在QT平台上使用PCA+SVM实现人脸识别

@使用PCA+SVM实现人脸识别，正样本使用的ORL face database，负样本是自己创建的，使用的QT编程软件
，以下是源码`

 #include "mainwindow.h"
    #include <opencv2/opencv.hpp>
    #include <opencv\cv.h>
    #include <opencv\cxcore.h>
    #include <opencv\highgui.h>
    #include <QApplication>
    #include<QMessageBox>
    using namespace cv;
    using namespace std;
    MainWindow::MainWindow(QWidget *parent)
        : QMainWindow(parent)
    {
    
    }
    MainWindow::~MainWindow()
    {
    }
    int  MainWindow::video_capture()
    {
         vector<Mat> images; //This variable will be loaded with a set of images to perform PCA on.
         char filename[100];
         int imageFaceNum = 400;
         int imageNonFaceNum = 22;
         int num = 0;
         int num_face=0;
         Size size;
         Mat src;
         Mat values(imageFaceNum+imageNonFaceNum, 1, CV_32SC1); //Values are the corresponding values to each of my images.
         for(int i=1; i<=imageFaceNum/10; i++)
         for(int j=1;j<=10;j++)
         {
          sprintf(filename,"D:\\VS C++\\QT\\PCA+SVM\\ORL face database\\s%d\\%d.bmp",i,j);
          values.at<int>(num,0) = 1;
          src=imread(filename,0);
          size=Size(src.cols,src.rows);
    
          equalizeHist(src,src);
          waitKey(30);
          imshow("samples",src);
          images.push_back(src);
          num++;
          num_face++;
    
         }
    
         for(int i=1; i<=imageNonFaceNum; i++)
         {
          Mat dst(size,CV_8UC1);
          sprintf(filename,"D:\\VS C++\\QT\\PCA+SVM\\no face database\\%d.jpg",i);
    
          values.at<int>(num,0) = 0;
          src=imread(filename,0);
         cv:: resize(src,dst,size);
    
         equalizeHist(dst,dst);
          waitKey(30);
          imshow("samples",dst);
          images.push_back(dst);
          num++;
         }
         int nEigens = images.size()-200; //Number of Eigen Vectors.作为基坐标的特征向量个数
         //Load the images into a Matrix
         Mat desc_mat(num_face, images[0].rows * images[0].cols, CV_8UC1);
         Mat desc_mat1(images.size(), images[0].rows * images[0].cols, CV_8UC1);
    
         for (int i=0; i<num_face; i++)
         {
           desc_mat.row(i) = images[i].reshape(1, 1) ;
         }
         for (int i=0; i<images.size(); i++)
         {
           desc_mat1.row(i) = images[i].reshape(1, 1) ;
         }
         Mat average;
         PCA pca(desc_mat, average, CV_PCA_DATA_AS_ROW, nEigens);
         Mat data(images.size(), nEigens, CV_32FC1); //This Mat will contain all the Eigenfaces that will be used later with SVM for detection
         //Project the images onto the PCA subspace
         for(int i=0; i<images.size(); i++)
         {
           Mat projectedMat(1, nEigens, CV_32FC1);
           pca.project(desc_mat1.row(i), projectedMat);
           data.row(i) = projectedMat.row(0);
         }
          cout<<"data"<<data.size()<<endl;
          cout<<"values"<<values.size()<<endl;
         CvMat d1 = (CvMat)data;
         CvMat d2 = (CvMat)values;
         CvSVM svm;
         CvSVMParams param;
         CvTermCriteria criteria;
         criteria = cvTermCriteria( CV_TERMCRIT_EPS, 1000, FLT_EPSILON );
         param = CvSVMParams( CvSVM::C_SVC, CvSVM::RBF, 10.0, 0.09, 1.0, 10.0, 0.5, 1.0, NULL, criteria );
         //☆☆☆☆☆☆☆☆☆(5)SVM学习☆☆☆☆☆☆☆☆☆☆☆☆
         svm.train( &d1, &d2, NULL, NULL, param );
         svm.save("svmdata.xml");


     Mat cs;
     src=imread("D:\\VS C++\\QT\\PCA+SVM\\1.bmp",0);
     equalizeHist(src,src);
     cs = src.reshape(1, 1) ;
     Mat projectedMat(1, nEigens, CV_32FC1);
     pca.project(cs, projectedMat);
     CvMat d3 = (CvMat)projectedMat;
     int ret = svm.predict(&d3);
     cout<<ret<<endl;
      waitKey(0);
      return 0;
    }

`