nnshaperecognizer.h

An open source handwriting recongnition package!!!

/*****************************************************************************************
* Copyright (c) 2006 Hewlett-Packard Development Company, L.P.
* Permission is hereby granted, free of charge, to any person obtaining a copy of
* this software and associated documentation files (the "Software"), to deal in
* the Software without restriction, including without limitation the rights to use,
* copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
* Software, and to permit persons to whom the Software is furnished to do so,
* subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
* INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
* PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
* HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
* CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE
* OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*****************************************************************************************/

/************************************************************************
 * SVN MACROS
 *
 * $LastChangedDate: 2008-09-02 12:09:40 +0530 (Tue, 02 Sep 2008) $
 * $Revision: 682 $
 * $Author: sharmnid $
 *
 ************************************************************************/
/************************************************************************
 * FILE DESCR: Definitions for NN Shape Recognition module
 *
 * CONTENTS:
 *
 * AUTHOR:     Saravanan R.
 *
 * DATE:       January 23, 2007
 * CHANGE HISTORY:
 * Author		Date			Description of change
 ************************************************************************/

#ifndef __NNSHAPERECOGNIZER_H
#define __NNSHAPERECOGNIZER_H

/**			Include files		*/
#include "LTKInc.h"
#include "LTKTypes.h"
#include "LTKMacros.h"
#include "LTKShapeRecognizer.h"
#include "LTKShapeRecoUtil.h"
#include "LTKShapeSample.h"
#include "LTKCheckSumGenerate.h"
#include "LTKDynamicTimeWarping.h"

/**			Forward declaration of classes		*/
class LTKTraceGroup;
class LTKPreprocessorInterface;
class LTKShapeSample;
class LTKShapeFeatureExtractor;
class LTKShapeFeature;

#define SIMILARITY(distance) (1 / (distance + EPS ))
#define SUPPORTED_MIN_VERSION	"2.1.0"

#ifdef _INTERNAL
class LTKAdapt;
#endif

typedef int (*FN_PTR_LOCAL_DISTANCE)(LTKShapeFeaturePtr, LTKShapeFeaturePtr,float&);
typedef int (*FN_PTR_CREATELTKLIPIPREPROCESSOR)(const LTKControlInfo& , LTKPreprocessorInterface** );
typedef int (*FN_PTR_DELETELTKLIPIPREPROCESSOR)(LTKPreprocessorInterface* );


/**
 * @ingroup NNShapeRecognizer.h
 * @brief The Header file for the NNShapeRecognizer
 * @class NNShapeRecognizer
 *<p> <p>
 */

class NNShapeRecognizer: public LTKShapeRecognizer
{

    public:
#ifdef _INTERNAL
        friend class LTKAdapt;
        int adapt(int shapeID );
        int adapt(const LTKTraceGroup& sampleTraceGroup, int shapeID );
        
        
    
    private:		

        //int NNShapeRecognizer::readInternalClassifierConfig();
        int readInternalClassifierConfig();

        int deleteAdaptInstance();

        

         /**
         * This method trains the classifier from the feature file whose path is passed as paramater
         *
         * Semantics
         *
         *
         * @param featureFilePath : string : Holds the path of Feature file
         *
         * @return none
         */

        int trainFromFeatureFile(const string& featureFilePath, 
                                    vector<LTKShapeSample>* trainSet=NULL);

        




#endif

        /** @name private data members */
        //@{
    private:
		
        FN_PTR_DELETELTKLIPIPREPROCESSOR m_deleteLTKLipiPreProcessor;
        //Function pointer for deleteLTKLipiPreProcessor

        // preproc lin handle
        void *m_libHandler;

        // feature extractor lin handle
        void *m_libHandlerFE;

        unsigned short m_numShapes;
        /**< @brief Number of shapes
         *     <p>
         *     It Defines the number of shapes to be recognized
         *
         *	DEFAULT: 0
         *
         *	Note: If the project is dynamic, then this numShapes was set to 0
         *		  If the project is not dynamic, then the numShapes was read from project configuration file
         *	</p>
         */

        string m_prototypeSelection;
        /**< @brief The Prototype Selection
         *	<p>
         *	if Prototype Selection = clustering, the training method used was clustering
         *					    = lvq,		 the training method used was LVQ
         *
         *	DEFAULT: LTKPreprocDefaults::NN_DEF_PROTOTYPESELECTION
         *	Possible values are "clustering" and "lvq"
         *	</p>
         */

        int m_prototypeReductionFactor;
        /**< @brief The prototype Reduction factor
         * 	<p>
         * 	if PrototypeReductionFactor 	= 0 every training sample is cluster on its own
         *     	                      		  	= 100 all training samples are represented by one prototype
         *                            		  	= 80 then all samples are represented by 20% of the training samples
         *
         *	DEFAULT: LTKPreprocDefaults::NN_DEF_PROTOTYPEREDUCTIONFACTOR
         *	RANGE:	0 TO 100
         * 	</p>
         */

        int m_numClusters;
        /**< @brief The number of clusters
         * 	<p>
         * 	if NumClusters = k, then k clusters are found from the training samples
         *
         *
         *
         *	DEFAULT: There is no default as this and prototype reduction factor are dependent
         *	RANGE:
         * 	</p>
         */


        string m_prototypeDistance;
        /**< @brief The Prototype Distance
         * <p>
         * if PrototypeDistance  =  eu, then the distance between the samples can be calculated using the Euclidean distance method
         *                       =  dtw, then the distance between the samples can be calculated using the DTW method
         *
         *	DEFAULT: LTKPreprocDefaults::NN_DEF_PROTOTYPEDISTANCE
         * 	Possible values are LTKMacros::DTW_DISTANCE and LTKMacros::EUCLIDEAN_DISTANCE.
         * </p>
         */

        int m_nearestNeighbors;
        /**< @brief Nearest Neighbors
         *	<p>
         *
         *	DEFAULT: LTKPreprocDefaults::NN_DEF_NEARESTNEIGHBORS
         *	</p>
         */


