机器学习与人脸识别8：opencv3 Eigenfaces, Fisherfaces,LBP 人脸识别C++代码实现_eigenfacerecognizer c++

OpenCV中有一个FaceRecognizer 类用于人脸识别；

当前实现的算法有

• Eigenfaces (see createEigenFaceRecognizer)
• Fisherfaces (see createFisherFaceRecognizer)
• Local Binary Patterns Histograms (see createLBPHFaceRecognizer)

Eigenfaces：

准备数据

准备几张用于识别的图像，并将图片信息写入csv文件；先手动写入吧 ，稍后用Python写入；

我准备了4张人脸

csv文件中分号前是图片相对路径，分好后是标签；就是说同一个人标签要相同；对应上面src_2.jpg 和 src_5.jpg是同一个人

代码实现：
#include "opencv2/core.hpp"
#include "opencv2/face.hpp"
#include "opencv2/highgui.hpp"
#include "opencv2/imgproc.hpp"
#include <iostream>
#include <fstream>
#include <sstream>
using namespace cv;
using namespace cv::face;
using namespace std;
static Mat norm_0_255(InputArray _src) {
    Mat src = _src.getMat();
    // Create and return normalized image:
    Mat dst;
    switch(src.channels()) {
    case 1:
        cv::normalize(_src, dst, 0, 255, NORM_MINMAX, CV_8UC1);
        break;
    case 3:
        cv::normalize(_src, dst, 0, 255, NORM_MINMAX, CV_8UC3);
        break;
    default:
        src.copyTo(dst);
        break;
    }
    return dst;
}
static void read_csv(const string& filename, vector<Mat>& images, vector<int>& labels, char separator = ';') {
    std::ifstream file(filename.c_str(), ifstream::in);
    if (!file) {
        string error_message = "No valid input file was given, please check the given filename.";
        CV_Error(Error::StsBadArg, error_message);
    }
    string line, path, classlabel;
    while (getline(file, line)) {
        stringstream liness(line);
        getline(liness, path, separator);
        getline(liness, classlabel);
        if(!path.empty() && !classlabel.empty()) {
            images.push_back(imread(path, 0));
            labels.push_back(atoi(classlabel.c_str()));
        }
    }
}
int main(int argc, const char *argv[])
{
    // 检查有效的命令行参数，打印用法
    // 如果没有给出任何参数
    if (argc < 2) {
        cout << "usage: " << argv[0] << " <csv.ext> <output_folder> " << endl;
        exit(1);
    }
    string output_folder = ".";
    if (argc == 3) {
        output_folder = string(argv[2]);
    }

    string fn_csv = string(argv[1]);        // 获取CSV的路径.
    vector<Mat> images;     // 这些矢量包含图像和相应的标签.
    vector<int> labels;

    try {
        read_csv(fn_csv, images, labels);
    } catch (cv::Exception& e) {
        cerr << "Error opening file \"" << fn_csv << "\". Reason: " << e.msg << endl;
        exit(1);
    }

    if(images.size() <= 1) {
        string error_message = "This demo needs at least 2 images to work. Please add more images to your data set!";
        CV_Error(Error::StsError, error_message);
    }

    // 获取第一张图片宽和高，后面都设置成相同参数
    int height = images[0].rows;

    Mat testSample = images[images.size() - 1];
    int testLabel = labels[labels.size() - 1];
    images.pop_back();
    labels.pop_back();
    // The following lines create an Eigenfaces model for
    // face recognition and train it with the images and
    // labels read from the given CSV file.
    // This here is a full PCA, if you just want to keep
    // 10 principal components (read Eigenfaces), then call
    // the factory method like this:
    //
    //      cv::createEigenFaceRecognizer(10);
    //
    // If you want to create a FaceRecognizer with a
    // confidence threshold (e.g. 123.0), call it with:
    //
    //      cv::createEigenFaceRecognizer(10, 123.0);
    //
    // If you want to use _all_ Eigenfaces and have a threshold,
    // then call the method like this:
    //
    //      cv::createEigenFaceRecognizer(0, 123.0);
    //
    Ptr<BasicFaceRecognizer> model = createEigenFaceRecognizer();
    model->train(images, labels);
    // The following line predicts the label of a given
    // test image:
    cv::imshow("test", testSample);
    cv::waitKey();

    int predictedLabel = model->predict(testSample);
    //
    // To get the confidence of a prediction call the model with:
    //
    //      int predictedLabel = -1;
    //      double confidence = 0.0;
    //      model->predict(testSample, predictedLabel, confidence);
    //
    string result_message = format("Predicted class = %d / Actual class = %d.",
                                   predictedLabel, testLabel);
    cout << result_message << endl;
    // Here is how to get the eigenvalues of this Eigenfaces model:
    Mat eigenvalues = model->getEigenValues();
    // And we can do the same to display the Eigenvectors (read Eigenfaces):
    Mat W = model->getEigenVectors();
    // Get the sample mean from the training data
    Mat mean = model->getMean();
    // Display