当前位置： 首页 > news >正文

C# OpenCvSharp DNN UNet 推理

news 2025/2/28 15:04:17

目录

效果

模型

项目

代码

下载

效果

模型

Inputs
-------------------------
name：data
tensor：Float[1, 3, 256, 256]
---------------------------------------------------------------

Outputs
-------------------------
name：predict
tensor：Float[1, 2, 256, 256]
---------------------------------------------------------------

项目

代码

using OpenCvSharp;
using OpenCvSharp.Dnn;
using System;
using System.Drawing;
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;

        int inpHeight;
        int inpWidth;

        Net opencv_net;
        Mat BN_image;

        Mat image;
        Mat result_image;

        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)
        {
            string modelTxt = "model/unet.prototxt";
            string modelBin = "model/unet.caffemodel";

            inpHeight = 256;
            inpWidth = 256;

            opencv_net = CvDnn.ReadNetFromCaffe(modelTxt, modelBin);

            image_path = "test_img/person.jpg";
            pictureBox1.Image = new Bitmap(image_path);

        }

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

            Mat src = new Mat(image_path);

            int max_image_length = src.Cols > src.Rows ? src.Cols : src.Rows;
            Mat max_image = Mat.Zeros(new OpenCvSharp.Size(max_image_length, max_image_length), MatType.CV_8UC3);
            Rect roi = new Rect(0, 0, src.Cols, src.Rows);
            src.CopyTo(new Mat(max_image, roi));

            Mat resize_image = max_image.Resize(new OpenCvSharp.Size(256, 256));

            BN_image = CvDnn.BlobFromImage(resize_image, 1 / 255.0, new OpenCvSharp.Size(inpWidth, inpHeight), new Scalar(127.5, 127.5, 127.5), true, false);

            //float* ptr = (float*)BN_image.Data;
            //for (int i = 0; i < 10; i++)
            //{
            //    Console.WriteLine(ptr[i]);
            //}

            opencv_net.SetInput(BN_image, "data");


            dt1 = DateTime.Now;

            Mat detection = opencv_net.Forward("predict");


            //float* ptr2 = (float*)detection.Data;
            //for (int i = 0; i < 10; i++)
            //{
            //    Console.WriteLine(ptr2[i]);
            //}

            dt2 = DateTime.Now;

            //得到的输出是一个四维的mat格式数据，大小为[1，2, 256, 256]
            //首先将他reshape，设置成一通道，512行，256列，其中前256行与后256行是互补关系，对应位置相加都为1
            //前256行为背景的概率，后256行为人像的概率
            Mat newMat = detection.Reshape(1, 512);
            //获取人像概率矩阵
            newMat = newMat.RowRange(256, 512);

            Mat result = new Mat();
            newMat.ConvertTo(result, MatType.CV_8U, 255.0);

            Cv2.Threshold(result, result, 127, 255, ThresholdTypes.Binary);

            Mat result2 = Mat.Zeros(256, 256, MatType.CV_8UC3) * 255;

            resize_image.CopyTo(result2, result);

            Cv2.ImShow("黑白", result);
            Cv2.ImShow("扣取", result2);

            pictureBox2.Image = new Bitmap(result2.ToMemoryStream());
            textBox1.Text = "推理耗时:" + (dt2 - dt1).TotalMilliseconds + "ms";
        }

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

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


/*
Inputs
-------------------------
name：data
tensor：Float[1, 3, 256, 256]
---------------------------------------------------------------

Outputs
-------------------------
name：predict
tensor：Float[1, 2, 256, 256]
---------------------------------------------------------------
 */

