Java OpenCV 人工智能01.3 机器学习 深度学习


OpenCV 包含的ML模块

序号 模块 中文 英文
1 ANN_MLP 人工神经网络–多层感知器 Artificial Neural Networks Multi-Layer Perception
2 DTree 决策树 Decision Tree
3 EM EM算法 Expectation Maximization
4 KNearest K-邻近算法 K-Nearest Neighbour Classifier
5 LogisticRegression 逻辑回归 logistic regression
6 NormalBayesClassifier 贝叶斯分类 Normal Bayes Classifier
7 RTrees 随机森林 Random Forest
8 SVM 支持向量机 Support Vector Machines
9 SVMSGD 随机梯度下降SVM分类器 Stochastic Gradient Descent SVM Classifier

1 ANN_MLP 人工神经网络-多层感知器

MLP 多层感知器

package com.xu.opencv;

import java.io.BufferedReader;
import java.io.FileReader;
import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;

import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.core.TermCriteria;
import org.opencv.ml.ANN_MLP;
import org.opencv.ml.Boost;
import org.opencv.ml.DTrees;
import org.opencv.ml.EM;
import org.opencv.ml.KNearest;
import org.opencv.ml.LogisticRegression;
import org.opencv.ml.Ml;
import org.opencv.ml.NormalBayesClassifier;
import org.opencv.ml.RTrees;
import org.opencv.ml.SVM;
import org.opencv.ml.SVMSGD;
import org.opencv.ml.TrainData;

/**
 * @author Administrator
 */
public class ML {

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

    public static void main(String[] args) throws Exception {
        ANN_MLP();
    }

    /**
     * Artificial Neural Networks -- Multi-Layer Perception 人工神经网络--多层感知器
     *
     * @throws Exception
     */
    public static void ANN_MLP() throws Exception {
        ANN_MLP ann = ANN_MLP.create();
        /**
         * 输入层：对应着每个图片的像素，所以是28*28
         * 隐含层：两个，神经元个数分别为 512 和 256
         * 输出层：和训练的标签对应，神经元为10个，即手写数字 0123456789
         */
        int[] layer = {28 * 28, 512, 256, 10};
        Mat layerSizes = new Mat(1, layer.length, CvType.CV_32FC1);
        for (int i = 0; i < layer.length; i++) {
            layerSizes.put(0, i, layer[i]);
        }
        ann.setLayerSizes(layerSizes);
        ann.setBackpropWeightScale(0.1);
        ann.setBackpropMomentumScale(0.1);
        ann.setTrainMethod(ANN_MLP.BACKPROP);
        ann.setActivationFunction(ANN_MLP.SIGMOID_SYM);
        ann.setTermCriteria(new TermCriteria(TermCriteria.MAX_ITER, 100000, 0.000001));

        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\train.txt", CvType.CV_32FC1, CvType.CV_32FC1);
        TrainData data = TrainData.create(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        ann.train(data);
        ann.save("D:\\OneDrive\\桌面\\ann.xml");
        ann.clear();
    }

    public static Map<String, Mat> read(String path, int train_type, int label_type) throws Exception {
        BufferedReader reader = new BufferedReader(new FileReader(path));
        Mat train = new Mat(600, 28 * 28, train_type);
        Mat label = new Mat(600, 10, label_type);
        Map<String, Mat> map = new HashMap<>(2);
        int index = 0;
        String line = null;
        while ((line = reader.readLine()) != null) {
            int[] data = Arrays.asList(line.split("->")[1].split(",")).stream().mapToInt(Integer::parseInt).toArray();
            for (int i = 0; i < 28 * 28; i++) {
                train.put(index, i, (float) data[i]);
            }
            int col = Integer.parseInt(line.split("->")[0]);
            label.put(index, col, (float) col);
            index++;
            if (index >= 600) {
                break;
            }
        }
        map.put("train", train);
        map.put("label", label);
        reader.close();
        return map;
    }

}

2 SVM 支持向量机

package com.xu.opencv;

import java.io.BufferedReader;
import java.io.FileReader;
import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;

import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.core.TermCriteria;
import org.opencv.ml.ANN_MLP;
import org.opencv.ml.Boost;
import org.opencv.ml.DTrees;
import org.opencv.ml.EM;
import org.opencv.ml.KNearest;
import org.opencv.ml.LogisticRegression;
import org.opencv.ml.Ml;
import org.opencv.ml.NormalBayesClassifier;
import org.opencv.ml.RTrees;
import org.opencv.ml.SVM;
import org.opencv.ml.SVMSGD;
import org.opencv.ml.TrainData;

