机器学习：SVM分割数据点(线性不可分多分类)

#include <iostream>
#include<opencv2/opencv.hpp>
#include<ctype.h>

#define	NTRAINING_SAMPLES	100			// Number of training samples per class
#define FRAC_LINEAR_SEP		0.9f	    // Fraction of samples which compose the linear separable part

using namespace cv;
using namespace std;
using namespace cv::ml;

int main(int argc, char* argv[])
{
	int size = 400; // height and widht of image  
	const int s = 1000; // number of data  
	int i, j, sv_num;
	Mat img;
	
	
	srand(time(NULL));     // 设置随机数种子
	RNG rng(12345);
	Point pts[s]; // 定义1000个点  
	float data[s * 2]; // 点的坐标  
	int res[s]; // 点的类别  

	const float* support;

	// 图像区域的初始化  
	img = Mat::zeros(Size(size, size), CV_8UC3);

	// 学习数据的生成  
	for (i = 0; i < s; ++i)
	{
		pts[i].x = rand() % size;
		pts[i].y = rand() % size;

		if (pts[i].y > 50 * cos(pts[i].x * CV_PI / 100) + 200)
		{
			line(img,Point(pts[i].x - 2, pts[i].y - 2), Point(pts[i].x + 2, pts[i].y + 2), Scalar(255, 0, 0));
			line(img, Point(pts[i].x + 2, pts[i].y - 2), Point(pts[i].x - 2, pts[i].y + 2), Scalar(255, 0, 0));
			res[i] = 1;
		}
		else
		{
			if (pts[i].x > 200)
			{
				line(img, Point(pts[i].x - 2, pts[i].y - 2), Point(pts[i].x + 2, pts[i].y + 2), Scalar(0, 255, 0));
				line(img, Point(pts[i].x + 2, pts[i].y - 2), Point(pts[i].x - 2, pts[i].y + 2), Scalar(0, 255, 0));
				res[i] = 2;
			}
			else
			{
				line(img, Point(pts[i].x - 2, pts[i].y - 2), Point(pts[i].x + 2, pts[i].y + 2), Scalar(0, 0, 255));
				line(img, Point(pts[i].x + 2, pts[i].y - 2), Point(pts[i].x - 2, pts[i].y + 2), Scalar(0, 0, 255));
				res[i] = 3;
			}
		}
	}
	// 学习数据的现实  
	namedWindow("SVM训练样本空间及分类", WINDOW_AUTOSIZE);
	imshow("SVM训练样本空间及分类", img);
	//waitKey(0);

	// 学习参数的生成  
	for (i = 0; i < s; ++i)
	{
		data[i * 2] = float(pts[i].x) / size;
		data[i * 2 + 1] = float(pts[i].y) / size;
	}
	//创建svm
	Ptr<SVM> svm = SVM::create();
	Mat data_mat(1000,2,CV_32FC1,data);
	Mat res_mat(s,1,CV_32SC1,res);
	svm->setTermCriteria(TermCriteria(TermCriteria::EPS,1000, FLT_EPSILON));
	svm->setType(SVM::C_SVC);
	svm->setKernel(SVM::RBF);
	svm->setDegree(10.0);//高斯核的参数设置
	svm->setGamma(8.0);
	svm->setCoef0(1.0);
	svm->setC(10.0);
	svm->setNu(0.5);
	svm->setP(0.1);
	Ptr<TrainData> traind = TrainData::create(data_mat,ROW_SAMPLE,res_mat);
	svm->train(traind);

	// 学习结果绘图  
	Vec3b red(0, 0, 255), green(0, 255, 0), blue(255, 0, 0);
	for (i = 0; i < size; i++)
	{
		for (j = 0; j < size; j++)
		{
			
			float ret = 0.0;
			float a[] = { float(j) / size,float(i) / size };
			Mat m(1,2,CV_32FC1,a);
			ret = svm->predict(m);

			switch ((int)ret)
			{
			case 1:
				img.at<Vec3b>(i, j) = red;
				break;
			case 2:
				img.at<Vec3b>(i, j) = green;
				break;
			case 3:
				img.at<Vec3b>(i, j) = blue;
				break;
			}
			
		}
	}

	// 为了显示学习结果，通过对输入图像区域的所有像素（特征向量）进行分类，然后对输入的像素用所属颜色等级的颜色绘图  
	for (i = 0; i < s; ++i)
	{
		Scalar rcolor;
		switch (res[i])
		{
		case 1:
			rcolor = Scalar(100, 0, 0);
			break;
		case 2:
			rcolor = Scalar(0, 100, 0);
			break;
		case 3:
			rcolor = Scalar(0, 0, 100);
			break;
		}
		line(img, Point(pts[i].x - 2, pts[i].y - 2), Point(pts[i].x + 2, pts[i].y + 2), rcolor);
		line(img, Point(pts[i].x + 2, pts[i].y - 2), Point(pts[i].x - 2, pts[i].y + 2), rcolor);
	}

	// 支持向量的绘制  
	
	imshow("SVM分类结果及支持向量", img);
	waitKey(0);
	
	return 0;
}