using OpenCvSharp;
using OpenCvSharp.Dnn;
using System;
using System.Drawing;
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;int inpHeight;int inpWidth;Net opencv_net;Mat BN_image;Mat image;Mat result_image;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){string modelTxt = "model/unet.prototxt";string modelBin = "model/unet.caffemodel";inpHeight = 256;inpWidth = 256;opencv_net = CvDnn.ReadNetFromCaffe(modelTxt, modelBin);image_path = "test_img/person.jpg";pictureBox1.Image = new Bitmap(image_path);}private unsafe void button2_Click(object sender, EventArgs e){if (image_path == ""){return;}textBox1.Text = "检测中，请稍等……";pictureBox2.Image = null;Application.DoEvents();Mat src = new Mat(image_path);int max_image_length = src.Cols > src.Rows ? src.Cols : src.Rows;Mat max_image = Mat.Zeros(new OpenCvSharp.Size(max_image_length, max_image_length), MatType.CV_8UC3);Rect roi = new Rect(0, 0, src.Cols, src.Rows);src.CopyTo(new Mat(max_image, roi));Mat resize_image = max_image.Resize(new OpenCvSharp.Size(256, 256));BN_image = CvDnn.BlobFromImage(resize_image, 1 / 255.0, new OpenCvSharp.Size(inpWidth, inpHeight), new Scalar(127.5, 127.5, 127.5), true, false);//float* ptr = (float*)BN_image.Data;//for (int i = 0; i < 10; i++)//{//    Console.WriteLine(ptr[i]);//}opencv_net.SetInput(BN_image, "data");dt1 = DateTime.Now;Mat detection = opencv_net.Forward("predict");//float* ptr2 = (float*)detection.Data;//for (int i = 0; i < 10; i++)//{//    Console.WriteLine(ptr2[i]);//}dt2 = DateTime.Now;//得到的输出是一个四维的mat格式数据，大小为[1，2, 256, 256]//首先将他reshape，设置成一通道，512行，256列，其中前256行与后256行是互补关系，对应位置相加都为1//前256行为背景的概率，后256行为人像的概率Mat newMat = detection.Reshape(1, 512);//获取人像概率矩阵newMat = newMat.RowRange(256, 512);Mat result = new Mat();newMat.ConvertTo(result, MatType.CV_8U, 255.0);Cv2.Threshold(result, result, 127, 255, ThresholdTypes.Binary);Mat result2 = Mat.Zeros(256, 256, MatType.CV_8UC3) * 255;resize_image.CopyTo(result2, result);Cv2.ImShow("黑白", result);Cv2.ImShow("扣取", result2);pictureBox2.Image = new Bitmap(result2.ToMemoryStream());textBox1.Text = "推理耗时:" + (dt2 - dt1).TotalMilliseconds + "ms";}private void pictureBox2_DoubleClick(object sender, EventArgs e){Common.ShowNormalImg(pictureBox2.Image);}private void pictureBox1_DoubleClick(object sender, EventArgs e){Common.ShowNormalImg(pictureBox1.Image);}}
}/*
Inputs
-------------------------
name：data
tensor：Float[1, 3, 256, 256]
---------------------------------------------------------------Outputs
-------------------------
name：predict
tensor：Float[1, 2, 256, 256]
---------------------------------------------------------------*/

下载

源码下载


http://www.mrgr.cn/news/61411.html

相关文章：

  • 2024年【通信安全员ABC证】最新解析及通信安全员ABC证新版试题
  • qt的c++环境配置和c++基础【正点原子】嵌入式Qt5 C++开发视频
  • 【AIGC】2024-arXiv-Lumiere：视频生成的时空扩散模型
  • 开始菜单增强工具 StartAllBack v3.7.10.4910 直装激活版
  • dubbo介绍
  • 13.音乐管理系统(基于SpringBoot + Vue)
  • YoloV9改进策略：Block改进|RFE模块，提高小物体的识别精度|即插即用|代码+修改过程
  • 抽取picomax的设备树
  • Leetcode 第 142 场双周赛题解
  • leetcode57:插入区间
  • 明日周刊-第25期
  • Docker方式部署ClickHouse
  • 大数据新视界 -- 大数据大厂之大数据重塑影视娱乐产业的未来（4 - 4）
  • 基于Mysql、JavaScript、PHP、ajax开发的MBTI性格测试网站（前端+后端）
  • Linux shell编程学习笔记87：blkid命令——获取块设备信息
  • 第7章 利用CSS和多媒体美化页面作业
  • Tree of Thoughts: Deliberate Problem Solving with Large Language Models
  • 正点原子阿尔法ARM开发板-IMX6ULL（十一）——IIC协议和SPI协议--AP3216C环境光传感器和ICM20608六轴传感器
  • RK3568平台开发系列讲解（I2C篇）通过I2C总线访问客户端方法
  • go sdk的安装或者升级