boofcv.alg.feature.detect.grid

Class FitGaussianPrune

java.lang.Object

boofcv.alg.feature.detect.grid.FitGaussianPrune

public class FitGaussianPrune
extends java.lang.Object

Given a histogram of pixel intensity values and a valid range of indexes in the histogram, compute the mean and standard deviation. Prune bins which go outside the preset threshold.

Author:

Peter Abeles

Field Summary

Fields

Modifier and Type	Field and Description
protected int	indexHigh
protected int	indexLow

Constructor Summary

Constructors

Constructor and Description
FitGaussianPrune(int maxIterations, double thresholdSigma, int minSeparation)

Method Summary

Methods

Modifier and Type	Method and Description
double	getMean()
double	getSigma()
void	process(IntensityHistogram h, int low, int high) Computes statistics of pixel intensity values inside histogram.

Methods inherited from class java.lang.Object

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

Field Detail

indexLow

protected int indexLow

indexHigh

protected int indexHigh

Constructor Detail

FitGaussianPrune

public FitGaussianPrune(int maxIterations,
                double thresholdSigma,
                int minSeparation)

Parameters:

maxIterations - Maximum number of iterations. Try 20

thresholdSigma - Number of standard deviations a point is away for it to be pruned. Try 4

minSeparation - A bin will not be pruned if it is this close to the mean. 1 or 3 typically

Method Detail

process

public void process(IntensityHistogram h,
           int low,
           int high)

Computes statistics of pixel intensity values inside histogram. Valid range is indexLow <= i < indexHigh

Parameters:

h - Histogram being processed

low - Index of the lower bound being analyzed

high - Index of the upper bound being analyzed

getMean

public double getMean()

getSigma

public double getSigma()