摘要：该算法包含表格边界框检测、表格分割和表格方向识别三个部分，首先，ppyoloe-plus-x 对边界框进行预测，并对置信度较高的表格边界框（box）进行裁剪。裁剪后的单个表格实例会送入到DBNet中进行语义分割，分割结果通过OpenCv轮廓处理获得表格关键点(
百度网盘AI大赛-表格检测的第2名方案。
该算法包含表格边界框检测、表格分割和表格方向识别三个部分，首先，ppyoloe-plus-x 对边界框进行预测，并对置信度较高的表格边界框（box）进行裁剪。裁剪后的单个表格实例会送入到DBNet中进行语义分割，分割结果通过OpenCv轮廓处理获得表格关键点(point)。之后，我们根据DBNet计算的关键点在裁剪后的单个表格实例上绘制表格边界。最后，PP-LCNet结合表格边界先验和表格实例图像，对表格的方向进行预测，并根据之前定义的几何轮廓点与语义轮廓点的对应关系，将几何轮廓点映射为语义轮廓点。
Model Properties
Inputs
name：input
tensor：Float[-1, 3, 624, 624]
Outputs
name：linear_1.tmp_1
tensor：Float[-1, 4]
yolo_edge_det.onnx
Model Properties
date：2024-10-28T08:16:43.725877
description：Ultralytics YOLO11l-seg model trained on coco-seg.yaml
author：Ultralytics
version：8.3.23
task：segment
license：AGPL-3.0 License (https://ultralytics.com/license)
docs：https://docs.ultralytics.com
stride：32
batch：1
imgsz：[800, 800]
names：{0: 'table'}
Inputs
name：images
tensor：Float[1, 3, 800, 800]
Outputs
name：output0
tensor：Float[1, 37, 13125]
name：output1
tensor：Float[1, 32, 200, 200]
yolo_obj_det.onnx
Model Properties
date：2024-10-28T13:52:42.181333
description：Ultralytics YOLO11l model trained on coco.yaml
author：Ultralytics
version：8.3.23
task：detect
stride：32
batch：1
imgsz：[928, 928]
names：{0: 'table'}
Inputs
name：images
tensor：Float[1, 3, 928, 928]
Outputs
name：output0
tensor：Float[1, 5, 17661]
C++封装DLL，C#调用
using System;
using System.Diagnostics;
using System.Drawing;
using System.Drawing.Imaging;
using System.Runtime.InteropServices;
using System.Text;
using System.Windows.Forms;
namespace TableDetection
{
public partial class Form1 : Form
{
publicForm1
{
InitializeComponent;
}
Stopwatch stopwatch = new Stopwatch;
Mat image;
string image_path;
string startupPath;
string fileFilter = "*.*|*.bmp;*.jpg;*.jpeg;*.tiff;*.tiff;*.png";
const string DllName = "TableDetectionSharp.dll";
IntPtr engine;
/*
//初始化
extern "C" _declspec(dllexport) int __cdecl init(void** engine, char* obj_model_path, char* edge_model_path, char* cls_model_path, char* msg);
//tableDet
extern "C" _declspec(dllexport) int __cdecl tableDet(void* engine, Mat* image, char* msg, int* out_imgs_size, Mat* out_img1, Mat* out_img2, Mat* out_img3);
//释放
extern "C" _declspec(dllexport) void __cdecl destroy(void* engine);
*/
[DllImport(DllName, EntryPoint = "init", CallingConvention = CallingConvention.Cdecl)]
internal extern static int init(ref IntPtr engine, string obj_model_path, string edge_model_path, string cls_model_path, StringBuilder msg);
[DllImport(DllName, EntryPoint = "tableDet", CallingConvention = CallingConvention.Cdecl)]
internal extern static int tableDet(IntPtr engine, IntPtr image, StringBuilder msg, ref int out_imgs_size, IntPtr out_img1, IntPtr out_img2, IntPtr out_img3);
[DllImport(DllName, EntryPoint = "destroy", CallingConvention = CallingConvention.Cdecl)]
