OpenMS
GaussFilterAlgorithm Class Reference

This class represents a Gaussian lowpass-filter which works on uniform as well as on non-uniform profile data. More...

#include <OpenMS/PROCESSING/SMOOTHING/GaussFilterAlgorithm.h>

Collaboration diagram for GaussFilterAlgorithm:
[legend]

Public Member Functions

 GaussFilterAlgorithm ()
 Constructor. More...
 
virtual ~GaussFilterAlgorithm ()
 Destructor. More...
 
bool filter (OpenMS::Interfaces::SpectrumPtr spectrum)
 Smoothes an Spectrum containing profile data. More...
 
bool filter (OpenMS::Interfaces::ChromatogramPtr chromatogram)
 Smoothes an Chromatogram containing profile data. More...
 
template<typename ConstIterT , typename IterT >
bool filter (ConstIterT mz_in_start, ConstIterT mz_in_end, ConstIterT int_in_start, IterT mz_out, IterT int_out)
 Smoothes two data arrays. More...
 
void initialize (double gaussian_width, double spacing, double ppm_tolerance, bool use_ppm_tolerance)
 

Protected Member Functions

template<typename InputPeakIterator >
double integrate_ (InputPeakIterator x, InputPeakIterator y, InputPeakIterator first, InputPeakIterator last)
 Computes the convolution of the raw data at position x and the gaussian kernel. More...
 

Protected Attributes

std::vector< double > coeffs_
 Coefficients. More...
 
double sigma_
 The standard derivation \( \sigma \). More...
 
double spacing_
 The spacing of the pre-tabulated kernel coefficients. More...
 
bool use_ppm_tolerance_
 
double ppm_tolerance_
 

Detailed Description

This class represents a Gaussian lowpass-filter which works on uniform as well as on non-uniform profile data.

Gaussian filters are important in many signal processing, image processing, and communication applications. These filters are characterized by narrow bandwidths, sharp cutoffs, and low passband ripple. A key feature of Gaussian filters is that the Fourier transform of a Gaussian is also a Gaussian, so the filter has the same response shape in both the time and frequency domains. The coefficients \( \emph{coeffs} \) of the Gaussian-window with length \( \emph{frameSize} \) are calculated from the gaussian distribution

\[ \emph{coeff}(x) = \frac{1}{\sigma \sqrt{2\pi}} e^{\frac{-x^2}{2\sigma^2}} \]

where \( x=[-\frac{frameSize}{2},...,\frac{frameSize}{2}] \) represents the window area and \( \sigma \) is the standard derivation.

Note
The wider the kernel width the smoother the signal (the more detail information get lost!). Use a gaussian filter kernel which has approximately the same width as your mass peaks, whereas the gaussian peak width corresponds approximately to 8*sigma.
The data must be sorted according to ascending m/z!

Constructor & Destructor Documentation

◆ GaussFilterAlgorithm()

Constructor.

◆ ~GaussFilterAlgorithm()

virtual ~GaussFilterAlgorithm ( )
virtual

Destructor.

Member Function Documentation

◆ filter() [1/3]

bool filter ( ConstIterT  mz_in_start,
ConstIterT  mz_in_end,
ConstIterT  int_in_start,
IterT  mz_out,
IterT  int_out 
)
inline

Smoothes two data arrays.

Convolutes the filter and the profile data and writes the results into the output iterators mz_out and int_out.

References OpenMS::Math::ppmToMass().

◆ filter() [2/3]

bool filter ( OpenMS::Interfaces::ChromatogramPtr  chromatogram)
inline

Smoothes an Chromatogram containing profile data.

◆ filter() [3/3]

bool filter ( OpenMS::Interfaces::SpectrumPtr  spectrum)
inline

Smoothes an Spectrum containing profile data.

◆ initialize()

void initialize ( double  gaussian_width,
double  spacing,
double  ppm_tolerance,
bool  use_ppm_tolerance 
)

◆ integrate_()

double integrate_ ( InputPeakIterator  x,
InputPeakIterator  y,
InputPeakIterator  first,
InputPeakIterator  last 
)
inlineprotected

Computes the convolution of the raw data at position x and the gaussian kernel.

References OpenSwath::norm().

Member Data Documentation

◆ coeffs_

std::vector<double> coeffs_
protected

Coefficients.

◆ ppm_tolerance_

double ppm_tolerance_
protected

◆ sigma_

double sigma_
protected

The standard derivation \( \sigma \).

◆ spacing_

double spacing_
protected

The spacing of the pre-tabulated kernel coefficients.

◆ use_ppm_tolerance_

bool use_ppm_tolerance_
protected