/**
 * @author Administrator
 */
public class ML {

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

    public static void main(String[] args) throws Exception {
        SVM();
    }

    /**
     * 支持向量机 Support Vector Machines
     *
     * @throws Exception
     */
    public static void SVM() throws Exception {
        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);
        svm.setTermCriteria(new TermCriteria(TermCriteria.EPS + TermCriteria.MAX_ITER, 1000, 0));
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\data.txt");
        svm.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        svm.save("D:\\OneDrive\\桌面\\svm.xml");
    }

    public static Map<String, Mat> read(String path) throws Exception {
        BufferedReader reader = new BufferedReader(new FileReader(path));
        String line = null;
        Mat train = new Mat(60000, 28 * 28, CvType.CV_32FC1);
        Mat label = new Mat(1, 60000, CvType.CV_32SC1);
        Map<String, Mat> map = new HashMap<>(2);
        int index = 0;
        while ((line = reader.readLine()) != null) {
            int[] data = Arrays.asList(line.split("->")[1].split(",")).stream().mapToInt(Integer::parseInt).toArray();
            for (int i = 0; i < 28 * 28; i++) {
                train.put(index, i, data[i]);
            }
            label.put(index, 0, Integer.parseInt(line.split("->")[0]));
            index++;
        }
        map.put("train", train);
        map.put("label", label);
        reader.close();
        return map;
    }

}

3 Java OpenCV 全部测试代码

package com.xu.opencv;

import java.io.BufferedReader;
import java.io.FileReader;
import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;

import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.core.TermCriteria;
import org.opencv.ml.ANN_MLP;
import org.opencv.ml.Boost;
import org.opencv.ml.DTrees;
import org.opencv.ml.EM;
import org.opencv.ml.KNearest;
import org.opencv.ml.LogisticRegression;
import org.opencv.ml.Ml;
import org.opencv.ml.NormalBayesClassifier;
import org.opencv.ml.RTrees;
import org.opencv.ml.SVM;
import org.opencv.ml.SVMSGD;
import org.opencv.ml.TrainData;

/**
 * @author Administrator
 */
public class ML {

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

    public static void main(String[] args) throws Exception {

        ANN_MLP();
    }

    /**
     * ANN_MLP 人工神经网络-多层感知器
     *
     * @throws Exception
     */
    public static void ANN_MLP() throws Exception {
        ANN_MLP ann = ANN_MLP.create();
        /**
         * 输入层：对应着每个图片的像素，所以是28*28
         * 隐含层：两个，神经元个数分别为 512 和 256
         * 输出层：和训练的标签对应，神经元为10个，即手写数字 0123456789
         */
        int[] layer = {28 * 28, 512, 256, 10};
        Mat layerSizes = new Mat(1, layer.length, CvType.CV_32FC1);
        for (int i = 0; i < layer.length; i++) {
            layerSizes.put(0, i, layer[i]);
        }
        ann.setLayerSizes(layerSizes);
        ann.setBackpropWeightScale(0.1);
        ann.setBackpropMomentumScale(0.1);
        ann.setTrainMethod(ANN_MLP.BACKPROP);
        ann.setActivationFunction(ANN_MLP.SIGMOID_SYM);
        ann.setTermCriteria(new TermCriteria(TermCriteria.MAX_ITER, 100000, 0.000001));

        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\train.txt", CvType.CV_32FC1, CvType.CV_32FC1);
        TrainData data = TrainData.create(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        ann.train(data);
        ann.save("D:\\OneDrive\\桌面\\ann.xml");

        Mat response = new Mat();
        ann.predict(map.get("train"), response);
        for (int i = 0; i < response.height(); i++) {
            System.out.println(response.get(i, 0)[0]);
        }

        ann.clear();
    }

    public static Map<String, Mat> read(String path, int train_type, int label_type) throws Exception {
        BufferedReader reader = new BufferedReader(new FileReader(path));
        Mat train = new Mat(600, 28 * 28, train_type);
        Mat label = new Mat(600, 10, label_type);
        Map<String, Mat> map = new HashMap<>(2);
        int index = 0;
        String line = null;
        while ((line = reader.readLine()) != null) {
            int[] data = Arrays.asList(line.split("->")[1].split(",")).stream().mapToInt(Integer::parseInt).toArray();
            for (int i = 0; i < 28 * 28; i++) {
                train.put(index, i, (float) data[i]);
            }
            int col = Integer.parseInt(line.split("->")[0]);
            label.put(index, col, (float) col);
            index++;
            if (index >= 600) {
                break;
            }
        }
        map.put("train", train);
        map.put("label", label);
        reader.close();
        return map;
    }

    /**
     * 决策树 Decision Tree
     *
     * @throws Exception
     */
    public static void DTrees() throws Exception {
        DTrees tree = DTrees.create();
        tree.setMaxCategories(10); // 设置分类数为 0-9
        tree.setMaxDepth(10);
        tree.setMinSampleCount(10);
        tree.setCVFolds(0);
        tree.setUseSurrogates(false);
        tree.setUse1SERule(false);
        tree.setTruncatePrunedTree(false);
        tree.setRegressionAccuracy(0);
        tree.setPriors(new Mat());
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\train.txt");
        tree.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        tree.save("D:\\OneDrive\\桌面\\DTrees.xml");
    }

