#include "cluster.h"
#include "commandlineflags.h"
#include "featdefs.h"
#include "intproto.h"
#include "oldlist.h"

Classes
struct	LABELEDLISTNODE
struct	MERGE_CLASS_NODE
Namespaces
namespace	tesseract
Typedefs
typedef struct LABELEDLISTNODE *	LABELEDLIST
typedef MERGE_CLASS_NODE *	MERGE_CLASS
Functions
void	ParseArguments (int argc, char **argv)
ShapeTable *	tesseract::LoadShapeTable (const STRING &file_prefix)
void	tesseract::WriteShapeTable (const STRING &file_prefix, const ShapeTable &shape_table)
MasterTrainer *	tesseract::LoadTrainingData (int argc, const char const argv, bool replication, ShapeTable *shape_table, STRING file_prefix)
const char *	GetNextFilename (int argc, const char const argv)
LABELEDLIST	FindList (LIST List, char *Label)
LABELEDLIST	NewLabeledList (const char *Label)
void	ReadTrainingSamples (const FEATURE_DEFS_STRUCT &feature_defs, const char feature_name, int max_samples, UNICHARSET unicharset, FILE file, LIST training_samples)
void	WriteTrainingSamples (const FEATURE_DEFS_STRUCT &FeatureDefs, char Directory, LIST CharList, const char program_feature_type)
void	FreeTrainingSamples (LIST CharList)
void	FreeLabeledList (LABELEDLIST LabeledList)
void	FreeLabeledClassList (LIST ClassListList)
CLUSTERER *	SetUpForClustering (const FEATURE_DEFS_STRUCT &FeatureDefs, LABELEDLIST CharSample, const char *program_feature_type)
LIST	RemoveInsignificantProtos (LIST ProtoList, BOOL8 KeepSigProtos, BOOL8 KeepInsigProtos, int N)
void	CleanUpUnusedData (LIST ProtoList)
void	MergeInsignificantProtos (LIST ProtoList, const char label, CLUSTERER Clusterer, CLUSTERCONFIG *Config)
MERGE_CLASS	FindClass (LIST List, const char *Label)
MERGE_CLASS	NewLabeledClass (const char *Label)
CLASS_STRUCT *	SetUpForFloat2Int (const UNICHARSET &unicharset, LIST LabeledClassList)
void	Normalize (float *Values)
void	FreeNormProtoList (LIST CharList)
void	AddToNormProtosList (LIST NormProtoList, LIST ProtoList, char CharName)
int	NumberOfProtos (LIST ProtoList, BOOL8 CountSigProtos, BOOL8 CountInsigProtos)
void	allocNormProtos ()
Variables
FEATURE_DEFS_STRUCT	feature_defs
CLUSTERCONFIG	Config

Typedef Documentation

typedef struct LABELEDLISTNODE * LABELEDLIST

typedef MERGE_CLASS_NODE* MERGE_CLASS

Definition at line 56 of file commontraining.h.

Function Documentation

void AddToNormProtosList	(	LIST *	NormProtoList,
		LIST	ProtoList,
		char *	CharName
	)

Definition at line 870 of file commontraining.cpp.

{
  PROTOTYPE* Proto;
  LABELEDLIST LabeledProtoList;

  LabeledProtoList = NewLabeledList(CharName);
  iterate(ProtoList)
  {
    Proto = (PROTOTYPE *) first_node (ProtoList);
    LabeledProtoList->List = push(LabeledProtoList->List, Proto);
  }
  *NormProtoList = push(*NormProtoList, LabeledProtoList);
}

void allocNormProtos ( )

void CleanUpUnusedData ( LIST ProtoList )

Definition at line 618 of file commontraining.cpp.

{
  PROTOTYPE* Prototype;

  iterate(ProtoList)
  {
    Prototype = (PROTOTYPE *) first_node (ProtoList);
    if(Prototype->Variance.Elliptical != NULL)
    {
      memfree(Prototype->Variance.Elliptical);
      Prototype->Variance.Elliptical = NULL;
    }
    if(Prototype->Magnitude.Elliptical != NULL)
    {
      memfree(Prototype->Magnitude.Elliptical);
      Prototype->Magnitude.Elliptical = NULL;
    }
    if(Prototype->Weight.Elliptical != NULL)
    {
      memfree(Prototype->Weight.Elliptical);
      Prototype->Weight.Elliptical = NULL;
    }
  }
}

MERGE_CLASS FindClass	(	LIST	List,
		const char *	Label
	)

Definition at line 713 of file commontraining.cpp.

