 8400, MatType.CV_32F);
            result_data = outs[0].T();
            List position_boxes = new List();
            List class_ids = new List();
            List confidences = new List();
            List rotations = new List();
            // Preprocessing output results
            for (int i = 0; i < result_data.Rows; i++)
            {
                Mat classes_scores = new Mat(result_data, new Rect(4, i, 15, 1));
                OpenCvSharp.Point max_classId_point, min_classId_point;
                double max_score, min_score;
                // Obtain the maximum value and its position in a set of data
                Cv2.MinMaxLoc(classes_scores, out min_score, out max_score,
                    out min_classId_point, out max_classId_point);
                // Confidence level between 0 ~ 1
                // Obtain identification box information
                if (max_score > 0.25)
                {
                    float cx = result_data.At(i, 0);
                    float cy = result_data.At(i, 1);
                    float ow = result_data.At(i, 2);
                    float oh = result_data.At(i, 3);
                    double x = (cx - 0.5 * ow) * factor;
                    double y = (cy - 0.5 * oh) * factor;
                    double width = ow * factor;
                    double height = oh * factor;
                    Rect2d box = new Rect2d();
                    box.X = x;
                    box.Y = y;
                    box.Width = width;
                    box.Height = height;
                    position_boxes.Add(box);
                    class_ids.Add(max_classId_point.X);
                    confidences.Add((float)max_score);
                    rotations.Add(result_data.At(i, 19));
                }
            }

            // NMS 
            int[] indexes = new int[position_boxes.Count];
            CvDnn.NMSBoxes(position_boxes, confidences, 0.25f, 0.7f, out indexes);
            List rotated_rects = new List();
            for (int i = 0; i < indexes.Length; i++)
            {
                int index = indexes[i];
                float w = (float)position_boxes[index].Width;
                float h = (float)position_boxes[index].Height;
                float x = (float)position_boxes[index].X + w / 2;
                float y = (float)position_boxes[index].Y + h / 2;
                float r = rotations[index];
                float w_ = w > h ? w : h;
                float h_ = w > h ? h : w;
                r = (float)((w > h ? r : (float)(r + Math.PI / 2)) % Math.PI);
                RotatedRect rotate = new RotatedRect(new Point2f(x, y), new Size2f(w_, h_), (float)(r * 180.0 / Math.PI));
                rotated_rects.Add(rotate);
            }

            result_image = image.Clone();

            for (int i = 0; i < indexes.Length; i++)
            {
                int index = indexes[i];
                Point2f[] points = rotated_rects[i].Points();

                for (int j = 0; j < 4; j++)
                {
                    Cv2.Line(result_image, (OpenCvSharp.Point)points[j], (OpenCvSharp.Point)points[(j + 1) % 4], new Scalar(0, 255, 0), 2);
                }

                Cv2.PutText(result_image, class_lables[class_ids[index]] + "-" + confidences[index].ToString("0.00"),
                    (OpenCvSharp.Point)points[0], HersheyFonts.HersheySimplex, 0.8, new Scalar(0, 0, 255), 2);
            }

            pictureBox2.Image = new Bitmap(result_image.ToMemoryStream());
            textBox1.Text = "推理耗时:" + (dt2 - dt1).TotalMilliseconds + "ms";

            button2.Enabled = true;
        }

        private void pictureBox2_DoubleClick(object sender, EventArgs e)
        {
            Common.ShowNormalImg(pictureBox2.Image);
        }

        private void pictureBox1_DoubleClick(object sender, EventArgs e)
        {
            Common.ShowNormalImg(pictureBox1.Image);
        }
    }
}