//        int m_dtwBanding;
        float m_dtwBanding;
        /**< @brief  DTW Banding
         *	<p>
         *
         *	DEFAULT: LTKPreprocDefaults::NN_DEF_BANDING
         *	</p>
         */

        int m_dtwEuclideanFilter;
        /**< @brief DTW Euclidean Filter
         *	<p>
         *
         *	DEFAULT: LTKPreprocDefaults::NN_DEF_DTWEUCLIDEANFILTER
         *	</p>
         */

        string m_featureExtractorName;
        /**< @brief The Feature Extractor
         * <p>
         *
         *	DEFAULT:
         *
         * </p>
         */

        bool m_projectTypeDynamic;
        /**< @brief Project Dynamic
         * 	<p>
         * 	if projectTypeDynamic  	= true, then the project is dynamic ie, the numShapes can take any number of value
         *		         	= false, then the project is not dynamic ie, the numShape can take value specified in project.cfg file
         *
         *	DEFAULT: false
         * 	</p>
         */

        LTKPreprocessorInterface *m_ptrPreproc;
        /**< @brief Pointer to preprocessor instance
         *	<p>
         *	Instance which is used to call the preprocessing methods before recognition
         *
         *	DEFAULT: NULL
         *	</p>
         */

        string m_nnCfgFilePath;
        /**< @brief Full path of NN configuration file
         * 	<p>
         *	Assigned value in the NNShapeRecognizer::initialize function
         * 	</p>
         */

        string m_nnMDTFilePath;
        /**< @brief Full path of Model data file
         *	<p>
         *   	Assigned value in the NNShapeRecognizer::initialize function
         * 	</p>
         */

        stringStringMap m_headerInfo;
        /**< @brief Header Information
         *	<p>
         *	</p>
         */

        LTKShapeRecoUtil m_shapeRecUtil;
        /**< @brief Pointer to LTKShapeRecoUtil class
         *	<p>
         *	Instance which used to call Shape Recognizer Utility functions
         *
         *	DEFAULT: NULL
         */

        string m_lipiRootPath;
        /**< @brief Path of the Lipi Root
         *	<p>
         *	DEFAULT: LipiEngine::getLipiPath()
         *	</p>
         */

        LTKShapeFeatureExtractor *m_ptrFeatureExtractor;
        /**< @brief Pointer to LTKShapeFeatureExtractor class
         *	<p>
         *	DEFAULT: NULL
         *	</p>
         */

        string m_preProcSeqn;
        /**< @brief Preprocessor Sequence
         *	<p>
         *	This string will holds what sequence the preprocessor methods to be executed
         *	</p>
         */

        vector<LTKShapeSample> m_prototypeSet;
        /**< @brief Prototype Set for LVQ
         *	<p>
         *	It contains the Vector of Clustered ShapeSamples
         *	</p>
         */

        LTKCaptureDevice m_captureDevice;

        struct NeighborInfo
        {
            int classId;
            float distance;
            int prototypeSetIndex;
        };

        /*
           struct MapModFunc
           {
           string moduleName;
           string funcName;
           };
         */

        vector<stringStringPair> m_preprocSequence;

        intIntMap m_shapeIDNumPrototypesMap;
        /**< @brief Map of shapeID and Number of Samples per shape
         *	<p>
         *
         *	</p>
         */

        int m_prototypeSetModifyCount;
        /**< @brief
         *	<p>
         *	Used to count number of modifications done to m_prototypeSet.
         *   Write to MDT after m_prototypeModifyCntCFG such modifications or at Exit.
         *	</p>
         */

        int m_MDTUpdateFreq;
        /**< @brief Update MDT after a specified number of modifications to m_prototypeSet
         *	<p>
         *	Specified in NN.cfg
         *
         *	</p>
         */

        //Cache Parameters used by Adapt
        vector<LTKShapeRecoResult> m_vecRecoResult;
        /**< @brief Store Recognize results 
         *	 used by subsequent call to Adapt
         *	<p>
         *	
         *
         *	</p>
         */


        vector <struct NeighborInfo> m_neighborInfoVec;
        /**< @brief Vector to store the distances of test sample to each of the samples in prototypeSet,
         *    classIDs and indices within the prototypeset
         *   Used during subsequent call to Adapt	
         *	<p>
         *	
         *
         *	</p>
         */

        LTKShapeSample m_cachedShapeSampleFeatures;
        /**< @brief Store ShapeSampleFeatures of the last inTraceGroup to Recognize 
         *   Used during subsequent call to Adapt
         *	<p>
         *	
         *
         *	</p>
         */

        float m_rejectThreshold;
        /**< @brief Threshold on the confidence to reject a test sample
         *	<p>
         *
         *	</p>
         */

        bool m_adaptivekNN;
        /**< @brief Adaptive kNN method to compute confidence
         *	<p>
         *	If m_adaptivekNN = true, the adaptive kNN method is used for confidence computation
         * 			   false, NN or kNN method is used, based on the value of m_nearestNeighbors
         *	</p>
         */

        //@}

        string m_currentVersion;