神经网络

src/main/java/ML.java

@@ -0,0 +1,78 @@
+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.Ml;
+import org.opencv.ml.TrainData;
+
+import java.net.URL;
+
+public class ML {
+
+    static {
+        URL systemResource = ClassLoader.getSystemResource("lib/x64/opencv_java460.dll");
+        System.load(systemResource.getPath());
+//        System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
+    }
+
+    public static void main(String[] args) {
+        // 训练数据 体重,身高
+        float[] train_data = { 186, 80, 185, 81, 160, 68, 168, 61, 160, 50, 161, 48 };
+        // 训练标签 0f,0f->男,1f,1f->女
+        float[] labels = { 0f, 0f, 0f, 0f, 0f, 0f, 1f, 1f, 1f, 1f };
+        // 测试数据 身高,体重
+        float[] test = { 184, 76, 160, 68, 161, 48 };
+
+        Mat train_mat = new Mat(6, 2, CvType.CV_32FC1);
+        train_mat.put(0, 0, train_data);
+
+        Mat lable_mat = new Mat(6, 2, CvType.CV_32FC1);
+        lable_mat.put(0, 0, labels);
+
+        Mat test_mat = new Mat(3, 2, CvType.CV_32FC1);
+        test_mat.put(0, 0, test);
+
+        ANN(train_mat, lable_mat, test_mat);
+    }
+
+    /**
+     * OpenCV-4.1.0 ANN 人工神经网络
+     * @Author: hyacinth
+     * @Title: ANN
+     * @param : tarin 训练数据
+     * @param : lable 训练标签
+     * @param : test  测试数据
+     * @Description: TODO
+     * @return void
+     * @date: 2019年11月22日20:23:23
+     */
+    public static void ANN(Mat tarin, Mat lable, Mat test) {
+        TrainData td = TrainData.create(tarin, Ml.ROW_SAMPLE, lable);
+        Mat layer = new Mat(1, 4, CvType.CV_32FC1);
+        // 含有两个隐含层的网络结构，输入、输出层各两个节点，每个隐含层含两个节点
+        layer.put(0, 0, new float[] { 2, 2, 2, 2 });
+        ANN_MLP ann = ANN_MLP.create();
+        ann.setLayerSizes(layer);
+        ann.setTrainMethod(ANN_MLP.BACKPROP);
+        ann.setBackpropWeightScale(0.1);
+        ann.setBackpropMomentumScale(0.1);
+        ann.setActivationFunction(ANN_MLP.SIGMOID_SYM, 1, 1);
+        TermCriteria criteria=new TermCriteria(TermCriteria.EPS + TermCriteria.MAX_ITER,300,0);
+        ann.setTermCriteria(criteria);
+        ann.train(td.getSamples(), Ml.ROW_SAMPLE, td.getResponses());
+        ann.save("C:\\Users\\Administrator\\Desktop\\number.xml");
+
+        Mat response = new Mat();
+        ann.predict(test, response, 0);
+        System.out.println(response.dump());
+        for (int i = 0; i < response.size().height; i++) {
+            if (response.get(i, 0)[0] + response.get(i, 1)[0] >= 1) {
+                System.out.println("女");
+            }
+            if (response.get(i, 0)[0] + response.get(i, 1)[0] < 1) {
+                System.out.println("男");
+            }
+        }
+
+    }
+}