  1. using OpenCvSharp;
  2. using OpenCvSharp.Dnn;
  3. using System;
  4. using System.Collections.Generic;
  5. using System.Drawing;
  6. using System.IO;
  7. using System.Linq;
  8. using System.Windows.Forms;
  9.  
  10. namespace OpenCvSharp_DNN_Demo
  11. {
  12.     public partial class frmMain : Form
  13.     {
  14.         public frmMain()
  15.         {
  16.             InitializeComponent();
  17.         }
  18.  
  19.         string fileFilter = "*.*|*.bmp;*.jpg;*.jpeg;*.tiff;*.tiff;*.png";
  20.         string image_path = "";
  21.  
  22.         DateTime dt1 = DateTime.Now;
  23.         DateTime dt2 = DateTime.Now;
  24.  
  25.         string modelpath;
  26.         string classer_path;
  27.  
  28.         List<string> class_names;
  29.  
  30.         Net opencv_net;
  31.         Mat BN_image;
  32.  
  33.         Mat image;
  34.         Mat result_image;
  35.  
  36.         string[] class_lables;
  37.  
  38.         private void button1_Click(object sender, EventArgs e)
  39.         {
  40.             OpenFileDialog ofd = new OpenFileDialog();
  41.             ofd.Filter = fileFilter;
  42.             if (ofd.ShowDialog() != DialogResult.OK) return;
  43.  
  44.             pictureBox1.Image = null;
  45.             pictureBox2.Image = null;
  46.             textBox1.Text = "";
  47.  
  48.             image_path = ofd.FileName;
  49.             pictureBox1.Image = new Bitmap(image_path);
  50.             image = new Mat(image_path);
  51.         }
  52.  
  53.         private void Form1_Load(object sender, EventArgs e)
  54.         {
  55.             modelpath = "model/yolov8s-obb.onnx";
  56.             classer_path = "model/lable.txt";
  57.  
  58.             opencv_net = CvDnn.ReadNetFromOnnx(modelpath);
  59.  
  60.             List<string> str = new List<string>();
  61.             StreamReader sr = new StreamReader(classer_path);
  62.             string line;
  63.             while ((line = sr.ReadLine()) != null)
  64.             {
  65.                 str.Add(line);
  66.             }
  67.             class_lables = str.ToArray();
  68.  
  69.             image_path = "test_img/1.png";
  70.             pictureBox1.Image = new Bitmap(image_path);
  71.  
  72.         }
  73.  
  74.         private  void button2_Click(object sender, EventArgs e)
  75.         {
  76.             if (image_path == "")
  77.             {
  78.                 return;
  79.             }
  80.             textBox1.Text = "检测中，请稍等……";
  81.             pictureBox2.Image = null;
  82.             button2.Enabled = false;
  83.             Application.DoEvents();
  84.  
  85.             image = new Mat(image_path);
  86.  
  87.             //图片缩放
  88.             image = new Mat(image_path);
  89.             int max_image_length = image.Cols > image.Rows ? image.Cols : image.Rows;
  90.             Mat max_image = Mat.Zeros(new OpenCvSharp.Size(max_image_length, max_image_length), MatType.CV_8UC3);
  91.             Rect roi = new Rect(0, 0, image.Cols, image.Rows);
  92.             image.CopyTo(new Mat(max_image, roi));
  93.  
  94.             float[] result_array;
  95.             float factor = (float)(max_image_length / 640.0);
  96.  
  97.             // 将图片转为RGB通道
  98.             Mat image_rgb = new Mat();
  99.             Cv2.CvtColor(max_image, image_rgb, ColorConversionCodes.BGR2RGB);
  100.             Mat resize_image = new Mat();
  101.             Cv2.Resize(image_rgb, resize_image, new OpenCvSharp.Size(640, 640));
  102.  
  103.             BN_image = CvDnn.BlobFromImage(resize_image, 1 / 255.0, new OpenCvSharp.Size(640, 640), new Scalar(0, 0, 0), true, false);
  104.  
  105.             //配置图片输入数据
  106.             opencv_net.SetInput(BN_image);
  107.  
  108.             //模型推理，读取推理结果
  109.             Mat[] outs = new Mat[1] { new Mat() };
  110.             string[] outBlobNames = opencv_net.GetUnconnectedOutLayersNames().ToArray();
  111.  
  112.             dt1 = DateTime.Now;
  113.  
  114.             opencv_net.Forward(outs, outBlobNames);
  115.  
  116.             dt2 = DateTime.Now;
  117.  
  118.             int num_proposal = outs[0].Size(1);
  119.             int nout = outs[0].Size(2);
  120.  
  121.             if (outs[0].Dims > 2)
  122.             {
  123.                 outs[0] = outs[0].Reshape(0, num_proposal);
  124.             }
  125.  
  126.             Mat result_data = new Mat(20, 8400, MatType.CV_32F);
  127.             result_data = outs[0].T();
  128.             List<Rect2d> position_boxes = new List<Rect2d>();
  129.             List<int> class_ids = new List<int>();
