Package georegression.fitting.curves

Class FitCurve_F32

java.lang.Object

georegression.fitting.curves.FitCurve_F32

@Generated("georegression.fitting.curves.FitCurve_F64")
public class FitCurve_F32
extends Object

Fits different types of curves to points

Constructor Summary

Constructors

Constructor	Description
FitCurve_F32()

Method Summary

Modifier and Type	Method	Description
static boolean	fit(float[] data, int offset, int length, PolynomialQuadratic2D_F32 output)	Fits points to a 2D quadratic polynomial.
static boolean	fitMM(float[] data, int offset, int length, PolynomialCubic1D_F32 output, @Nullable org.ejml.data.FMatrix4x4 A)	Fits points to the cubic polynomial.
static boolean	fitMM(float[] data, int offset, int length, PolynomialQuadratic1D_F32 output, @Nullable org.ejml.data.FMatrix3x3 work)	Fits points to the quadratic polynomial.
static boolean	fitQRP(float[] data, int offset, int length, PolynomialQuadratic1D_F32 output)	Fits points to the quadratic polynomial using a QRP linear solver.

Methods inherited from class java.lang.Object

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

Constructor Details

FitCurve_F32

public FitCurve_F32()

Method Details

fitMM

public static boolean fitMM(float[] data,
 int offset,
 int length,
 PolynomialQuadratic1D_F32 output,
 @Nullable
 @Nullable org.ejml.data.FMatrix3x3 work)

Fits points to the quadratic polynomial. It's recommended that you apply a linear transform to the points to ensure that they have zero mean and a standard deviation of 1. Then reverse the transform on the result. A solution is found using the pseudo inverse and inverse by minor matrices.

For a more numerically stable algorithm see fitQRP(float[], int, int, PolynomialQuadratic1D_F32)

Parameters:

data - Interleaved data [input[0], output[0], ....

offset - first index in data

length - number of elements in data that are to be read. must be divisible by 2

output - (Optional) storage for the curve

Returns:

The fitted curve

fitQRP

public static boolean fitQRP(float[] data,
 int offset,
 int length,
 PolynomialQuadratic1D_F32 output)

Fits points to the quadratic polynomial using a QRP linear solver.

Parameters:

data - Interleaved data [input[0], output[0], ....

offset - first index in data

length - number of elements in data that are to be read. must be divisible by 2

output - (Optional) storage for the curve

Returns:

The fitted curve

See Also:

• FitPolynomialSolverTall_F32

fitMM

public static boolean fitMM(float[] data,
 int offset,
 int length,
 PolynomialCubic1D_F32 output,
 @Nullable
 @Nullable org.ejml.data.FMatrix4x4 A)

Fits points to the cubic polynomial. It's recommended that you apply a linear transform to the points to ensure that they have zero mean and a standard deviation of 1. Then reverse the transform on the result. A solution is found using the pseudo inverse and inverse by minor matrices.

For a more numerically stable algorithm see fitQRP(float[], int, int, PolynomialQuadratic1D_F32)

Parameters:

data - Interleaved data [input[0], output[0], ....

offset - first index in data

length - number of elements in data that are to be read. must be divisible by 2

output - (Optional) storage for the curve

Returns:

The fitted curve

fit

public static boolean fit(float[] data,
 int offset,
 int length,
 PolynomialQuadratic2D_F32 output)

Fits points to a 2D quadratic polynomial. There are two inputs and one output for each data point. It's recommended that you apply a linear transform to the points.

Parameters:

data - Interleaved data [inputA[0], inputB[0], output[0], ....

offset - first index in data

length - number of elements in data that are to be read. must be divisible by 2

output - (Optional) storage for the curve

Returns:

The fitted curve