mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2025-01-21 17:13:09 +08:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

Replace malloc / free by new / delete for CHAR_DESC_STRUCT

Browse Source 
Signed-off-by: Stefan Weil <sw@weilnetz.de>
This commit is contained in:
Stefan Weil 2021-03-27 18:43:14 +01:00
parent 0f90ccb9cd
commit aa8dda89a3
5 changed files with 35 additions and 68 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
src/classify/blobclass.cpp
View File

@ -34,13 +34,13 @@ namespace tesseract {
// blob_text: Ground truth text for the blob.
void Classify::LearnBlob(const std::string &fontname, TBLOB *blob, const DENORM &cn_denorm,
const INT_FX_RESULT_STRUCT &fx_info, const char *blob_text) {
CHAR_DESC CharDesc = NewCharDescription(feature_defs_);
std::unique_ptr<CHAR_DESC_STRUCT> CharDesc(new CHAR_DESC_STRUCT(feature_defs_));
CharDesc->FeatureSets[0] = ExtractMicros(blob, cn_denorm);
CharDesc->FeatureSets[1] = ExtractCharNormFeatures(fx_info);
CharDesc->FeatureSets[2] = ExtractIntCNFeatures(*blob, fx_info);
CharDesc->FeatureSets[3] = ExtractIntGeoFeatures(*blob, fx_info);
if (ValidCharDescription(feature_defs_, CharDesc)) {
if (ValidCharDescription(feature_defs_, CharDesc.get())) {
// Label the features with a class name and font name.
tr_file_data_ += "\n";
tr_file_data_ += fontname;
@ -49,11 +49,10 @@ void Classify::LearnBlob(const std::string &fontname, TBLOB *blob, const DENORM
tr_file_data_ += "\n";
// write micro-features to file and clean up
WriteCharDescription(feature_defs_, CharDesc, tr_file_data_);
WriteCharDescription(feature_defs_, CharDesc.get(), tr_file_data_);
} else {
tprintf("Blob learned was invalid!\n");
}
FreeCharDescription(CharDesc);
} // LearnBlob
// Writes stored training data to a .tr file based on the given filename.

50
src/classify/featdefs.cpp
View File

@ -91,47 +91,6 @@ void InitFeatureDefs(FEATURE_DEFS_STRUCT *featuredefs) {
}
}
/*---------------------------------------------------------------------------*/
/**
* Release the memory consumed by the specified character
* description and all of the features in that description.
*
* @param CharDesc character description to be deallocated
*
* Globals:
* - none
*/
void FreeCharDescription(CHAR_DESC CharDesc) {
if (CharDesc) {
for (size_t i = 0; i < CharDesc->NumFeatureSets; i++) {
FreeFeatureSet(CharDesc->FeatureSets[i]);
}
free(CharDesc);
}
} /* FreeCharDescription */
/*---------------------------------------------------------------------------*/
/**
* Allocate a new character description, initialize its
* feature sets to be empty, and return it.
*
* Globals:
* - none
*
* @return New character description structure.
*/
CHAR_DESC NewCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs) {
CHAR_DESC CharDesc;
CharDesc = static_cast<CHAR_DESC>(malloc(sizeof(CHAR_DESC_STRUCT)));
CharDesc->NumFeatureSets = FeatureDefs.NumFeatureTypes;
for (size_t i = 0; i < CharDesc->NumFeatureSets; i++) {
CharDesc->FeatureSets[i] = nullptr;
}
return (CharDesc);
} /* NewCharDescription */
/*---------------------------------------------------------------------------*/
/**
* Appends a textual representation of CharDesc to str.
@ -147,7 +106,7 @@ CHAR_DESC NewCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs) {
* @param str string to append CharDesc to
* @param CharDesc character description to write to File
*/
void WriteCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC CharDesc, std::string &str) {
void WriteCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC_STRUCT *CharDesc, std::string &str) {
int NumSetsToWrite = 0;
for (size_t Type = 0; Type < CharDesc->NumFeatureSets; Type++) {
@ -169,7 +128,7 @@ void WriteCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC Char
// Return whether all of the fields of the given feature set
// are well defined (not inf or nan).
bool ValidCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC CharDesc) {
bool ValidCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC_STRUCT *CharDesc) {
bool anything_written = false;
bool well_formed = true;
for (size_t Type = 0; Type < CharDesc->NumFeatureSets; Type++) {
@ -210,17 +169,16 @@ bool ValidCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC Char
* @param File open text file to read character description from
* @return Character description read from File.
*/
CHAR_DESC ReadCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, FILE *File) {
CHAR_DESC_STRUCT *ReadCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, FILE *File) {
int NumSetsToRead;
char ShortName[FEAT_NAME_SIZE];
CHAR_DESC CharDesc;
int Type;
ASSERT_HOST(tfscanf(File, "%d", &NumSetsToRead) == 1);
ASSERT_HOST(NumSetsToRead >= 0);
ASSERT_HOST(NumSetsToRead <= FeatureDefs.NumFeatureTypes);
CharDesc = NewCharDescription(FeatureDefs);
auto CharDesc = new CHAR_DESC_STRUCT(FeatureDefs);
for (; NumSetsToRead > 0; NumSetsToRead--) {
tfscanf(File, "%s", ShortName);
Type = ShortNameToFeatureType(FeatureDefs, ShortName);

