www.pudn.com > firev0.01.rar > gabor.hpp
/*
This file is part of the FIRE -- Flexible Image Retrieval System
FIRE is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
FIRE is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with FIRE; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
/**
* @file gabor.hpp
* @author Thomas Deselaers
* @date Mon Jun 23 19:29:12 2003
*
* @brief Various functions dealing with gabor transformation.
*
*/
#ifndef __gabor_hpp
#define __gabor_hpp
#include
#include
#include
#include "hdimage.hpp"
#include "rgbimage.hpp"
#include "image.hpp"
#include "basetools.hpp"
/**
* save the gabor features from an image into a text file.
* @param image the ::img::HDImage containing the gaborfeatures.
* @param filename the filename of the file, where the features shall be saved.
*
*/
int saveGaborFeatures(::img::HDImage image, ::std::string filename) ;
/**
* load gabor features from a file
*
*/
::std::vector< ::std::vector* > loadGaborFeatures(string filename, const bool withcoords=true);
/**
* Extract gabor features from a color image
*
* @param filename the file to save the gaborfeatures to
* @param inImage the image to be processed
* @param decomp the decomposed version of the image
* @param threshold the threshold for local variance to determine which features to save
* @param winsize the window size for local variance calculation
*
* @return the number of features extracted
*/
int extractGaborFeatures(::std::string filename, const ::img::RGBImage & inImage, const ::img::HDImage &decomp,double threshold, unsigned int winsize) ;
/**
* Extract gabor features from a color image
*
* @param filename the file to save the gaborfeatures to
* @param inImage the image to be processed
* @param decomp the decomposed version of the image
* @param nOfFeatures the number of features to be extracted. The ones with highest local variance are extracted.
* @param winsize the window size for local variance calculation
*
* @return the number of features extracted
*/
int extractGaborFeatures(::std::string filename, const ::img::RGBImage & inImage, const ::img::HDImage &decomp,int nOfFeatures, unsigned int winsize);
/**
* Extract gabor features from a single layer image
*
* @param filename the file to save the gaborfeatures to
* @param inImage the image to be processed
* @param decomp the decomposed version of the image
* @param threshold the threshold for local variance to determine which features to save
* @param winsize the window size for local variance calculation
*
* @return the number of features extracted
*/
int extractGaborFeatures(::std::string filename, const ::img::BaseImage & inImage, const ::img::HDImage &decomp,double threshold, unsigned int winsize);
/**
* Extract gabor features from a single layer image
*
* @param filename the file to save the gaborfeatures to
* @param inImage the image to be processed
* @param decomp the decomposed version of the image
* @param nOfFeatures the number of features to be extracted. The ones with highest local variance are extracted.
* @param winsize the window size for local variance calculation
*
* @return the number of features extracted
*/
int extractGaborFeatures(::std::string filename, const ::img::BaseImage & inImage, const ::img::HDImage &decomp,int nOfFeatures, unsigned int winsize);
/**
* return the integer power of two
*
* @param input the number for which the next larger 2^x is search with x integer
*
* @return the first exponential of two larger than the input
*/
unsigned int next2Power(const unsigned int input) ;
/**
* return the position in a one dimensional array which is interpreted
* as two dimensional. No checking is done if this is possible.
*
* @param y the y position
* @param x the x position
* @param width the assumend width
*
* @return the position in a one dimensional array
*/
///return the index for a field at position (y,x) if width is assumed.
inline const unsigned int idx(const unsigned int& y,const unsigned int& x,const unsigned int& width) ;
/**
* Copy the Results from the Gabor filter into two layers of the output image
*
* @param result the ::img::HDImage where all the gabor answers will be collected
* @param HStoTransform the gaboranswers for the ::std::complex layer. In the format the NR-Routine fourn needs it
* @param VtoTransform the gaboransers for the value layer. In the format the NR-Routine fourn needs it
* @param height The Height of the important part of the image. (to skip padded values)
* @param width the width of the important part of the image. (to skip padded values)
* @param maxnn the dimensionality of the complete image (width=height=maxnn=2^x, x integer)
* @param filterno The number of the filter beeing applied, to know which layers have to be writen in result
*/
void copyToResultRGB(::img::HDImage &result,const ::std::vector &HStoTransform, const ::std::vector &VtoTransform,const unsigned int height,const unsigned int width, const unsigned int maxnn, const unsigned int filterno) ;
/**
* Save the ::std::complex and the value image to be able to see the results. Normalization is done within here
*
* @param HStoTransform the gaboranswers for the ::std::complex layer. In the format the NR-Routine fourn needs it
* @param VtoTransform the gaboransers for the value layer. In the format the NR-Routine fourn needs it
* @param freq The frequency of the used filter (used for filename generation)
* @param deviation The deviation of the used filter (used for filename generation)
* @param pha The phase of the used filter (used for filename generation)
* @param ratio The ratio of the used filter (used for filename generation)
* @param height The Height of the important part of the image. (to skip padded values)
* @param width the width of the important part of the image. (to skip padded values)
* @param maxnn the dimensionality of the complete image (width=height=maxnn=2^x, x integer)
*/
void saveHSVTransformed(const ::std::vector &HStoTransform,
const ::std::vector &VtoTransform,
double freq, double deviation, double pha, double ratio,
unsigned int height, unsigned int width,unsigned int maxnn) ;
/**
* Save the filter as png. For visual control of results.
*
* @param filter the filter to be saved
* @param freq The frequency of the used filter (used for filename generation)
* @param deviation The deviation of the used filter (used for filename generation)
* @param phase The phase of the used filter (used for filename generation)
* @param ratio The ratio of the used filter (used for filename generation)
*/
void saveFilter(const ::std::vector > &filter, double freq, double deviation, double phase, double ratio);
/**
* Make Gabor-Jet decomposition for an image.
*
* @param in the image to be analyzed
* @param numpha the number of phases used for this
*
* @return the ::img::HDImage contains in each "layer" a response from either
* a Gaborfilter on a ::std::complex image or the v image.
*/
///direct reimplementation of ldecompose.c from keysers
::img::HDImage decompose(::img::RGBImage in, unsigned int numpha=8, int numfreq=-1);
/**
* Make Gabor-Jet decomposition for an image.
*
* @param in the image to be analyzed
* @param numpha the number of phases used for this
*
* @return the ::img::HDImage contains in each "layer" a response from either
* a Gaborfilter.
*/
///direct reimplementation of ldecompose.c from keysers
::img::HDImage decompose(::img::BaseImage in, unsigned int numpha=8, int numfreq=-1);
#endif