【图像识别】利用BP神经网络做特征向量分类

现在手头没数据集，用之前的几张图片测试了一下。
data 数组前两个是同一种树叶：

后两个向量是同一种树叶：

没数据集了，用的差不多的一种树叶对比的，就是上面double x[]的特征：

虽然看面积就知道和第一个相似，程序运行结果：

[0.100747, 0.00233442, 0.690329, 1.6875, 0.0483126]:[0.9902157551076237, 0.0030781815284147123, 0.009777284822493251]
[0.104574, 0.0029732, 0.660244, 1.71795, 0.0530419]:[0.9816079000858899, 0.0030103326451449953, 0.018348749704753257]
[0.159876, 0.016334, 0.710315, 2.91176, 0.0400851]:[6.278126573769545E-4, 0.0017362230457877918, 0.9993780509911905]
[0.128983, 0.00860048, 0.72265, 2.39355, 0.0454218]:[0.020148315067172743, 0.0021197880406902316, 0.9798882999520514]
[0.147046, 0.00364964, 0.67903, 1.66344, 0.0540356]:[0.9914561686871933, 0.003028309020178626, 0.008562771909244912]

前四个是训练集，“：”后面是类别，之前是想分八类，所以用了三个输出节点。不影响。
最后一个（第五个）是测试结果，可以看到被分到第一类里了，就是和前两个分一类，占99.14%。
明天去采集数据集，看看效果怎么样。先实现一下4分类吧。。

import java.util.Arrays;
public class BpDeepTest{
    public static void main(String[] args){
        //初始化神经网络的基本配置
        //第一个参数是一个整型数组，表示神经网络的层数和每层节点数，比如{3,10,10,10,10,2}表示输入层是3个节点，输出层是2个节点，中间有4层隐含层，每层10个节点
        //第二个参数是学习步长，第三个参数是动量系数
        BpDeep bp = new BpDeep(new int[]{5,10,3}, 0.12, 0.8);

        //设置样本数据，对应上面的4个二维坐标数据
        double[][] data = new double[][]{{0.100747,0.00233442,0.690329,1.6875,0.0483126},{0.104574,0.0029732,0.660244,1.71795,0.0530419},{0.159876,0.016334,0.710315,2.91176,0.0400851},{0.128983,0.00860048,0.72265,2.39355,0.0454218}};
        //设置目标数据，对应4个坐标数据的分类
        double[][] target = new double[][]{{1,0,0},{1,0,0},{0,0,1},{0,0,1}};

        //迭代训练5000次
        for(int n=0;n<5000;n++)
            for(int i=0;i//根据训练结果来检验样本数据
        for(int j=0;jdouble[] result = bp.computeOut(data[j]);
            System.out.println(Arrays.toString(data[j])+":"+Arrays.toString(result));
        }

        //根据训练结果来预测一条新数据的分类

        //0.147046,0.00364964,0.67903,1.66344,0.0540356
        double[] x = new double[]{0.147046,0.00364964,0.67903,1.66344,0.0540356};
        double[] result = bp.computeOut(x);
        System.out.println(Arrays.toString(x)+":"+Arrays.toString(result));
    }
}

import java.util.Random;
public class BpDeep{
    public double[][] layer;//神经网络各层节点
    public double[][] layerErr;//神经网络各节点误差
    public double[][][] layer_weight;//各层节点权重
    public double[][][] layer_weight_delta;//各层节点权重动量
    public double mobp;//动量系数
    public double rate;//学习系数

    public BpDeep(int[] layernum, double rate, double mobp){
        this.mobp = mobp;
        this.rate = rate;
        layer = new double[layernum.length][];
        layerErr = new double[layernum.length][];
        layer_weight = new double[layernum.length][][];
        layer_weight_delta = new double[layernum.length][][];
        Random random = new Random();
        for(int l=0;lnew double[layernum[l]];
            layerErr[l]=new double[layernum[l]];
            if(l+1new double[layernum[l]+1][layernum[l+1]];
                layer_weight_delta[l]=new double[layernum[l]+1][layernum[l+1]];
                for(int j=0;j1;j++)
                    for(int i=0;i1];i++)
                        layer_weight[l][j][i]=random.nextDouble();//随机初始化权重
            }   
        }
    }
    //逐层向前计算输出
    public double[] computeOut(double[] in){
        for(int l=1;lfor(int j=0;jdouble z=layer_weight[l-1][layer[l-1].length][j];
                for(int i=0;i1].length;i++){
                    layer[l-1][i]=l==1?in[i]:layer[l-1][i];
                    z+=layer_weight[l-1][i][j]*layer[l-1][i];
                }
                layer[l][j]=1/(1+Math.exp(-z));
            }
        }
        return layer[layer.length-1];
    }
    //逐层反向计算误差并修改权重
    public void updateWeight(double[] tar){
        int l=layer.length-1;
        for(int j=0;j1-layer[l][j])*(tar[j]-layer[l][j]);

        while(l-->0){
            for(int j=0;jdouble z = 0.0;
                for(int i=0;i1].length;i++){
                    z=z+l>0?layerErr[l+1][i]*layer_weight[l][j][i]:0;
                    layer_weight_delta[l][j][i]= mobp*layer_weight_delta[l][j][i]+rate*layerErr[l+1][i]*layer[l][j];//隐含层动量调整
                    layer_weight[l][j][i]+=layer_weight_delta[l][j][i];//隐含层权重调整
                    if(j==layerErr[l].length-1){
                        layer_weight_delta[l][j+1][i]= mobp*layer_weight_delta[l][j+1][i]+rate*layerErr[l+1][i];//截距动量调整
                        layer_weight[l][j+1][i]+=layer_weight_delta[l][j+1][i];//截距权重调整
                    }
                }
                layerErr[l][j]=z*layer[l][j]*(1-layer[l][j]);//记录误差
            }
        }
    }

    public void train(double[] in, double[] tar){
        double[] out = computeOut(in);
        updateWeight(tar);
    }
}