java OpenCV查找k最近邻居C ++到Java转换

Question

I am trying to find the k nearest neighbors with the Knn classifier in OpenCV. I found this C++ Code:

class atsKNN{
public :
void knn(cv::Mat& trainingData, cv::Mat& trainingClasses, cv::Mat& testData, cv::Mat& testClasses, int K) 
{
    cv::KNearest knn(trainingData, trainingClasses, cv::Mat(), false, K);
    cv::Mat predicted(testClasses.rows, 1, CV_32F);
    for(int i = 0; i < testData.rows; i++) {
            const cv::Mat sample = testData.row(i);
            predicted.at<float>(i,0) = knn.find_nearest(sample, K);
    }

    float percentage = evaluate(predicted, testClasses) * 100;
    cout << "K Nearest Neighbor Evaluated Accuracy = " << percentage << "%" << endl;
    prediction = predicted;
}
void showplot(cv::Mat testData)
{
    plot_binary(testData, prediction, "Predictions Backpropagation");
}
private:
cv::Mat prediction;

};

The comments mention it works really good but i am having problems Converting it to Java. There is no Documentation for Java. I tried using a C++ to Java Converter but the resulting Code does not work.

here is the code it produced:

public class atsKNN
{
public final void knn(cv.Mat trainingData, cv.Mat trainingClasses, cv.Mat testData, cv.Mat testClasses, int K)
{
    cv.KNearest knn = new cv.KNearest(trainingData, trainingClasses, cv.Mat(), false, K);
    cv.Mat predicted = new cv.Mat(testClasses.rows, 1, CV_32F);
    for (int i = 0; i < testData.rows; i++)
    {
            final cv.Mat sample = testData.row(i);
            predicted.<Float>at(i,0) = knn.find_nearest(sample, K);
    }

    float percentage = evaluate(predicted, testClasses) * 100;
    System.out.print("K Nearest Neighbor Evaluated Accuracy = ");
    System.out.print(percentage);
    System.out.print("%");
    System.out.print("\n");
    prediction = predicted;
}
public final void showplot(cv.Mat testData)
{
    plot_binary(testData, prediction, "Predictions Backpropagation");
}
private cv.Mat prediction = new cv.Mat();

}

edit: The line predicted.at(i,0) = knn.find_nearest(sample, K); has most definitely to be wrong. There is now function at in object Mat. Also there is no "evaluate function". Another thing is where does the prediction Mat belong to?In java you can not just put it in the end of the class. Thanks=)

Answer 1

The following code is for finding the digits

here's some code to try:

import org.opencv.core.*;
import org.opencv.imgproc.*;
import org.opencv.imgcodecs.Imgcodecs;
import org.opencv.ml.*;
import org.opencv.utils.*;
import java.util.*;


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

public static void main(String[] args) {
    // samples/data/digits.png, have a look at it.
    Mat digits = Imgcodecs.imread("digits.png", 0);
    // setup train/test data:
    Mat trainData = new Mat(),
        testData = new Mat();
    List<Integer> trainLabs = new ArrayList<Integer>(),
        testLabs = new ArrayList<Integer>();
    // 10 digits a 5 rows:
    for (int r=0; r<50; r++) { 
        // 100 digits per row:
        for (int c=0; c<100; c++) {
            // crop out 1 digit:
            Mat num = digits.submat(new Rect(c*20,r*20,20,20));
            // we need float data for knn:
            num.convertTo(num, CvType.CV_32F);
            // 50/50 train/test split:
            if (c % 2 == 0) {
                // for opencv ml, each feature has to be a single row:
                trainData.push_back(num.reshape(1,1));
                // add a label for that feature (the digit number):
                trainLabs.add(r/5);
            } else {
                testData.push_back(num.reshape(1,1));
                testLabs.add(r/5);
            }
        }                
    }

    // make a Mat of the train labels, and train knn:
    KNearest knn = KNearest.create();
    knn.train(trainData, Ml.ROW_SAMPLE, Converters.vector_int_to_Mat(trainLabs));
    // now test predictions:
    for (int i=0; i<testData.rows(); i++)
    {
        Mat one_feature = testData.row(i);
        int testLabel = testLabs.get(i);

        Mat res = new Mat();
        float p = knn.findNearest(one_feature, 1, res);
        System.out.println(testLabel + " " + p + " " + res.dump());
    }

    //// hmm, the 'real world' test case probably looks more like this:
    //// make sure, you follow the very same preprocessing steps used in the train phase:
    //  Mat one_feature = Imgcodecs.imread("one_digit.png", 0);
    //  Mat feature; one_feature.convertTo(feature, CvTypes.CV_32F);
    //  Imgproc.resize(feature, feature, Size(20,20));
    //  int predicted = knn.findNearest(feature.reshape(1,1), 1);
}
}