Package sc.fiji.snt.analysis

Class CircularModels

java.lang.Object

sc.fiji.snt.analysis.CircularModels

public final class CircularModels
extends Object

Shared utilities for circular modeling, including von Mises fits.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static enum	CircularModels.Domain	Domain of the angle data.
static final record	CircularModels.VonMisesFit	Immutable summary of a von Mises fit on circular (directional) or axial (orientation) data.

Method Summary

Modifier and Type	Method	Description
static CircularModels.VonMisesFit	fitFromHistogram(double[] weights, double angleStepDeg, CircularModels.Domain domain)	Fit von Mises to a weighted histogram over angle bins.
static CircularModels.VonMisesFit	fitFromPairs(double[] anglesDeg, double[] weights, CircularModels.Domain domain)	Fit von Mises to angle/weight pairs (angles in degrees).
static double	kappaFromRBar(double R)	Returns κ from \bar R using standard fast approximations as per Fisher et. al., Statistical Analysis of Spherical Data and Matlab's CircStat package by Philipp Berens
static double	norm180(double deg)	Normalizes deg to [0,180[
static double	norm360(double deg)	Normalizes deg to [0,360).

Methods inherited from class java.lang.Object

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

Method Details

kappaFromRBar

public static double kappaFromRBar(double R)

Returns κ from \bar R using standard fast approximations as per Fisher et. al., Statistical Analysis of Spherical Data and Matlab's CircStat package by Philipp Berens

norm360

public static double norm360(double deg)

Normalizes deg to [0,360).

norm180

public static double norm180(double deg)

Normalizes deg to [0,180[

fitFromHistogram

public static CircularModels.VonMisesFit fitFromHistogram(double[] weights,
 double angleStepDeg,
 CircularModels.Domain domain)

Fit von Mises to a weighted histogram over angle bins. Angles are assumed to be bin centers: (i+0.5)*angleStepDeg.

Parameters:

weights - non-negative bin heights, length = nBins

angleStepDeg - bin width in degrees (360/nBins)

domain - DIRECTIONAL (0–360) or AXIAL (0–180)

fitFromPairs

public static CircularModels.VonMisesFit fitFromPairs(double[] anglesDeg,
 double[] weights,
 CircularModels.Domain domain)

Fit von Mises to angle/weight pairs (angles in degrees).

Parameters:

anglesDeg - angles in degrees; domain interpreted by argument

weights - non-negative weights; null => all ones