#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/ml/ml.hpp>
using namespace cv;
int main()
{
//【1】创建可视化的窗口
int width = 512;
int height = 512;
cv::Mat image = Mat::zeros(height, width, CV_8UC3);
//【2】设置训练数据
float labels[4] = {1.0, -1.0, -1.0, -1.0};
cv::Mat labelsMat(4, 1, CV_32FC1, labels); //[0]构造了一个列向量的矩阵头
float trainingData[4][2] = { {501, 10}, {255, 10}, {501, 255}, {10, 501} };
cv::Mat trainingDataMat(4, 2, CV_32FC1, trainingData); //[1]构造一个执行数组trainingData的矩阵头
//【3】设置支持向量机的参数
CvSVMParams params;
params.svm_type = CvSVM::C_SVC; //[2]SVM的类型
params.kernel_type = CvSVM::LINEAR; //[3]核函数的类型::线性
params.term_crit = cvTermCriteria(CV_TERMCRIT_ITER, 100, 1e-6); //[4]终止准则函数,当迭代到最大值时,终止迭代
//【4】训练SVM
CvSVM SVM; //[5]实例化一个SVM类的实例
SVM.train(trainingDataMat, labelsMat, Mat(), Mat(), params); //[6]参数为:输入数据,响应,xx,xx,SVM的参数
cv::Vec3b green(0,255,0); //[7]Vec3b这是一个3维的列向量,可以存储3个字符型数据
cv::Vec3b blue (255,0,0);
//【5】显示由SVM决定的区域
for (int i = 0; i < image.rows; ++i)
for (int j = 0; j < image.cols; ++j)
{
Mat sampleMat = (Mat_<float>(1,2) << i,j);
float response = SVM.predict(sampleMat); //[8]predict()函数是用来做预测的,输入参数为样本,返回值为1或者-1
if (response == 1)
image.at<Vec3b>(j, i) = green;
else if (response == -1)
image.at<Vec3b>(j, i) = blue;
}
//【6】画出训练数据
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);
//【7】显示支持向量
thickness = 2;
lineType = 8;
int c = SVM.get_support_vector_count();
std::cout<<"支持向量的个数为 = "<<c<<std::endl;
for(int i = 0; i < c; ++i)
{
const float* v = SVM.get_support_vector(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);
}
/*****************************************************************************************************
程序功能:
SVM线性不可分的情况示例程序解读
程序说明:
1)本训练程序中,一共有两类样本,是一个基于SVM的二分类问题;其中每类样本集的数量为100个样本,其
中90个样本是线性可分的,10个是线性不可分的
2)这200个样本的具体数据存储在trainData内;trainData是一个200行2列的矩阵,其中第一列存储的样本是
X值;第二列存储的样本是Y值;每一列的前90个元素是第一类线性可分的部分,后90个元素是第二类
线性可分的部分,中间的20个元素是线性不可分的部分。
开发环境:
VS2012 + OpenGl(GLUT3.7) + OpenCv2.4.9 + Halcon10.0
时间地点:
陕西师范大学----2017.3.2
作 者:
九月
*****************************************************************************************************/
#include <iostream>
#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/ml/ml.hpp>
#define NTRAINING_SAMPLES 100 //[1]每个样本集的数量
#define FRAC_LINEAR_SEP 0.9f //[2]其中线性可分的部分
using namespace cv;
using namespace std;
int main()
{
//【1】创建一个可视化的图像
const int WIDTH = 512;
const int HEIGHT = 512;
cv::Mat I = Mat::zeros(HEIGHT, WIDTH, CV_8UC3); //[3]Matlab风格的矩阵的初始化
Mat trainData(2*NTRAINING_SAMPLES, 2, CV_32FC1); //[4]训练样本数据的存储矩阵
Mat labels (2*NTRAINING_SAMPLES, 1, CV_32FC1); //[5]训练样本类别的存储矩阵
RNG rng(100); //[6]设定随机数的种子
//[7]设定线性可分部分的训练数据量
int nLinearSamples = (int) (FRAC_LINEAR_SEP * NTRAINING_SAMPLES);
//【2】设定第一类中的数据
