基于qt和opencv3实现机器学习之：利用正态贝叶斯分类

来源：互联网发布：js 单选按钮改变事件编辑：程序博客网时间：2024/06/08 15:04

和上一篇svm(http://blog.csdn.net/u013289254/article/details/70568790)的比较相似

第一步：在pro文件里面设置路径

INCLUDEPATH += /usr/local/include \                  /usr/local/include/opencv \                  /usr/local/include/opencv2    LIBS += /usr/local/lib/libopencv_highgui.so \          /usr/local/lib/libopencv_core.so    \          /usr/local/lib/libopencv_imgproc.so \          /usr/local/lib/libopencv_imgcodecs.so \          /usr/local/lib/libopencv_ml.so

第二步：建立cpp文件

#include "opencv2/opencv.hpp"#include "opencv2/imgproc.hpp"#include "opencv2/highgui.hpp"#include "opencv2/ml.hpp"using namespace cv;using namespace cv::ml;int main(int, char**){    int width = 512, height = 512;    Mat image = Mat::zeros(height, width, CV_8UC3);  //创建窗口可视化    // 设置训练数据    int labels[10] = { 1, -1, 1, 1,-1,1,-1,1,-1,-1 };    Mat labelsMat(10, 1, CV_32SC1, labels);    float trainingData[10][2] = { { 501, 150 }, { 255, 10 }, { 501, 255 }, { 10, 501 }, { 25, 80 },    { 150, 300 }, { 77, 200 } , { 300, 300 } , { 45, 250 } , { 200, 200 } };    Mat trainingDataMat(10, 2, CV_32FC1, trainingData);    // 创建贝叶斯分类器    Ptr<NormalBayesClassifier> model=NormalBayesClassifier::create();    // 设置训练数据    Ptr<TrainData> tData =TrainData::create(trainingDataMat, ROW_SAMPLE, labelsMat);    //训练分类器    model->train(tData);    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);  //生成测试数据        float response = model->predict(sampleMat);  //进行预测，返回1或-1        if (response == 1)            image.at<Vec3b>(i, j) = green;        else if (response == -1)            image.at<Vec3b>(i, j) = blue;    }    // 显示训练数据    int thickness = -1;    int lineType = 8;    Scalar c1 = Scalar::all(0); //标记为1的显示成黑点    Scalar c2 = Scalar::all(255); //标记成-1的显示成白点    //绘图时，先宽后高，对应先列后行    for (int i = 0; i < labelsMat.rows; i++)    {        const float* v = trainingDataMat.ptr<float>(i); //取出每行的头指针        Point pt = Point((int)v[0], (int)v[1]);        if (labels[i] == 1)            circle(image, pt, 5, c1, thickness, lineType);        else            circle(image, pt, 5, c2, thickness, lineType);    }    imshow("normal Bayessian classifier Simple Example", image); // show it to the user    waitKey(0);}

第三步：运行程序，出现的结果

0 0