Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

Kmeans support needed

Hello,

cv::kmeans(data, 8, labels, cv::TermCriteria(CV_TERMCRIT_ITER, 10, 1.0), 3, cv::KMEANS_PP_CENTERS, &centers);

In kmeans Opencv, I have the following questions :

1)- What is the format of the mat labels? INT * 3 /pixel ? 2)- What is the format of centers ? 3)- For the TermCriteria, Some use CV_TERMCRIT_ITER|CV_TERMCRIT_EPS, so which to use for color reduction, and the two parametere, some enters 10000, 0.0001

Thank you, cjacquel

Kmeans support needed

Hello,

cv::kmeans(data, 8, labels, cv::TermCriteria(CV_TERMCRIT_ITER, 10, 1.0), 3, cv::KMEANS_PP_CENTERS, &centers);

In kmeans Opencv, I have the following questions :

1)- What is the format of the mat labels? INT * 3 /pixel ? 2)- What is the format of centers ? 3)- For the TermCriteria, Some use CV_TERMCRIT_ITER|CV_TERMCRIT_EPS, so which to use for color reduction, and the two parametere, parameters, some enters 10000, 0.0001

Thank you, cjacquel

click to hide/show revision 3
None

updated 2017-10-06 02:30:04 -0600

berak gravatar image

Kmeans support needed

Hello,

cv::kmeans(data, 8, labels, cv::TermCriteria(CV_TERMCRIT_ITER, 10, 1.0), 3, cv::KMEANS_PP_CENTERS, &centers);

In kmeans Opencv, I have the following questions :

1)-

  1. What is the format of the mat labels? INT * 3 /pixel ? 2)- ?
  2. What is the format of centers ?
  3. 3)-
  4. For the TermCriteria, Some use CV_TERMCRIT_ITER|CV_TERMCRIT_EPS, so which to use for color reduction, and the two parameters, some enters 10000, 0.0001

Thank you, cjacquel

Kmeans support needed

Hello,

cv::kmeans(data, 8, labels, cv::TermCriteria(CV_TERMCRIT_ITER, 10, 1.0), 3, cv::KMEANS_PP_CENTERS, &centers);

In kmeans Opencv, I have the following questions :

  1. What is the format of the mat labels? INT * 3 /pixel ?
  2. What is the format of centers ?
  3. For the TermCriteria, Some use CV_TERMCRIT_ITER|CV_TERMCRIT_EPS, so which to use for color reduction, and the two parameters, some enters 10000, 0.0001

Thank you, cjacquel