edu.jhu.bme.smile.commons.math
Class FieldRegularize

java.lang.Object
  edu.jhu.bme.smile.commons.math.FieldRegularize

public class FieldRegularize
extends java.lang.Object

public class FieldRegularize This class contains static methods to regularize (smooth) a 2d field using polynomial basis sets. Float.NaN values indicate points at which the data in the field should be ignored.

Author:

bennett

Constructor Summary

FieldRegularize()

Method Summary

static boolean	isFinite(float f)
static Matrix	pseudoInverse(Matrix data)
static float[][]	regularize2DChebyshev(int degX, int degY, float[][] field)
static float[][]	regularize2DChebyshevWLS(int degX, int degY, float[][] field, float[][] weight)
static float[][]	regularize2DChebyshevWLS(int degX, int degY, float[][] field, float[][] weight, Matrix result)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

FieldRegularize

public FieldRegularize()

Method Detail

isFinite

public static final boolean isFinite(float f)

Parameters:

f - - a floating point number

Returns:

boolean - true if f is not NaN and is finite

regularize2DChebyshev

public static final float[][] regularize2DChebyshev(int degX,
                                                    int degY,
                                                    float[][] field)

Parameters:

degX - - the number of degrees of freedom in the polynomial for the the x (1st) dimension

degY - - the number of degrees of freedom in the polynomial for the the y (2nd) dimension

field - - the data to be regularized

Returns:

- polynomial fit to the value with Chebyshev polynomials

regularize2DChebyshevWLS

public static final float[][] regularize2DChebyshevWLS(int degX,
                                                       int degY,
                                                       float[][] field,
                                                       float[][] weight)

Parameters:

degX - - the number of degrees of freedom in the polynomial for the the x (1st) dimension

degY - - the number of degrees of freedom in the polynomial for the the y (2nd) dimension

field - - the data to be regularized

weight - - pixel-by-pixel weightings for polynomial weighting

Returns:

- polynomial fit to the value with Chebyshev polynomials using weighted least squares

regularize2DChebyshevWLS

public static final float[][] regularize2DChebyshevWLS(int degX,
                                                       int degY,
                                                       float[][] field,
                                                       float[][] weight,
                                                       Matrix result)

Parameters:

degX - - the number of degrees of freedom in the polynomial for the the x (1st) dimension

degY - - the number of degrees of freedom in the polynomial for the the y (2nd) dimension

field - - the data to be regularized

weight - - pixel-by-pixel weightings for polynomial weighting

result - - Matrix to be used for storage of the resulting polynomial fit

Returns:

- polynomial fit to the value with Chebyshev polynomials using weighted least squares

pseudoInverse

public static Matrix pseudoInverse(Matrix data)

Parameters:

data - - input matrix

Returns:

the pseudo-inverse of the input matrix