mftraining.cpp File Reference

#include <string.h>
#include <stdio.h>
#include <math.h>
#include "classify.h"
#include "cluster.h"
#include "clusttool.h"
#include "commontraining.h"
#include "danerror.h"
#include "efio.h"
#include "emalloc.h"
#include "featdefs.h"
#include "fontinfo.h"
#include "genericvector.h"
#include "indexmapbidi.h"
#include "intproto.h"
#include "mastertrainer.h"
#include "mergenf.h"
#include "mf.h"
#include "ndminx.h"
#include "ocrfeatures.h"
#include "oldlist.h"
#include "protos.h"
#include "shapetable.h"
#include "tessopt.h"
#include "tprintf.h"
#include "unicity_table.h"

Go to the source code of this file.

Macros
#define	_USE_MATH_DEFINES
#define	PROGRAM_FEATURE_TYPE   "mf"

Functions
	DECLARE_STRING_PARAM_FLAG (test_ch)
int	main (int argc, char **argv)

Variables
const int	kMaxShapeLabelLength = 10

Macro Definition Documentation

#define _USE_MATH_DEFINES

Definition at line 35 of file mftraining.cpp.

#define PROGRAM_FEATURE_TYPE   "mf"

Definition at line 72 of file mftraining.cpp.

Function Documentation

DECLARE_STRING_PARAM_FLAG

(

test_ch

)

int main	(	int	argc,
		char **	argv
	)

This program reads in a text file consisting of feature samples from a training page in the following format:

   FontName UTF8-char-str xmin ymin xmax ymax page-number
    NumberOfFeatureTypes(N)
      FeatureTypeName1 NumberOfFeatures(M)
         Feature1
         ...
         FeatureM
      FeatureTypeName2 NumberOfFeatures(M)
         Feature1
         ...
         FeatureM
      ...
      FeatureTypeNameN NumberOfFeatures(M)
         Feature1
         ...
         FeatureM
   FontName CharName ...

The result of this program is a binary inttemp file used by the OCR engine.

Parameters

argc	number of command line arguments
argv	array of command line arguments

Returns

none

Note

Exceptions: none

History: Fri Aug 18 08:56:17 1989, DSJ, Created.

History: Mon May 18 1998, Christy Russson, Revistion started.

Definition at line 388 of file tesseractmain.cpp.

                                {
  const char* lang = "eng";
  const char* image = NULL;
  const char* outputbase = NULL;
  const char* datapath = NULL;
  bool list_langs = false;
  bool print_parameters = false;
  int arg_i = 1;
  tesseract::PageSegMode pagesegmode = tesseract::PSM_AUTO;
  tesseract::OcrEngineMode enginemode = tesseract::OEM_DEFAULT;
  /* main() calls functions like ParseArgs which call exit().
   * This results in memory leaks if vars_vec and vars_values are
   * declared as auto variables (destructor is not called then). */
  static GenericVector<STRING> vars_vec;
  static GenericVector<STRING> vars_values;

#ifdef NDEBUG
  // Disable debugging and informational messages from Leptonica.
  setMsgSeverity(L_SEVERITY_ERROR);
#endif

#if defined(HAVE_TIFFIO_H) && defined(_WIN32)
  /* Show libtiff warnings on console (not in GUI). */
  TIFFSetWarningHandler(Win32WarningHandler);
#endif /* HAVE_TIFFIO_H &&  _WIN32 */

  ParseArgs(argc, argv, &lang, &image, &outputbase, &datapath, &list_langs,
            &print_parameters, &vars_vec, &vars_values, &arg_i, &pagesegmode,
            &enginemode);

  bool banner = false;
  if (outputbase != NULL && strcmp(outputbase, "-") &&
      strcmp(outputbase, "stdout")) {
    banner = true;
  }

  PERF_COUNT_START("Tesseract:main")

  // Call GlobalDawgCache here to create the global DawgCache object before
  // the TessBaseAPI object. This fixes the order of destructor calls:
  // first TessBaseAPI must be destructed, DawgCache must be the last object.
  tesseract::Dict::GlobalDawgCache();

  // Avoid memory leak caused by auto variable when exit() is called.
  static tesseract::TessBaseAPI api;

  api.SetOutputName(outputbase);

  int init_failed = api.Init(datapath, lang, enginemode, &(argv[arg_i]),
                             argc - arg_i, &vars_vec, &vars_values, false);
  if (init_failed) {
    fprintf(stderr, "Could not initialize tesseract.\n");
    return EXIT_FAILURE;
  }

  SetVariablesFromCLArgs(&api, argc, argv);

  if (list_langs) {
    PrintLangsList(&api);
    return EXIT_SUCCESS;
  }

  if (print_parameters) {
     FILE* fout = stdout;
     fprintf(stdout, "Tesseract parameters:\n");
     api.PrintVariables(fout);
     api.End();
     return EXIT_SUCCESS;
  }

  FixPageSegMode(&api, pagesegmode);

  if (pagesegmode == tesseract::PSM_AUTO_ONLY) {
    int ret_val = EXIT_SUCCESS;

    Pix* pixs = pixRead(image);
    if (!pixs) {
      fprintf(stderr, "Cannot open input file: %s\n", image);
      return 2;
    }

    api.SetImage(pixs);

    tesseract::Orientation orientation;
    tesseract::WritingDirection direction;
    tesseract::TextlineOrder order;
    float deskew_angle;

    tesseract::PageIterator* it = api.AnalyseLayout();
    if (it) {
      it->Orientation(&orientation, &direction, &order, &deskew_angle);
      tprintf(
          "Orientation: %d\nWritingDirection: %d\nTextlineOrder: %d\n"
          "Deskew angle: %.4f\n",
          orientation, direction, order, deskew_angle);
    } else {
      ret_val = EXIT_FAILURE;
    }

    delete it;

    pixDestroy(&pixs);
    return ret_val;
  }

  // set in_training_mode to true when using one of these configs:
  // ambigs.train, box.train, box.train.stderr, linebox, rebox
  bool b = false;
  bool in_training_mode =
      (api.GetBoolVariable("tessedit_ambigs_training", &b) && b) ||
      (api.GetBoolVariable("tessedit_resegment_from_boxes", &b) && b) ||
      (api.GetBoolVariable("tessedit_make_boxes_from_boxes", &b) && b);

  // Avoid memory leak caused by auto variable when exit() is called.
  static tesseract::PointerVector<tesseract::TessResultRenderer> renderers;

  if (in_training_mode) {
    renderers.push_back(NULL);
  } else {
    PreloadRenderers(&api, &renderers, pagesegmode, outputbase);
  }

  if (!renderers.empty()) {
    if (banner) PrintBanner();
    bool succeed = api.ProcessPages(image, NULL, 0, renderers[0]);
    if (!succeed) {
      fprintf(stderr, "Error during processing.\n");
      return EXIT_FAILURE;
    }
  }

  PERF_COUNT_END

  return EXIT_SUCCESS;
}

Variable Documentation

const int kMaxShapeLabelLength = 10

Definition at line 75 of file mftraining.cpp.