{
  MERGE_CLASS   MergeClass;

  iterate (List)
  {
    MergeClass = (MERGE_CLASS) first_node (List);
    if (strcmp (MergeClass->Label, Label) == 0)
      return (MergeClass);
  }
  return (NULL);

}       /* FindClass */

LABELEDLIST FindList	(	LIST	List,
		char *	Label
	)

Definition at line 320 of file commontraining.cpp.

{
  LABELEDLIST   LabeledList;

  iterate (List)
  {
    LabeledList = (LABELEDLIST) first_node (List);
    if (strcmp (LabeledList->Label, Label) == 0)
      return (LabeledList);
  }
  return (NULL);

}       /* FindList */

void FreeLabeledClassList ( LIST ClassListList )

Definition at line 744 of file commontraining.cpp.

{
  MERGE_CLASS   MergeClass;

  iterate (ClassList)           /* iterate thru all of the fonts */
  {
    MergeClass = (MERGE_CLASS) first_node (ClassList);
    free (MergeClass->Label);
    FreeClass(MergeClass->Class);
    delete MergeClass;
  }
  destroy (ClassList);

}       /* FreeLabeledClassList */

void FreeLabeledList ( LABELEDLIST LabeledList )

Definition at line 483 of file commontraining.cpp.

                                              {
/*
 **     Parameters:
 **             LabeledList     labeled list to be freed
 **     Globals: none
 **     Operation:
 **             This routine deallocates all of the memory consumed by
 **             a labeled list.  It does not free any memory which may be
 **             consumed by the items in the list.
 **     Return: none
 **     Exceptions: none
 **     History: Fri Aug 18 17:52:45 1989, DSJ, Created.
 */
  destroy(LabeledList->List);
  free(LabeledList->Label);
  free(LabeledList);
}  /* FreeLabeledList */

void FreeNormProtoList ( LIST CharList )

Definition at line 854 of file commontraining.cpp.

{
  LABELEDLIST   char_sample;

  iterate (CharList)            /* iterate thru all of the fonts */
  {
    char_sample = (LABELEDLIST) first_node (CharList);
    FreeLabeledList (char_sample);
  }
  destroy (CharList);

}       // FreeNormProtoList

void FreeTrainingSamples ( LIST CharList )

Definition at line 453 of file commontraining.cpp.

                                        {
/*
 **     Parameters:
 **             FontList        list of all fonts in document
 **     Globals: none
 **     Operation:
 **             This routine deallocates all of the space allocated to
 **             the specified list of training samples.
 **     Return: none
 **     Exceptions: none
 **     History: Fri Aug 18 17:44:27 1989, DSJ, Created.
 */
  LABELEDLIST char_sample;
  FEATURE_SET FeatureSet;
  LIST FeatureList;


  iterate(CharList) {  /* iterate thru all of the fonts */
    char_sample = (LABELEDLIST) first_node(CharList);
    FeatureList = char_sample->List;
    iterate(FeatureList) {  /* iterate thru all of the classes */
      FeatureSet = (FEATURE_SET) first_node(FeatureList);
      FreeFeatureSet(FeatureSet);
    }
    FreeLabeledList(char_sample);
  }
  destroy(CharList);
}  /* FreeTrainingSamples */

const char* GetNextFilename	(	int	argc,
		const char const	argv
	)

Definition at line 297 of file commontraining.cpp.

                                                                {
  /*
   **   Parameters: none
   **   Globals:
   **           tessoptind                      defined by tessopt sys call
   **   Operation:
   **           This routine returns the next command line argument.  If
   **           there are no remaining command line arguments, it returns
   **           NULL.  This routine should only be called after all option
   **           arguments have been parsed and removed with ParseArguments.
   **   Return: Next command line argument or NULL.
   **   Exceptions: none
   **   History: Fri Aug 18 09:34:12 1989, DSJ, Created.
   */
  if (tessoptind < argc)
    return argv[tessoptind++];
  else
    return NULL;
}       /* GetNextFilename */

void MergeInsignificantProtos	(	LIST	ProtoList,
		const char *	label,
		CLUSTERER *	Clusterer,
		CLUSTERCONFIG *	Config
	)

Definition at line 553 of file commontraining.cpp.

