org.opencv.ml.EM Maven / Gradle / Ivy
Show all versions of opencv Show documentation
//
// This file is auto-generated. Please don't modify it!
//
package org.opencv.ml;
import org.opencv.core.Algorithm;
import org.opencv.core.Mat;
import org.opencv.core.TermCriteria;
// C++: class EM
/**
* The class implements the EM algorithm as described in the beginning of this
* section. It is inherited from "Algorithm".
*
* @see org.opencv.ml.EM : public Algorithm
*/
public class EM extends Algorithm {
protected EM(long addr) { super(addr); }
public static final int
COV_MAT_SPHERICAL = 0,
COV_MAT_DIAGONAL = 1,
COV_MAT_GENERIC = 2,
COV_MAT_DEFAULT = COV_MAT_DIAGONAL,
DEFAULT_NCLUSTERS = 5,
DEFAULT_MAX_ITERS = 100,
START_E_STEP = 1,
START_M_STEP = 2,
START_AUTO_STEP = 0;
//
// C++: EM::EM(int nclusters = EM::DEFAULT_NCLUSTERS, int covMatType = EM::COV_MAT_DIAGONAL, TermCriteria termCrit = TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, EM::DEFAULT_MAX_ITERS, FLT_EPSILON))
//
/**
* The constructor of the class
*
* @param nclusters The number of mixture components in the Gaussian mixture
* model. Default value of the parameter is EM.DEFAULT_NCLUSTERS=5.
* Some of EM implementation could determine the optimal number of mixtures
* within a specified value range, but that is not the case in ML yet.
* @param covMatType Constraint on covariance matrices which defines type of
* matrices. Possible values are:
*
* - EM.COV_MAT_SPHERICAL A scaled identity matrix mu_k * I.
* There is the only parameter mu_k to be estimated for each matrix.
* The option may be used in special cases, when the constraint is relevant, or
* as a first step in the optimization (for example in case when the data is
* preprocessed with PCA). The results of such preliminary estimation may be
* passed again to the optimization procedure, this time with
covMatType=EM.COV_MAT_DIAGONAL.
* - EM.COV_MAT_DIAGONAL A diagonal matrix with positive diagonal
* elements. The number of free parameters is
d for each matrix.
* This is most commonly used option yielding good estimation results.
* - EM.COV_MAT_GENERIC A symmetric positively defined matrix. The number
* of free parameters in each matrix is about d^2/2. It is not
* recommended to use this option, unless there is pretty accurate initial
* estimation of the parameters and/or a huge number of training samples.
*
* @param termCrit The termination criteria of the EM algorithm. The EM
* algorithm can be terminated by the number of iterations termCrit.maxCount
* (number of M-steps) or when relative change of likelihood logarithm is less
* than termCrit.epsilon. Default maximum number of iterations is
* EM.DEFAULT_MAX_ITERS=100.
*
* @see org.opencv.ml.EM.EM
*/
public EM(int nclusters, int covMatType, TermCriteria termCrit)
{
super( EM_0(nclusters, covMatType, termCrit.type, termCrit.maxCount, termCrit.epsilon) );
return;
}
/**
* The constructor of the class
*
* @see org.opencv.ml.EM.EM
*/
public EM()
{
super( EM_1() );
return;
}
//
// C++: void EM::clear()
//
public void clear()
{
clear_0(nativeObj);
return;
}
//
// C++: bool EM::isTrained()
//
public boolean isTrained()
{
boolean retVal = isTrained_0(nativeObj);
return retVal;
}
//
// C++: Vec2d EM::predict(Mat sample, Mat& probs = Mat())
//
/**
* Returns a likelihood logarithm value and an index of the most probable
* mixture component for the given sample.
*
* The method returns a two-element double vector. Zero element is
* a likelihood logarithm value for the sample. First element is an index of the
* most probable mixture component for the given sample.
*
* @param sample A sample for classification. It should be a one-channel matrix
* of 1 x dims or dims x 1 size.
* @param probs Optional output matrix that contains posterior probabilities of
* each component given the sample. It has 1 x nclusters size and
* CV_64FC1 type.
*
* @see org.opencv.ml.EM.predict
*/
public double[] predict(Mat sample, Mat probs)
{
double[] retVal = predict_0(nativeObj, sample.nativeObj, probs.nativeObj);
return retVal;
}
/**
* Returns a likelihood logarithm value and an index of the most probable
* mixture component for the given sample.
