学习KNN（三）KNN+HOG实现手写数字识别

#include <iostream>
#include <opencv2/opencv.hpp>
#include <opencv2/ml/ml.hpp>  
using namespace std;
using namespace cv;

char ad[128]={0};

int main()
{
    Mat samFeatureMat ,samLabelMat;
    int k=5,testnum=0,truenum=0;
    HOGDescriptor hog(Size(20,20),Size(10,10),Size(2,2),Size(2,2),9);//利用构造函数，给对象赋值。
    int DescriptorDim;//HOG描述子的维数
    //读取训练数据 4000张
    for (int i = 0; i < 10; i++)
    {
        for (int j =0;j<400;j++)
        {
            sprintf_s(ad, "D:\\data\\%d\\%d.jpg",i,j);
            Mat srcimage = imread(ad);
            vector<float> descriptors;//HOG描述子向量
            hog.compute(srcimage,descriptors);

            if ( i == 0&&j==0)
            {
                DescriptorDim = descriptors.size();
                samFeatureMat = Mat::zeros(4000  , DescriptorDim, CV_32FC1);
                samLabelMat = Mat::zeros(4000  , 1, CV_32FC1);
            }

            for(int k=0; k<DescriptorDim; k++)
            {
                samFeatureMat.at<float>(i*400+j,k) = descriptors[k];
                samLabelMat.at<float>(i*400+j,0) = i;
            }
        }
    }
    samFeatureMat.convertTo(samFeatureMat,CV_32F);
    CvKNearest knn( samFeatureMat, samLabelMat, cv::Mat(), false, k );  
    cv::Mat nearests( 1, k, CV_32F);  
    //读取测试数据  1000张
    for (int i = 0; i < 10; i++)
    {
        for (int j =400;j<500;j++)
        {
            testnum++;
            sprintf_s(ad, "D:\\data\\%d\\%d.jpg",i,j);
            Mat testFeatureMat =Mat::zeros(1,DescriptorDim, CV_32FC1);
            Mat testdata = imread(ad);
            vector<float> descriptors;//HOG描述子向量
            hog.compute(testdata,descriptors);
            for(int k=0; k<DescriptorDim; k++)
            {
                testFeatureMat.at<float>(0,k) = descriptors[k];
            }
            testFeatureMat.convertTo(testFeatureMat,CV_32F);
            int  response = knn.find_nearest(testFeatureMat,k,0,0,&nearests,0); 
            if (response==i)
            {
                truenum++;
            }
        }
    }
    cout<<"测试总数"<<testnum<<endl;
    cout<<"正确分类数"<<truenum<<endl;
    cout<<"准确率："<<(float)truenum/testnum*100<<"%"<<endl;
    return 0;
}

    一个检测窗口中可以滑出36个块，一个块中可以划分25个cell，一个cell中产生9个方向，那么36\*25\*9=8100。