                                                                           {
  PROTOTYPE     *Prototype;
  bool debug = strcmp(FLAGS_test_ch.c_str(), label) == 0;

  LIST pProtoList = ProtoList;
  iterate(pProtoList) {
    Prototype = (PROTOTYPE *) first_node (pProtoList);
    if (Prototype->Significant || Prototype->Merged)
      continue;
    FLOAT32 best_dist = 0.125;
    PROTOTYPE* best_match = NULL;
    // Find the nearest alive prototype.
    LIST list_it = ProtoList;
    iterate(list_it) {
      PROTOTYPE* test_p = (PROTOTYPE *) first_node (list_it);
      if (test_p != Prototype && !test_p->Merged) {
        FLOAT32 dist = ComputeDistance(Clusterer->SampleSize,
                                       Clusterer->ParamDesc,
                                       Prototype->Mean, test_p->Mean);
        if (dist < best_dist) {
          best_match = test_p;
          best_dist = dist;
        }
      }
    }
    if (best_match != NULL && !best_match->Significant) {
      if (debug)
        tprintf("Merging red clusters (%d+%d) at %g,%g and %g,%g\n",
                best_match->NumSamples, Prototype->NumSamples,
                best_match->Mean[0], best_match->Mean[1],
                Prototype->Mean[0], Prototype->Mean[1]);
      best_match->NumSamples = MergeClusters(Clusterer->SampleSize,
                                             Clusterer->ParamDesc,
                                             best_match->NumSamples,
                                             Prototype->NumSamples,
                                             best_match->Mean,
                                             best_match->Mean, Prototype->Mean);
      Prototype->NumSamples = 0;
      Prototype->Merged = 1;
    } else if (best_match != NULL) {
      if (debug)
        tprintf("Red proto at %g,%g matched a green one at %g,%g\n",
                Prototype->Mean[0], Prototype->Mean[1],
                best_match->Mean[0], best_match->Mean[1]);
      Prototype->Merged = 1;
    }
  }
  // Mark significant those that now have enough samples.
  int min_samples = (inT32) (Config->MinSamples * Clusterer->NumChar);
  pProtoList = ProtoList;
  iterate(pProtoList) {
    Prototype = (PROTOTYPE *) first_node (pProtoList);
    // Process insignificant protos that do not match a green one
    if (!Prototype->Significant && Prototype->NumSamples >= min_samples &&
        !Prototype->Merged) {
      if (debug)
        tprintf("Red proto at %g,%g becoming green\n",
                Prototype->Mean[0], Prototype->Mean[1]);
      Prototype->Significant = true;
    }
  }
}       /* MergeInsignificantProtos */

MERGE_CLASS NewLabeledClass ( const char * Label )

Definition at line 730 of file commontraining.cpp.

{
  MERGE_CLASS   MergeClass;

  MergeClass = new MERGE_CLASS_NODE;
  MergeClass->Label = (char*)Emalloc (strlen (Label)+1);
  strcpy (MergeClass->Label, Label);
  MergeClass->Class = NewClass (MAX_NUM_PROTOS, MAX_NUM_CONFIGS);
  return (MergeClass);

}       /* NewLabeledClass */

LABELEDLIST NewLabeledList ( const char * Label )

Definition at line 352 of file commontraining.cpp.

{
  LABELEDLIST   LabeledList;

  LabeledList = (LABELEDLIST) Emalloc (sizeof (LABELEDLISTNODE));
  LabeledList->Label = (char*)Emalloc (strlen (Label)+1);
  strcpy (LabeledList->Label, Label);
  LabeledList->List = NIL_LIST;
  LabeledList->SampleCount = 0;
  LabeledList->font_sample_count = 0;
  return (LabeledList);

}       /* NewLabeledList */

void Normalize ( float * Values )

Definition at line 837 of file commontraining.cpp.