    /**
     * EM算法 Expectation - Maximization
     *
     * @throws Exception
     */
    public static void EM() throws Exception {
        EM em = EM.create();
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\data.txt");
        em.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        em.save("D:\\OneDrive\\桌面\\em.xml");
    }


    /**
     * K-邻近算法 K-Nearest Neighbour Classifier
     *
     * @throws Exception
     */
    public static void KNN() throws Exception {
        KNearest knn = KNearest.create();
        knn.setDefaultK(5);
        knn.setIsClassifier(true);
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\train.txt");
        TrainData data = TrainData.create(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        knn.train(data);
        knn.save("D:\\OneDrive\\桌面\\knn.xml");
    }

    /**
     * 逻辑回归 logistic regression
     *
     * @throws Exception
     */
    public static void LogisticRegression() throws Exception {
        LogisticRegression logic = LogisticRegression.create();
        logic.setLearningRate(0.00001);
        logic.setIterations(100);
        logic.setRegularization(LogisticRegression.REG_DISABLE);
        logic.setTrainMethod(LogisticRegression.MINI_BATCH);
        logic.setMiniBatchSize(1);
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\train.txt");
        logic.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        logic.save("D:\\OneDrive\\桌面\\logic.xml");
    }

    /**
     * 贝叶斯分类 Normal Bayes Classifier
     *
     * @throws Exception
     */
    public static void NormalBayesClassifier() throws Exception {
        NormalBayesClassifier nbc = NormalBayesClassifier.create();
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\data.txt");
        nbc.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        nbc.save("D:\\OneDrive\\桌面\\nbc.xml");
    }

    /**
     * 随机森林 Random Forest
     *
     * @throws Exception
     */
    public static void RTrees() throws Exception {
        RTrees tree = RTrees.create();
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\data.txt");
        tree.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        tree.save("D:\\OneDrive\\桌面\\RTrees.xml");
    }

    /**
     * 支持向量机 Support Vector Machines
     *
     * @throws Exception
     */
    public static void SVM() throws Exception {
        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);
        svm.setTermCriteria(new TermCriteria(TermCriteria.EPS + TermCriteria.MAX_ITER, 1000, 0));
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\data.txt");
        svm.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        svm.save("D:\\OneDrive\\桌面\\svm.xml");
    }

    /**
     * SVMSGD 随机梯度下降SVM分类器
     *
     * @throws Exception
     */
    public static void SVMSGD() throws Exception {
        SVMSGD svmsgd = SVMSGD.create();
        Map<String, Mat> map = read("D:\\OneDrive\\桌面\\data.txt");
        svmsgd.train(map.get("train"), Ml.ROW_SAMPLE, map.get("label"));
        svmsgd.save("D:\\OneDrive\\桌面\\svmsgd.xml");
    }

    public static Map<String, Mat> read(String path) throws Exception {
        BufferedReader reader = new BufferedReader(new FileReader(path));
        String line = null;
        Mat train = new Mat(60000, 28 * 28, CvType.CV_32FC1);
        Mat label = new Mat(1, 60000, CvType.CV_32SC1);
        Map<String, Mat> map = new HashMap<>(2);
        int index = 0;
        while ((line = reader.readLine()) != null) {
            int[] data = Arrays.asList(line.split("->")[1].split(",")).stream().mapToInt(Integer::parseInt).toArray();
            for (int i = 0; i < 28 * 28; i++) {
                train.put(index, i, data[i]);
            }
            label.put(index, 0, Integer.parseInt(line.split("->")[0]));
            index++;
        }
        map.put("train", train);
        map.put("label", label);
        reader.close();
        return map;
    }

}
# opencv # 人工智能 # 深度学习
Logo
脑启社区

脑启社区是一个专注类脑智能领域的开发者社区。欢迎加入社区，共建类脑智能生态。社区为开发者提供了丰富的开源类脑工具软件、类脑算法模型及数据集、类脑知识库、类脑技术培训课程以及类脑应用案例等资源。

更多推荐

cover

经典算力墙

avatar 脑启社区

生物计算中的多模态数据：提示工程架构师的AI提示融合方法

什么是生物多模态数据？它的融合难点在哪里？生物多模态数据是指来自不同生物层级、不同技术手段的异质数据模态类型数据形式示例核心价值分子模态序列/数值/结构基因组（DNA突变）、转录组（mRNA表达）、蛋白质组（3D结构）解释疾病的分子机制细胞/组织模态图像/单细胞矩阵病理切片（组织形态）、单细胞RNA-seq（细胞类型）连接分子与宏观表型临床模态文本/数值病历（年龄、吸烟史）、肿瘤标志物（CA125

avatar 脑启社区
cover

AI 算力竞赛白热化：GPU 之外，哪些硬件技术正在重塑行业格局？

avatar 脑启社区