Java OpenCV-4.x 人工智能01 機器學習 支持向量機 SVM

OpenCV 中的機器學習(Machine Learning，ML)算法。

序號	中文名稱	英文名稱
01	決策樹	Decision Tree
02	EM算法	Expectation - Maximization
03	貝葉斯分類	Normal Bayes Classifier
04	K-鄰近算法	K-Nearest Neighbour Classifier
05	支持向量機	Support Vector Machine
06	Boost樹算法	Boosted Tree Classifier
07	隨機森林算法	Random Trees Classifier
08	人工神經網絡	Artificial Neural Networks
09	梯度Boost樹算法	Gradient Boosted Trees
10	絕對隨機森林算法	Extremely Randomized Trees Classifier
11	SGD支持向量機	Stochastic Gradient Descent Support Vector Machine

Support Vector Machine SVM 介紹

package com.xu.opencv;

import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.core.TermCriteria;
import org.opencv.ml.Ml;
import org.opencv.ml.SVM;
import org.opencv.ml.TrainData;

/**  
 * 
 * @Author: hyacinth
 * @Title: ML.java   
 * @Package com.xu.opencv   
 * @Description:
 * @Date: 2019年11月22日20:23:23 
 * @Version V-1.0 
 * @Copyright: 2019 hyacinth
 * 
 */
public class ML {

	static {
		System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
	}

	public static void main(String[] args) {
		// 訓練數據 體重,身高
		float[] train_data = { 186, 80, 185, 81, 165, 60, 168, 61, 160, 50, 161, 48 };
		// 測試數據 身高,體重
		float[] test = { 184, 79, 160, 62, 159, 50 };
		// 訓練數據 0=男 ,1=女
		int[] label = { 0, 0, 0, 0, 1, 1 };     
		
		Mat train_mat = new Mat(6, 2, CvType.CV_32FC1);  
		train_mat.put(0, 0, train_data);  

		Mat train_lable = new Mat(6, 1, CvType.CV_32SC1);  
		train_lable.put(0, 0, label);   

		Mat test_lable = new Mat(3, 2, CvType.CV_32FC1);  
		test_lable.put(0, 0, test);  

		SVM(train_mat,train_lable,test_lable);
	}

	/**
	 * OpenCV-4.1.0 SVM 支持向量機
	 * @Author: hyacinth
	 * @Title: SVM
	 * @param : tarin 訓練數據
	 * @param : lable 訓練標籤
	 * @param : test  測試數據
	 * @Description: TODO       
	 * @return void  
	 * @date: 2019年11月22日20:23:23
	 */
	public static void SVM(Mat tarin,Mat lable,Mat test) {
		SVM svm = SVM.create();
		svm.setC(1);
		svm.setP(0);
		svm.setNu(0);
		svm.setCoef0(0);
		svm.setGamma(1);
		svm.setDegree(0);
		svm.setType(SVM.C_SVC);
		svm.setKernel(SVM.LINEAR);
		TermCriteria criteria=new TermCriteria(TermCriteria.EPS + TermCriteria.MAX_ITER,1000,0);
		svm.setTermCriteria(criteria);

		TrainData trainData = TrainData.create(tarin,Ml.ROW_SAMPLE,lable);
		svm.train(trainData.getSamples(), Ml.ROW_SAMPLE,trainData.getResponses());
		svm.save("C:\\Users\\Administrator\\Desktop\\number.xml");

		Mat response = new Mat();
		svm.predict(test, response, 0);

		System.out.println(response.dump());

		for (int i = 0; i < response.height(); i++) {
			if (response.get(i, 0)[0] == 0)
				System.out.println("男");
			if (response.get(i, 0)[0] == 1)
				System.out.println("女");
		}

	}

}