39
src/classify/featdefs.h
View File

@ -37,35 +37,46 @@ extern TESS_API const char *const kGeoFeatureType;
feature consists of a number of parameters. All features within a
feature set contain the same number of parameters.*/
struct CHAR_DESC_STRUCT {
uint32_t NumFeatureSets;
FEATURE_SET FeatureSets[NUM_FEATURE_TYPES];
};
using CHAR_DESC = CHAR_DESC_STRUCT *;
struct FEATURE_DEFS_STRUCT {
int32_t NumFeatureTypes;
const FEATURE_DESC_STRUCT *FeatureDesc[NUM_FEATURE_TYPES];
};
using FEATURE_DEFS = FEATURE_DEFS_STRUCT *;
struct CHAR_DESC_STRUCT {
/// Allocate a new character description, initialize its
/// feature sets to be empty, and return it.
CHAR_DESC_STRUCT(const FEATURE_DEFS_STRUCT &FeatureDefs) {
NumFeatureSets = FeatureDefs.NumFeatureTypes;
for (size_t i = 0; i < NumFeatureSets; i++) {
FeatureSets[i] = nullptr;
}
}
/// Release the memory consumed by the specified character
/// description and all of the features in that description.
~CHAR_DESC_STRUCT() {
for (size_t i = 0; i < NumFeatureSets; i++) {
FreeFeatureSet(FeatureSets[i]);
}
}
uint32_t NumFeatureSets;
FEATURE_SET FeatureSets[NUM_FEATURE_TYPES];
};
/*----------------------------------------------------------------------
Generic functions for manipulating character descriptions
----------------------------------------------------------------------*/
TESS_API
void InitFeatureDefs(FEATURE_DEFS_STRUCT *featuredefs);
TESS_API
void FreeCharDescription(CHAR_DESC CharDesc);
bool ValidCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC_STRUCT *CharDesc);
CHAR_DESC NewCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs);
bool ValidCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC CharDesc);
void WriteCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC CharDesc, std::string &str);
void WriteCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, CHAR_DESC_STRUCT *CharDesc, std::string &str);
TESS_API
CHAR_DESC ReadCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, FILE *File);
CHAR_DESC_STRUCT *ReadCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs, FILE *File);
TESS_API
uint32_t ShortNameToFeatureType(const FEATURE_DEFS_STRUCT &FeatureDefs, const char *ShortName);

3
src/training/common/commontraining.cpp
View File

@ -377,7 +377,6 @@ void ReadTrainingSamples(const FEATURE_DEFS_STRUCT &feature_definitions, const c
char unichar[UNICHAR_LEN + 1];
LABELEDLIST char_sample;
FEATURE_SET feature_samples;
CHAR_DESC char_desc;
uint32_t feature_type = ShortNameToFeatureType(feature_definitions, feature_name);
// Zero out the font_sample_count for all the classes.
@ -407,7 +406,7 @@ void ReadTrainingSamples(const FEATURE_DEFS_STRUCT &feature_definitions, const c
char_sample = NewLabeledList(unichar);
*training_samples = push(*training_samples, char_sample);
}
char_desc = ReadCharDescription(feature_definitions, file);
auto char_desc = ReadCharDescription(feature_definitions, 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);

4
src/training/common/mastertrainer.cpp
View File

@ -168,7 +168,7 @@ void MasterTrainer::ReadTrainingSamples(const char *page_name,
tprintf("Bad format in tr file, reading box coords\n");
continue;
}
CHAR_DESC char_desc = ReadCharDescription(feature_defs, fp);
auto char_desc = ReadCharDescription(feature_defs, fp);
auto *sample = new TrainingSample;
sample->set_font_id(font_id);
sample->set_page_num(page_number + page_images_.size());
@ -176,7 +176,7 @@ void MasterTrainer::ReadTrainingSamples(const char *page_name,
sample->ExtractCharDesc(int_feature_type, micro_feature_type, cn_feature_type, geo_feature_type,
char_desc);
AddSample(verification, unichar.c_str(), sample);
FreeCharDescription(char_desc);
delete char_desc;
}
charsetsize_ = unicharset_.size();
fclose(fp);