{
  register float Slope;
  register float Intercept;
  register float Normalizer;

  Slope      = tan (Values [2] * 2 * PI);
  Intercept  = Values [1] - Slope * Values [0];
  Normalizer = 1 / sqrt (Slope * Slope + 1.0);

  Values [0] = Slope * Normalizer;
  Values [1] = - Normalizer;
  Values [2] = Intercept * Normalizer;
} // Normalize

int NumberOfProtos	(	LIST	ProtoList,
		BOOL8	CountSigProtos,
		BOOL8	CountInsigProtos
	)

Definition at line 888 of file commontraining.cpp.

{
  int N = 0;
  PROTOTYPE     *Proto;

  iterate(ProtoList)
  {
    Proto = (PROTOTYPE *) first_node ( ProtoList );
    if (( Proto->Significant && CountSigProtos )        ||
        ( ! Proto->Significant && CountInsigProtos ) )
      N++;
  }
  return(N);
}

void ParseArguments	(	int *	argc,
		char ***	argv
	)

Definition at line 89 of file commontraining.cpp.

                                             {
  STRING usage;
  if (*argc) {
    usage += (*argv)[0];
  }
  usage += " [.tr files ...]";
  tesseract::ParseCommandLineFlags(usage.c_str(), argc, argv, true);
  // Record the index of the first non-flag argument to 1, since we set
  // remove_flags to true when parsing the flags.
  tessoptind = 1;
  // Set some global values based on the flags.
  Config.MinSamples =
      MAX(0.0, MIN(1.0, double(FLAGS_clusterconfig_min_samples_fraction)));
  Config.MaxIllegal =
      MAX(0.0, MIN(1.0, double(FLAGS_clusterconfig_max_illegal)));
  Config.Independence =
      MAX(0.0, MIN(1.0, double(FLAGS_clusterconfig_independence)));
  Config.Confidence =
      MAX(0.0, MIN(1.0, double(FLAGS_clusterconfig_confidence)));
  // Set additional parameters from config file if specified.
  if (!FLAGS_configfile.empty()) {
    tesseract::ParamUtils::ReadParamsFile(
        FLAGS_configfile.c_str(),
        tesseract::SET_PARAM_CONSTRAINT_NON_INIT_ONLY,
        ccutil.params());
  }
}

void ReadTrainingSamples	(	const FEATURE_DEFS_STRUCT &	feature_defs,
		const char *	feature_name,
		int	max_samples,
		UNICHARSET *	unicharset,
		FILE *	file,
		LIST *	training_samples
	)

Definition at line 383 of file commontraining.cpp.

                                                             {
/*
**  Parameters:
**    file    open text file to read samples from
**  Globals: none
**  Operation:
**    This routine reads training samples from a file and
**    places them into a data structure which organizes the
**    samples by FontName and CharName.  It then returns this
**    data structure.
**  Return: none
**  Exceptions: none
**  History: Fri Aug 18 13:11:39 1989, DSJ, Created.
**       Tue May 17 1998 simplifications to structure, illiminated
**        font, and feature specification levels of structure.
*/
  char    buffer[2048];
  char    unichar[UNICHAR_LEN + 1];
  LABELEDLIST char_sample;
  FEATURE_SET feature_samples;
  CHAR_DESC char_desc;
  int   i;
  int feature_type = ShortNameToFeatureType(feature_defs, feature_name);
  // Zero out the font_sample_count for all the classes.
  LIST it = *training_samples;
  iterate(it) {
    char_sample = reinterpret_cast<LABELEDLIST>(first_node(it));
    char_sample->font_sample_count = 0;
  }

  while (fgets(buffer, 2048, file) != NULL) {
    if (buffer[0] == '\n')
      continue;

    sscanf(buffer, "%*s %s", unichar);
    if (unicharset != NULL && !unicharset->contains_unichar(unichar)) {
      unicharset->unichar_insert(unichar);
      if (unicharset->size() > MAX_NUM_CLASSES) {
        tprintf("Error: Size of unicharset in training is "
                "greater than MAX_NUM_CLASSES\n");
        exit(1);
      }
    }
    char_sample = FindList(*training_samples, unichar);
    if (char_sample == NULL) {
      char_sample = NewLabeledList(unichar);
      *training_samples = push(*training_samples, char_sample);
    }
    char_desc = ReadCharDescription(feature_defs, file);
    feature_samples = char_desc->FeatureSets[feature_type];
    if (char_sample->font_sample_count < max_samples || max_samples <= 0) {
      char_sample->List = push(char_sample->List, feature_samples);
      char_sample->SampleCount++;
      char_sample->font_sample_count++;
    } else {
      FreeFeatureSet(feature_samples);
    }
    for (i = 0; i < char_desc->NumFeatureSets; i++) {
      if (feature_type != i)
        FreeFeatureSet(char_desc->FeatureSets[i]);
    }
    free(char_desc);
  }
}  // ReadTrainingSamples

