OpenCvSharp+Yolov5Net+Onnx 完整Demo

效果

工程

代码

using Microsoft.ML.OnnxRuntime;
using OpenCvSharp;
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading;
using System.Windows.Forms;

namespace OpenCvSharp_Yolov5Net_Onnx
{
    public partial class Form1 : Form
    {
        public Form1()
        {
            InitializeComponent();
        }

        private string PicFilter = "*.*|*.bmp;*.jpg;*.jpeg;*.tiff;*.tiff;*.png";
        private void button1_Click(object sender, EventArgs e)
        {
            OpenFileDialog ofd = new OpenFileDialog();
            ofd.Filter = PicFilter;
            if (ofd.ShowDialog() != DialogResult.OK) return;

            var imagefile = ofd.FileName;
            var image = Image.FromFile(imagefile);
            Yolov5Net.Scorer.YoloScorer<Yolov5Net.Scorer.Models.YoloCocoP5Model> scorer = new Yolov5Net.Scorer.YoloScorer<Yolov5Net.Scorer.Models.YoloCocoP5Model>("yolov5s.onnx");

            List<Yolov5Net.Scorer.YoloPrediction> predictions = scorer.Predict(image);

            var graphics = Graphics.FromImage(image);

            foreach (var prediction in predictions) // iterate predictions to draw results
            {
                double score = Math.Round(prediction.Score, 2);

                graphics.DrawRectangles(new Pen(prediction.Label.Color, 2), new[] { prediction.Rectangle });

                PointF p = new PointF(prediction.Rectangle.X - 3, prediction.Rectangle.Y - 23);

                graphics.DrawString(prediction.Label.Name + " (" + score + ")"
                    , new Font("Arial", 16, GraphicsUnit.Pixel)
                    , new SolidBrush(prediction.Label.Color)
                    , p);
            }

            pictureBox1.Image = image;
        }

        FrameSource frame;
        Thread t;
        private void button2_Click(object sender, EventArgs e)
        {
            //获取视频设备
            frame = Cv2.CreateFrameSource_Camera(0);
            t = new Thread(CameraDet);
            t.Start();
        }

        private void CameraDet()
        {

            Mat mat = new Mat();
            //yolov5+onnx检测
            var scorer = new Yolov5Net.Scorer.YoloScorer<Yolov5Net.Scorer.Models.YoloCocoP5Model>("yolov5s.onnx");

            //逐帧放入
            while (true)
            {
                if (frame.IsDisposed)
                {
                    return;
                }
                //从摄像头读视频帧
                frame.NextFrame(mat);
                //先将Mat图像转换成Stream流，再将流转成Bitmap
                var image = new Bitmap(mat.ToMemoryStream()) as Image;
                //var image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(mat);
                List<Yolov5Net.Scorer.YoloPrediction> predictions = scorer.Predict(image);
                //在图片上绘制目标标识
                var graphics = Graphics.FromImage(image);
                foreach (var prediction in predictions) // iterate predictions to draw results
                {
                    double score = Math.Round(prediction.Score, 2);

                    graphics.DrawRectangles(new Pen(prediction.Label.Color, 2),
                        new[] { prediction.Rectangle });

                    var (x, y) = (prediction.Rectangle.X - 3, prediction.Rectangle.Y - 23);

                    graphics.DrawString($"{prediction.Label.Name} ({score})",
                        new Font("Arial", 16, GraphicsUnit.Pixel), new SolidBrush(prediction.Label.Color),
                        new PointF(x, y));

                    pictureBox1.Image = image;
                }
            }
        }

        private void Form1_FormClosing(object sender, FormClosingEventArgs e)
        {
            if (frame != null)
            {
                frame.Dispose();
                frame = null;
                t.Abort();
            }
        }

    }
}

Demo下载​​​​​​​