internal extern static int destroy(IntPtr engine);
private void Form1_Load(object sender, EventArgs e)
{
startupPath = Application.StartupPath;
string obj_model_path = startupPath + "\\model\\yolo_obj_det.onnx";
string edge_model_path = startupPath + "\\model\\yolo_edge_det.onnx";
string cls_model_path = startupPath + "\\model\\paddle_cls.onnx";
StringBuilder msg = new StringBuilder(512);
int res = init(ref engine, obj_model_path, edge_model_path, cls_model_path, msg);
if(res == -1)
{
MessageBox.Show(msg.ToString);
return;
}
else
{
Console.WriteLine(msg.ToString);
}
image_path = startupPath + "\\test_img\\real5.jpg";
pictureBox1.Image = new Bitmap(image_path);
image = new Mat(image_path);
}
private void button1_Click(object sender, EventArgs e)
{
OpenFileDialog ofd = new OpenFileDialog;
ofd.Filter = fileFilter;
if(ofd.ShowDialog != DialogResult.OK)return;
pictureBox1.Image = ;
pictureBox2.Image = ;
textBox1.Text = "";
image_path = ofd.FileName;
}
Mat out_img1;
Mat out_img2;
Mat out_img3;
private void button2_Click(object sender, EventArgs e)
{
if(image_path == "")
{
return;
}
button2.Enabled =false;
Application.DoEvents;
Cv2.DestroyAllWindows;
if(image != ) image.Dispose;
if(out_img1 != ) out_img1.Dispose;
if(out_img2 != ) out_img2.Dispose;
if(out_img3 != ) out_img3.Dispose;
if(pictureBox1.Image != ) pictureBox1.Image.Dispose;
int out_imgs_size = 0;
out_img1 = new Mat;
out_img2 = new Mat;
out_img3 = new Mat;
stopwatch.Restart;
int res = tableDet(engine, image.CvPtr, msg, ref out_imgs_size, out_img1.CvPtr, out_img2.CvPtr, out_img3.CvPtr);
if(res == 0)
{
stopwatch.Stop;
double costTime = stopwatch.Elapsed.TotalMilliseconds;
if(out_imgs_size >= 1)
{
pictureBox2.Image = new Bitmap(out_img1.ToMemoryStream);
}
if(out_imgs_size >= 2)
{
Cv2.ImShow("2", out_img2);
}
if(out_imgs_size >= 3)
{
Cv2.ImShow("3", out_img3);
}
textBox1.Text = $"耗时:{costTime:F2}ms";
}
else
{
textBox1.Text = "识别失败";
}
button2.Enabled =true;
}
private void Form1_FormClosed(object sender, FormClosedEventArgs e)
{
destroy(engine);
}
private void button3_Click(object sender, EventArgs e)
{
if(pictureBox2.Image == )
{
return;
}
Bitmap output = new Bitmap(pictureBox2.Image);
var sdf = new SaveFileDialog;
sdf.Title = "保存";
sdf.Filter = "Images (*.jpg)|*.jpg|Images (*.png)|*.png|Images (*.bmp)|*.bmp";
if(sdf.ShowDialog == DialogResult.OK)
{
switch (sdf.FilterIndex)
{
case1:
{
output.Save(sdf.FileName, ImageFormat.Jpeg);
break;
}
case2:
{
output.Save(sdf.FileName, ImageFormat.Png);
break;
}
case3:
{
output.Save(sdf.FileName, ImageFormat.Bmp);
break;
}
}
MessageBox.Show("保存成功，位置：" + sdf.FileName);
}
}
}
}
来源：opendotnet
免责声明：本站系转载，并不代表本网赞同其观点和对其真实性负责。如涉及作品内容、版权和其它问题，请在30日内与本站联系，我们将在第一时间删除内容!