LIST RemoveInsignificantProtos	(	LIST	ProtoList,
		BOOL8	KeepSigProtos,
		BOOL8	KeepInsigProtos,
		int	N
	)

Definition at line 645 of file commontraining.cpp.

{
  LIST NewProtoList = NIL_LIST;
  LIST pProtoList;
  PROTOTYPE* Proto;
  PROTOTYPE* NewProto;
  int i;

  pProtoList = ProtoList;
  iterate(pProtoList)
  {
    Proto = (PROTOTYPE *) first_node (pProtoList);
    if ((Proto->Significant && KeepSigProtos) ||
        (!Proto->Significant && KeepInsigProtos))
    {
      NewProto = (PROTOTYPE *)Emalloc(sizeof(PROTOTYPE));

      NewProto->Mean = (FLOAT32 *)Emalloc(N * sizeof(FLOAT32));
      NewProto->Significant = Proto->Significant;
      NewProto->Style = Proto->Style;
      NewProto->NumSamples = Proto->NumSamples;
      NewProto->Cluster = NULL;
      NewProto->Distrib = NULL;

      for (i=0; i < N; i++)
        NewProto->Mean[i] = Proto->Mean[i];
      if (Proto->Variance.Elliptical != NULL)
      {
        NewProto->Variance.Elliptical = (FLOAT32 *)Emalloc(N * sizeof(FLOAT32));
        for (i=0; i < N; i++)
          NewProto->Variance.Elliptical[i] = Proto->Variance.Elliptical[i];
      }
      else
        NewProto->Variance.Elliptical = NULL;
      //---------------------------------------------
      if (Proto->Magnitude.Elliptical != NULL)
      {
        NewProto->Magnitude.Elliptical = (FLOAT32 *)Emalloc(N * sizeof(FLOAT32));
        for (i=0; i < N; i++)
          NewProto->Magnitude.Elliptical[i] = Proto->Magnitude.Elliptical[i];
      }
      else
        NewProto->Magnitude.Elliptical = NULL;
      //------------------------------------------------
      if (Proto->Weight.Elliptical != NULL)
      {
        NewProto->Weight.Elliptical = (FLOAT32 *)Emalloc(N * sizeof(FLOAT32));
        for (i=0; i < N; i++)
          NewProto->Weight.Elliptical[i] = Proto->Weight.Elliptical[i];
      }
      else
        NewProto->Weight.Elliptical = NULL;

      NewProto->TotalMagnitude = Proto->TotalMagnitude;
      NewProto->LogMagnitude = Proto->LogMagnitude;
      NewProtoList = push_last(NewProtoList, NewProto);
    }
  }
  FreeProtoList(&ProtoList);
  return (NewProtoList);
}       /* RemoveInsignificantProtos */

CLUSTERER* SetUpForClustering	(	const FEATURE_DEFS_STRUCT &	FeatureDefs,
		LABELEDLIST	CharSample,
		const char *	program_feature_type
	)

Definition at line 502 of file commontraining.cpp.

                                                                {
/*
 **     Parameters:
 **             char_sample: LABELEDLIST that holds all the feature information for a
 **             given character.
 **     Globals:
 **             None
 **     Operation:
 **             This routine reads samples from a LABELEDLIST and enters
 **             those samples into a clusterer data structure.  This
 **             data structure is then returned to the caller.
 **     Return:
 **             Pointer to new clusterer data structure.
 **     Exceptions:
 **             None
 **     History:
 **             8/16/89, DSJ, Created.
 */
  uinT16 N;
  int i, j;
  FLOAT32 *Sample = NULL;
  CLUSTERER *Clusterer;
  inT32 CharID;
  LIST FeatureList = NULL;
  FEATURE_SET FeatureSet = NULL;

  int desc_index = ShortNameToFeatureType(FeatureDefs, program_feature_type);
  N = FeatureDefs.FeatureDesc[desc_index]->NumParams;
  Clusterer = MakeClusterer(N, FeatureDefs.FeatureDesc[desc_index]->ParamDesc);