  130.             List<float> confidences = new List<float>();
  131.             List<float> rotations = new List<float>();
  132.             // Preprocessing output results
  133.             for (int i = 0; i < result_data.Rows; i++)
  134.             {
  135.                 Mat classes_scores = new Mat(result_data, new Rect(4, i, 15, 1));
  136.                 OpenCvSharp.Point max_classId_point, min_classId_point;
  137.                 double max_score, min_score;
  138.                 // Obtain the maximum value and its position in a set of data
  139.                 Cv2.MinMaxLoc(classes_scores, out min_score, out max_score,
  140.                     out min_classId_point, out max_classId_point);
  141.                 // Confidence level between 0 ~ 1
  142.                 // Obtain identification box information
  143.                 if (max_score > 0.25)
  144.                 {
  145.                     float cx = result_data.At<float>(i, 0);
  146.                     float cy = result_data.At<float>(i, 1);
  147.                     float ow = result_data.At<float>(i, 2);
  148.                     float oh = result_data.At<float>(i, 3);
  149.                     double x = (cx - 0.5 * ow) * factor;
  150.                     double y = (cy - 0.5 * oh) * factor;
  151.                     double width = ow * factor;
  152.                     double height = oh * factor;
  153.                     Rect2d box = new Rect2d();
  154.                     box.X = x;
  155.                     box.Y = y;
  156.                     box.Width = width;
  157.                     box.Height = height;
  158.                     position_boxes.Add(box);
  159.                     class_ids.Add(max_classId_point.X);
  160.                     confidences.Add((float)max_score);
  161.                     rotations.Add(result_data.At<float>(i, 19));
  162.                 }
  163.             }
  164.  
  165.             // NMS 
  166.             int[] indexes = new int[position_boxes.Count];
  167.             CvDnn.NMSBoxes(position_boxes, confidences, 0.25f, 0.7f, out indexes);
  168.             List<RotatedRect> rotated_rects = new List<RotatedRect>();
  169.             for (int i = 0; i < indexes.Length; i++)
  170.             {
  171.                 int index = indexes[i];
  172.                 float w = (float)position_boxes[index].Width;
  173.                 float h = (float)position_boxes[index].Height;
  174.                 float x = (float)position_boxes[index].X + w / 2;
  175.                 float y = (float)position_boxes[index].Y + h / 2;
  176.                 float r = rotations[index];
  177.                 float w_ = w > h ? w : h;
  178.                 float h_ = w > h ? h : w;
  179.                 r = (float)((w > h ? r : (float)(r + Math.PI / 2)) % Math.PI);
  180.                 RotatedRect rotate = new RotatedRect(new Point2f(x, y), new Size2f(w_, h_), (float)(r * 180.0 / Math.PI));
  181.                 rotated_rects.Add(rotate);
  182.             }
  183.  
  184.             result_image = image.Clone();
  185.  
  186.             for (int i = 0; i < indexes.Length; i++)
  187.             {
  188.                 int index = indexes[i];
  189.                 Point2f[] points = rotated_rects[i].Points();
  190.  
  191.                 for (int j = 0; j < 4; j++)
  192.                 {
  193.                     Cv2.Line(result_image, (OpenCvSharp.Point)points[j], (OpenCvSharp.Point)points[(j + 1) % 4], new Scalar(0, 255, 0), 2);
  194.                 }
  195.  
  196.                 Cv2.PutText(result_image, class_lables[class_ids[index]] + "-" + confidences[index].ToString("0.00"),
  197.                     (OpenCvSharp.Point)points[0], HersheyFonts.HersheySimplex, 0.8, new Scalar(0, 0, 255), 2);
  198.             }
  199.  
  200.             pictureBox2.Image = new Bitmap(result_image.ToMemoryStream());
  201.             textBox1.Text = "推理耗时:" + (dt2 - dt1).TotalMilliseconds + "ms";
  202.  
  203.             button2.Enabled = true;
  204.         }
  205.  
  206.         private void pictureBox2_DoubleClick(object sender, EventArgs e)
  207.         {
  208.             Common.ShowNormalImg(pictureBox2.Image);
  209.         }
  210.  
  211.         private void pictureBox1_DoubleClick(object sender, EventArgs e)
  212.         {
  213.             Common.ShowNormalImg(pictureBox1.Image);
  214.         }
  215.     }
  216. }

下载

源码下载

天天代码码天天
微信公众号
.NET 人工智能实践
注：本文转载自blog.csdn.net的天天代码码天天的文章"https://lw112190.blog.csdn.net/article/details/136298717"。版权归原作者所有，此博客不拥有其著作权，亦不承担相应法律责任。如有侵权，请联系我们删除。
复制链接
相关推荐
发表评论

评论记录：

未查询到任何数据！