【问题标题】:OpenCV SVM Training DataOpenCV SVM 训练数据
【发布时间】:2015-09-26 00:59:24
【问题描述】:

我想通过在 C++ 和 Visual Studio 2013 中使用 opencv 3.00 库来学习 svm 实现。 我的代码:

#include<stdio.h>
#include<math.h>
#include<opencv\cv.h>
#include<opencv\highgui.h>
#include<opencv2\objdetect\objdetect.hpp>
#include<opencv2\highgui\highgui.hpp>
#include<opencv2\imgproc\imgproc.hpp>
#include<vector>
#include <windows.h>
#include <atlstr.h>
#include <iostream>
#include <sstream>
#include <iomanip>
#include <opencv2\imgproc\imgproc.hpp>
#include <opencv2\core\core.hpp>
#include <opencv2\highgui\highgui.hpp>
#include <opencv\cvaux.hpp>

using namespace cv;
using namespace std;

#include <opencv2\ml.hpp>

using namespace cv;

int main()
{
    // Data for visual representation
    int width = 512, height = 512;
    Mat image = Mat::zeros(height, width, CV_8UC3);

    // Set up training data
    float labels[4] = { 1.0, -1.0, -1.0, -1.0 };
    Mat labelsMat(4, 1, CV_32FC1, labels);

    float trainingData[4][2] = { { 501, 10 }, { 255, 10 }, { 501, 255 }, { 10, 501 } };
    Mat trainingDataMat(4, 2, CV_32FC1, trainingData);

    // Set up SVM's parameters

    Ptr<ml::SVM> svm = ml::SVM::create();
    // edit: the params struct got removed,
    // we use setter/getter now:
    svm->setType(ml::SVM::C_SVC);
    svm->setKernel(ml::SVM::LINEAR);
    svm->setGamma(3);

    svm->train(trainingDataMat, ml::ROW_SAMPLE, labelsMat);

    Mat res;   // output


    Vec3b green(0, 255, 0), blue(255, 0, 0);
    // Show the decision regions given by the SVM
    for (int i = 0; i < image.rows; ++i)
        for (int j = 0; j < image.cols; ++j)
        {
        Mat sampleMat = (Mat_<float>(1, 2) << j, i);
        float response = svm->predict(sampleMat, res);

        if (response == 1)
            image.at<Vec3b>(i, j) = green;
        else if (response == -1)
            image.at<Vec3b>(i, j) = blue;
        }

    // Show the training data
    int thickness = -1;
    int lineType = 8;
    circle(image, Point(501, 10), 5, Scalar(0, 0, 0), thickness, lineType);
    circle(image, Point(255, 10), 5, Scalar(255, 255, 255), thickness, lineType);
    circle(image, Point(501, 255), 5, Scalar(255, 255, 255), thickness, lineType);
    circle(image, Point(10, 501), 5, Scalar(255, 255, 255), thickness, lineType);

    // Show support vectors
    thickness = 2;
    lineType = 8;
    Mat sv = svm->getSupportVectors();

    for (int i = 0; i < sv.rows; ++i)
    {
        const float* v = sv.ptr<float>(i);
        circle(image, Point((int)v[0], (int)v[1]), 6, Scalar(128, 128, 128), thickness, lineType);
    }

    imwrite("result.png", image);        // save the image

    imshow("SVM Simple Example", image); // show it to the user
    waitKey(0);

} 

运行这段代码后,我得到了那个错误:

OpenCV Error: Bad argument < in the case of classification problem the responses must be categorical; 
either specify varType when creating TrainData, or pass integer responses > in cv::ml::SVMImpl::train, 
file C:\builds\master_PackSlave-win64-vc12-shared\opencv\modules\ml\src\svm.cpp, line 1610

我调试了那个代码。调试器停在这一行:svm-&gt;train(trainingDataMat, ml::ROW_SAMPLE, labelsMat);

上面写着:

 First-chance exception at 0x000007FEFDA5AAAD in train.exe: Microsoft C++ exception: cv::Exception at memory location 0x00000000001CEE50.
    Unhandled exception at 0x000007FEFDA5AAAD in train.exe: Microsoft C++ exception: cv::Exception at memory location 0x00000000001CEE50.

另外,它说:

(Win32): Loaded 'C:\OpenCV3.0.0\opencv\build\x64\vc12\bin\opencv_world300d.dll'. Cannot find or open the PDB file.

其实我的理解是这个问题和内存有关。

【问题讨论】:

    标签: c++ opencv visual-studio-2013 opencv3.0


    【解决方案1】:

    responses 的类型不能是floatdouble

    改变

    float labels[4] = { 1.0, -1.0, -1.0, -1.0 };
    Mat labelsMat(4, 1, CV_32FC1, labels);
    

    int labels[4] = { 1, -1, -1, -1 };
    Mat labelsMat(4, 1, CV_32S, labels);
    

    顺便说一句,如果你使用线性内核,唯一的参数是C,所以你不需要setGamma


    另一个问题是获得预测响应的方式。由于每次只有一个样本要预测,所以如果要使用返回值作为响应,请不要将res 传递给predict

    你可以改变

    float response = svm->predict(sampleMat, res);
    

    float response = svm->predict(sampleMat);
    

    否则,如果要使用res,则返回值不再是响应值。但您可以改为从res 获得回复。

    你可以改变

    if (response == 1)
        image.at<Vec3b>(i, j) = green;
    else if (response == -1)
        image.at<Vec3b>(i, j) = blue;
    }
    

    if (res.at<float>(0) == 1)
        image.at<Vec3b>(i, j) = green;
    else if (res.at<float>(0) == -1)
        image.at<Vec3b>(i, j) = blue;
    }
    

    【讨论】:

    • 尽管使用了 CV32S,但我仍然遇到同样的错误:Mat label = Mat::zeros(res.rows, 1, CV_32SC1); res.col(i).copyTo(标签); Ptr<:svm> svm = ml::SVM::create(); svm->setType(ml::SVM::C_SVC); svm->setKernel(ml::SVM::LINEAR); svm->setTermCriteria(TermCriteria(TermCriteria::MAX_ITER, 100, 1e-6)); svm->train(train, ml::ROW_SAMPLE, label);
    • 我忘记提及的一件事是,在我的标签中,我的值从 0 到 8,即有 9 个类别或类
    猜你喜欢
    • 2015-08-14
    • 2015-05-04
    • 2015-07-25
    • 2015-04-18
    • 2015-01-22
    • 2017-08-25
    • 2014-12-15
    • 1970-01-01
    相关资源
    最近更新 更多