  FeatureList = char_sample->List;
  CharID = 0;
  iterate(FeatureList) {
    FeatureSet = (FEATURE_SET) first_node(FeatureList);
    for (i = 0; i < FeatureSet->MaxNumFeatures; i++) {
      if (Sample == NULL)
        Sample = (FLOAT32 *)Emalloc(N * sizeof(FLOAT32));
      for (j = 0; j < N; j++)
        Sample[j] = FeatureSet->Features[i]->Params[j];
      MakeSample (Clusterer, Sample, CharID);
    }
    CharID++;
  }
  if ( Sample != NULL ) free( Sample );
  return( Clusterer );

}       /* SetUpForClustering */

CLASS_STRUCT* SetUpForFloat2Int	(	const UNICHARSET &	unicharset,
		LIST	LabeledClassList
	)

SetUpForFloat2Int

Definition at line 774 of file commontraining.cpp.

                                                       {
  MERGE_CLASS   MergeClass;
  CLASS_TYPE            Class;
  int                           NumProtos;
  int                           NumConfigs;
  int                           NumWords;
  int                           i, j;
  float                 Values[3];
  PROTO                 NewProto;
  PROTO                 OldProto;
  BIT_VECTOR            NewConfig;
  BIT_VECTOR            OldConfig;

  //    printf("Float2Int ...\n");

  CLASS_STRUCT* float_classes = new CLASS_STRUCT[unicharset.size()];
  iterate(LabeledClassList)
  {
    UnicityTableEqEq<int>   font_set;
    MergeClass = (MERGE_CLASS) first_node (LabeledClassList);
    Class = &float_classes[unicharset.unichar_to_id(MergeClass->Label)];
    NumProtos = MergeClass->Class->NumProtos;
    NumConfigs = MergeClass->Class->NumConfigs;
    font_set.move(&MergeClass->Class->font_set);
    Class->NumProtos = NumProtos;
    Class->MaxNumProtos = NumProtos;
    Class->Prototypes = (PROTO) Emalloc (sizeof(PROTO_STRUCT) * NumProtos);
    for(i=0; i < NumProtos; i++)
    {
      NewProto = ProtoIn(Class, i);
      OldProto = ProtoIn(MergeClass->Class, i);
      Values[0] = OldProto->X;
      Values[1] = OldProto->Y;
      Values[2] = OldProto->Angle;
      Normalize(Values);
      NewProto->X = OldProto->X;
      NewProto->Y = OldProto->Y;
      NewProto->Length = OldProto->Length;
      NewProto->Angle = OldProto->Angle;
      NewProto->A = Values[0];
      NewProto->B = Values[1];
      NewProto->C = Values[2];
    }

    Class->NumConfigs = NumConfigs;
    Class->MaxNumConfigs = NumConfigs;
    Class->font_set.move(&font_set);
    Class->Configurations = (BIT_VECTOR*) Emalloc (sizeof(BIT_VECTOR) * NumConfigs);
    NumWords = WordsInVectorOfSize(NumProtos);
    for(i=0; i < NumConfigs; i++)
    {
      NewConfig = NewBitVector(NumProtos);
      OldConfig = MergeClass->Class->Configurations[i];
      for(j=0; j < NumWords; j++)
        NewConfig[j] = OldConfig[j];
      Class->Configurations[i] = NewConfig;
    }
  }
  return float_classes;
} // SetUpForFloat2Int

void WriteTrainingSamples	(	const FEATURE_DEFS_STRUCT &	FeatureDefs,
		char *	Directory,
		LIST	CharList,
		const char *	program_feature_type
	)

Variable Documentation

CLUSTERCONFIG Config

Definition at line 51 of file commontraining.cpp.

FEATURE_DEFS_STRUCT feature_defs

Definition at line 52 of file commontraining.cpp.

Classes

Namespaces

Typedefs

Functions

Variables

Typedef Documentation

Function Documentation

Variable Documentation