*
* The method returns a two-element double vector. Zero element is
* a likelihood logarithm value for the sample. First element is an index of the
* most probable mixture component for the given sample.
*
* @param sample A sample for classification. It should be a one-channel matrix
* of 1 x dims or dims x 1 size.
*
* @see org.opencv.ml.EM.predict
*/
public double[] predict(Mat sample)
{
double[] retVal = predict_1(nativeObj, sample.nativeObj);
return retVal;
}
//
// C++: bool EM::train(Mat samples, Mat& logLikelihoods = Mat(), Mat& labels = Mat(), Mat& probs = Mat())
//
/**
* Estimates the Gaussian mixture parameters from a samples set.
*
* Three versions of training method differ in the initialization of Gaussian
* mixture model parameters and start step:
*
* - train - Starts with Expectation step. Initial values of the model
* parameters will be estimated by the k-means algorithm.
*
- trainE - Starts with Expectation step. You need to provide initial
* means a_k of mixture components. Optionally you can pass initial
* weights pi_k and covariance matrices S_k of mixture
* components.
*
- trainM - Starts with Maximization step. You need to provide initial
* probabilities p_(i,k) to use this option.
*
*
* The methods return true if the Gaussian mixture model was
* trained successfully, otherwise it returns false.
*
* Unlike many of the ML models, EM is an unsupervised learning algorithm and it
* does not take responses (class labels or function values) as input. Instead,
* it computes the *Maximum Likelihood Estimate* of the Gaussian mixture
* parameters from an input sample set, stores all the parameters inside the
* structure: p_(i,k) in probs, a_k in
* means, S_k in covs[k], pi_k in
* weights, and optionally computes the output "class label" for
* each sample: labels_i=arg max_k(p_(i,k)), i=1..N (indices of the
* most probable mixture component for each sample).
*
* The trained model can be used further for prediction, just like any other
* classifier. The trained model is similar to the "CvNormalBayesClassifier".
*
* @param samples Samples from which the Gaussian mixture model will be
* estimated. It should be a one-channel matrix, each row of which is a sample.
* If the matrix does not have CV_64F type it will be converted to
* the inner matrix of such type for the further computing.
* @param logLikelihoods The optional output matrix that contains a likelihood
* logarithm value for each sample. It has nsamples x 1 size and
* CV_64FC1 type.
* @param labels The optional output "class label" for each sample:
* labels_i=arg max_k(p_(i,k)), i=1..N (indices of the most probable
* mixture component for each sample). It has nsamples x 1 size and
* CV_32SC1 type.
* @param probs The optional output matrix that contains posterior probabilities
* of each Gaussian mixture component given the each sample. It has nsamples
* x nclusters size and CV_64FC1 type.
*
* @see org.opencv.ml.EM.train
*/
public boolean train(Mat samples, Mat logLikelihoods, Mat labels, Mat probs)
{
boolean retVal = train_0(nativeObj, samples.nativeObj, logLikelihoods.nativeObj, labels.nativeObj, probs.nativeObj);
return retVal;
}
/**
* Estimates the Gaussian mixture parameters from a samples set.
*
* Three versions of training method differ in the initialization of Gaussian
* mixture model parameters and start step:
*
* - train - Starts with Expectation step. Initial values of the model
* parameters will be estimated by the k-means algorithm.
*
- trainE - Starts with Expectation step. You need to provide initial
* means a_k of mixture components. Optionally you can pass initial
* weights pi_k and covariance matrices S_k of mixture
* components.
*
- trainM - Starts with Maximization step. You need to provide initial
* probabilities p_(i,k) to use this option.
*
*
* The methods return true if the Gaussian mixture model was
* trained successfully, otherwise it returns false.
*
* Unlike many of the ML models, EM is an unsupervised learning algorithm and it
* does not take responses (class labels or function values) as input. Instead,
* it computes the *Maximum Likelihood Estimate* of the Gaussian mixture
* parameters from an input sample set, stores all the parameters inside the
* structure: p_(i,k) in probs, a_k in
* means, S_k in covs[k], pi_k in
* weights, and optionally computes the output "class label" for
* each sample: labels_i=arg max_k(p_(i,k)), i=1..N (indices of the
* most probable mixture component for each sample).