Mat trainClass = trainData.rowRange(0, nLinearSamples);//[8]从整个数据取出[0,89]行的数据
Mat c = trainClass.colRange(0, 1); //[9]取出第一列
//[10]随机生成X的值:[0,0.4*WIDTH]
rng.fill(c, RNG::UNIFORM, Scalar(1), Scalar(0.4 * WIDTH));
c = trainClass.colRange(1,2); //[11]取出第二列
rng.fill(c, RNG::UNIFORM, Scalar(1), Scalar(HEIGHT)); //[12]随机生成Y的值
//【3】设定第二类中的数据 //[13]从整个数据中取出[110,199]行
trainClass = trainData.rowRange(2*NTRAINING_SAMPLES-nLinearSamples, 2*NTRAINING_SAMPLES);
c = trainClass.colRange(0 , 1); //[14]取出第一列
//[15]随机生成X的值[0.6*WIDTH,WIDTH]
rng.fill(c, RNG::UNIFORM, Scalar(0.6*WIDTH), Scalar(WIDTH));
//[16]取出第二列
c = trainClass.colRange(1,2);
rng.fill(c, RNG::UNIFORM, Scalar(1), Scalar(HEIGHT)); //[17]随机生成Y的值
//【4】线性不可分的数据
trainClass = trainData.rowRange( nLinearSamples, 2*NTRAINING_SAMPLES-nLinearSamples);
c = trainClass.colRange(0,1);
rng.fill(c, RNG::UNIFORM, Scalar(0.4*WIDTH), Scalar(0.6*WIDTH));
c = trainClass.colRange(1,2);
rng.fill(c, RNG::UNIFORM, Scalar(1), Scalar(HEIGHT));
//【5】为所有数据设置标签:
//1)前100个为第一类数据
//2)后100个为第二类数据
labels.rowRange( 0, NTRAINING_SAMPLES).setTo(1); // Class 1
labels.rowRange(NTRAINING_SAMPLES, 2*NTRAINING_SAMPLES).setTo(2); // Class 2
//【6】设置支持向量机SVM的参数
CvSVMParams params;
params.svm_type = SVM::C_SVC;
params.C = 0.1;
params.kernel_type = SVM::LINEAR;
params.term_crit = TermCriteria(CV_TERMCRIT_ITER, (int)1e7, 1e-6);
//【7】训练SVM
std::cout << "Starting training process" << endl;
CvSVM svm;
svm.train(trainData, labels, Mat(), Mat(), params);
std::cout << "Finished training process" << endl;
//【8】显示判决区域
cv::Vec3b green(0,100,0);
cv::Vec3d blue (100,0,0);
for (int i = 0; i < I.rows; ++i)
for (int j = 0; j < I.cols; ++j)
{
Mat sampleMat = (Mat_<float>(1,2) << i, j);
float response = svm.predict(sampleMat); //[1]判决函数
if (response == 1) I.at<Vec3b>(j, i) = green; //[2]第一类
else if (response == 2) I.at<Vec3b>(j, i) = blue; //[3]第二类
}
//【9】显示训练数据
int thick = -1;
int lineType = 8;
float px;
float py;
// Class 1
for (int i = 0; i < NTRAINING_SAMPLES; ++i)
{
px = trainData.at<float>(i,0);
py = trainData.at<float>(i,1);
circle(I, Point( (int) px, (int) py ), 3, Scalar(0, 255, 0), thick, lineType);
}
// Class 2
for (int i = NTRAINING_SAMPLES; i <2*NTRAINING_SAMPLES; ++i)
{
px = trainData.at<float>(i,0);
py = trainData.at<float>(i,1);
circle(I, Point( (int) px, (int) py ), 3, Scalar(255, 0, 0), thick, lineType);
}
//【10】显示支持向量点
thick = 2;
lineType = 8;
int x = svm.get_support_vector_count();
for (int i = 0; i < x; ++i)
{
const float* v = svm.get_support_vector(i);
circle( I, Point( (int) v[0], (int) v[1]), 6, Scalar(128, 128, 128), thick, lineType);
}
imwrite("result.png", I); // save the Image
imshow("SVM for Non-Linear Training Data", I); // show it to the user
waitKey(0);
}