http://www.emgu.com
Starts with Expectation step. Initial values of the model parameters will be estimated by the k-means algorithm.

Namespace: Emgu.CV.ML
Assembly: Emgu.CV.ML (in Emgu.CV.ML.dll) Version: 2.4.10.1935 (2.4.10.1935)

Syntax

C#
public static bool CvEMTrain(
	IntPtr model,
	IntPtr samples,
	IntPtr labels,
	IntPtr probs,
	IntPtr logLikelihoods
)
Visual Basic
Public Shared Function CvEMTrain ( 
	model As IntPtr,
	samples As IntPtr,
	labels As IntPtr,
	probs As IntPtr,
	logLikelihoods As IntPtr
) As Boolean
Visual C++
public:
static bool CvEMTrain(
	IntPtr model, 
	IntPtr samples, 
	IntPtr labels, 
	IntPtr probs, 
	IntPtr logLikelihoods
)
F#
static member CvEMTrain : 
        model : IntPtr * 
        samples : IntPtr * 
        labels : IntPtr * 
        probs : IntPtr * 
        logLikelihoods : IntPtr -> bool 

Parameters

model
Type: System..::..IntPtr
The EM model
samples
Type: System..::..IntPtr
The training data. A 32-bit floating-point, single-channel matrix, one vector per row
labels
Type: System..::..IntPtr
Can be IntPtr.Zero if not needed. Optionally computed output "class label" for each sample
probs
Type: System..::..IntPtr
Initial probabilities p_{i,k} of sample i to belong to mixture component k. It is a one-channel floating-point matrix of nsamples x nclusters size.
logLikelihoods
Type: System..::..IntPtr
The optional output matrix that contains a likelihood logarithm value for each sample. It has nsamples x 1 size and CV_64FC1 type.

Return Value

Type: Boolean
The methods return true if the Gaussian mixture model was trained successfully, otherwise it returns false.

See Also