*
* The trained model can be used further for prediction, just like any other
* classifier. The trained model is similar to the "CvNormalBayesClassifier".
*
* @param samples Samples from which the Gaussian mixture model will be
* estimated. It should be a one-channel matrix, each row of which is a sample.
* If the matrix does not have CV_64F type it will be converted to
* the inner matrix of such type for the further computing.
*
* @see org.opencv.ml.EM.train
*/
public boolean train(Mat samples)
{
boolean retVal = train_1(nativeObj, samples.nativeObj);
return retVal;
}
//
// C++: bool EM::trainE(Mat samples, Mat means0, Mat covs0 = Mat(), Mat weights0 = Mat(), Mat& logLikelihoods = Mat(), Mat& labels = Mat(), Mat& probs = Mat())
//
public boolean trainE(Mat samples, Mat means0, Mat covs0, Mat weights0, Mat logLikelihoods, Mat labels, Mat probs)
{
boolean retVal = trainE_0(nativeObj, samples.nativeObj, means0.nativeObj, covs0.nativeObj, weights0.nativeObj, logLikelihoods.nativeObj, labels.nativeObj, probs.nativeObj);
return retVal;
}
public boolean trainE(Mat samples, Mat means0)
{
boolean retVal = trainE_1(nativeObj, samples.nativeObj, means0.nativeObj);
return retVal;
}
//
// C++: bool EM::trainM(Mat samples, Mat probs0, Mat& logLikelihoods = Mat(), Mat& labels = Mat(), Mat& probs = Mat())
//
public boolean trainM(Mat samples, Mat probs0, Mat logLikelihoods, Mat labels, Mat probs)
{
boolean retVal = trainM_0(nativeObj, samples.nativeObj, probs0.nativeObj, logLikelihoods.nativeObj, labels.nativeObj, probs.nativeObj);
return retVal;
}
public boolean trainM(Mat samples, Mat probs0)
{
boolean retVal = trainM_1(nativeObj, samples.nativeObj, probs0.nativeObj);
return retVal;
}
@Override
protected void finalize() throws Throwable {
delete(nativeObj);
}
// C++: EM::EM(int nclusters = EM::DEFAULT_NCLUSTERS, int covMatType = EM::COV_MAT_DIAGONAL, TermCriteria termCrit = TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, EM::DEFAULT_MAX_ITERS, FLT_EPSILON))
private static native long EM_0(int nclusters, int covMatType, int termCrit_type, int termCrit_maxCount, double termCrit_epsilon);
private static native long EM_1();
// C++: void EM::clear()
private static native void clear_0(long nativeObj);
// C++: bool EM::isTrained()
private static native boolean isTrained_0(long nativeObj);
// C++: Vec2d EM::predict(Mat sample, Mat& probs = Mat())
private static native double[] predict_0(long nativeObj, long sample_nativeObj, long probs_nativeObj);
private static native double[] predict_1(long nativeObj, long sample_nativeObj);
// C++: bool EM::train(Mat samples, Mat& logLikelihoods = Mat(), Mat& labels = Mat(), Mat& probs = Mat())
private static native boolean train_0(long nativeObj, long samples_nativeObj, long logLikelihoods_nativeObj, long labels_nativeObj, long probs_nativeObj);
private static native boolean train_1(long nativeObj, long samples_nativeObj);
// C++: bool EM::trainE(Mat samples, Mat means0, Mat covs0 = Mat(), Mat weights0 = Mat(), Mat& logLikelihoods = Mat(), Mat& labels = Mat(), Mat& probs = Mat())
private static native boolean trainE_0(long nativeObj, long samples_nativeObj, long means0_nativeObj, long covs0_nativeObj, long weights0_nativeObj, long logLikelihoods_nativeObj, long labels_nativeObj, long probs_nativeObj);
private static native boolean trainE_1(long nativeObj, long samples_nativeObj, long means0_nativeObj);
// C++: bool EM::trainM(Mat samples, Mat probs0, Mat& logLikelihoods = Mat(), Mat& labels = Mat(), Mat& probs = Mat())
private static native boolean trainM_0(long nativeObj, long samples_nativeObj, long probs0_nativeObj, long logLikelihoods_nativeObj, long labels_nativeObj, long probs_nativeObj);
private static native boolean trainM_1(long nativeObj, long samples_nativeObj, long probs0_nativeObj);
// native support for java finalize()
private static native void delete(long nativeObj);
}