OpenCV C++ usage of cv::Moments

Question

I try to find centroid using OpenCV C++'s cv::Moments. Whatever arguments I submit to it, All I receive back are zeros. Clearly I do something very simply wrong. Output of code:

23 of 500 elements in unit 3
point values 2.976444 18.248287
matrix size 23
moments 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
moments 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000

And code:

printf("%d of %d elements in unit %d\n",k,number_of_features,i);
cv::Mat x(k, 1, cv::DataType<cv::Point2f>::type);
k=0;
for(int j=0;j <number_of_features;j++) {
    if(i  == labels.at<int>(j)) {
        x.at<cv::Point2f>(k++) = samples.at<cv::Point2f>(i); 
    }
}
printf("point values %f %f\n", x.at<cv::Point2f>(0).x,x.at<cv::Point2f>(0).y);
cv::Size s=x.size();
printf("matrix size %d\n",s.height);
cv::Moments m=cv::moments(x);
printf("moments %f %f %f %f %f %f %f %f\n",m.m00,m.m01,m.m20,m.m11,m.m02,m.m30,m.m21,m.m03);
double h[7];
cv::HuMoments(m,h);
printf("moments %f %f %f %f %f %f %f\n",h[0],h[1],h[2],h[3],h[4],h[5],h[6]);

Strangely I cannot find any identical code sample from Google. All I see are C style approaches.

Answer 1

Using moments to find a centroid is a bit overkill imho. You can use the following algorithm to do it:

sumX = 0; sumY = 0;
size = array_points.size;
if(size > 0){

    foreach(point in array_points){
        sumX += point.x;
        sumY += point.y;
    }

 centroid.x = sumX/size;
 centroid.y = sumY/size;
}

Or with the help of Opencv's boundingRect:

//pseudo-code:

Rect bRect = Imgproc.boundingRect(array_points);

centroid.x = bRect.x + (bRect.width / 2);
centroid.y = bRect.y + (bRect.height / 2);

Answer 2

I would recommend you to visit Official Tutorial on moments. Learn and run that code first.

http://docs.opencv.org/doc/tutorials/imgproc/shapedescriptors/moments/moments.html#moments

Once it is success, try to implement whatever you want.