使用opencv中基于高斯混合模型（GMM）的EM算法进行数据点分类demo

环境：

Win10+VS2015+opencv3.4.x

opencv生成随机数据点，使用基于高斯混合模型（GMM）的EM算法进行数据点分类

源码：

void GMM_EM( ) {
	Mat img(600, 600, CV_8UC3);//图像
	RNG rng(12345);//随机数生成器，初始化可以传入一个64位的整型参数作为随机数产生器的初值
	 //颜色索引表，根据分类数量设定数组大小
	Scalar colorTab[5] = {
		Scalar(255,0,255),
		Scalar(0,0,255),
		Scalar(0,255,0),
		Scalar(255,0,0),
		Scalar(255,255,0),

	};
	int numCluster = rng.uniform(2, 5);  //随机聚类数，2至5类
	int sampleCount = rng.uniform(50, 1000);    //随机样本数量50-1000个
	Mat points(sampleCount, 2, CV_32FC1);  //样本矩阵，sampleCount行，2列，通道1
	 //生成随机点
	for (int k = 0; k < numCluster; k++)
	{
		Point center;
		center.x = rng.uniform(0, img.cols);
		center.y = rng.uniform(0, img.rows);
		Mat pointChunk = points.rowRange(k*sampleCount / numCluster, (k == numCluster - 1 ? sampleCount : (k + 1)*sampleCount / numCluster));  //为指定的行跨度创建一个矩阵头
		rng.fill(pointChunk, RNG::NORMAL, Scalar(center.x, center.y), Scalar(img.cols*0.05, img.rows*0.05));  //随机数填充矩阵
	 
	}
	//将原数组（矩阵）打乱
	randShuffle(points, 1, &rng);
	Mat lables;  //标签矩阵
 
	Ptr em_model = EM::create();
	em_model->setClustersNumber(numCluster);
	em_model->setCovarianceMatrixType(EM::COV_MAT_SPHERICAL);  // 协方差矩阵
	em_model->setTermCriteria(TermCriteria(TermCriteria::EPS + TermCriteria::COUNT, 100, 0.1));//停止条件
	em_model->trainEM(points, noArray(), lables, noArray());
	//分类像素显示
	Mat sample(1, 2, CV_32FC1);
	for (int i=0;i(0) = (float)i;
			sample.at(1) = (float)j;
			int res = cvRound(em_model->predict2(sample, noArray())[1]);
			Scalar color = colorTab[res];
			circle(img, Point(i, j), 1, color*0.75, -1);
		}
	}
	//原样本点显示
	for (int i = 0; i < sampleCount; i++) {
		circle(img, Point((int)points.at(i,0) , (int) points.at(i,1)), 1, Scalar(255,255,255), -1);
	}
	namedWindow("GMM-EM demo", CV_WINDOW_AUTOSIZE);
	imshow("GMM-EM demo", img);
	waitKey(0);
}

分类结果：

 

对比另一篇博文基于K-Means进行数据点分类：https://blog.csdn.net/ganwenbo2011/article/details/90574376