Classes |
struct | TESS_CHAR |
class | TessBaseAPI |
class | CubeRecoContext |
class | CubeClassifier |
class | CubeTessClassifier |
struct | DocQualCallbacks |
class | EquationDetect |
class | LTRResultIterator |
class | ChoiceIterator |
class | MutableIterator |
class | PageIterator |
class | UnicodeSpanSkipper |
struct | Cluster |
class | SimpleClusterer |
struct | GeometricClassifierState |
struct | Interval |
class | RowInfo |
struct | LineHypothesis |
class | RowScratchRegisters |
class | ParagraphTheory |
class | ParagraphModelSmearer |
class | ResultIterator |
class | TesseractCubeCombiner |
struct | TesseractStats |
class | Tesseract |
class | ImageThresholder |
class | BoxWord |
class | CCStruct |
class | DetLineFit |
class | DPPoint |
struct | FontSpacingInfo |
struct | FontInfo |
struct | FontSet |
struct | ParamsTrainingHypothesis |
class | ParamsTrainingBundle |
class | UnicharIdArrayUtils |
class | AmbigSpec |
class | UnicharAmbigs |
class | BitVector |
class | CCUtilMutex |
class | CCUtil |
class | PointerVector |
class | IndexMap |
class | IndexMapBiDi |
struct | ParamsVectors |
class | ParamUtils |
class | Param |
class | IntParam |
class | BoolParam |
class | StringParam |
class | DoubleParam |
class | TessdataManager |
class | Classify |
class | ErrorCounter |
class | IntFeatureDist |
class | IntFeatureMap |
class | IntFeatureSpace |
class | ClassPruner |
struct | ShapeDist |
class | MasterTrainer |
class | SampleIterator |
struct | ShapeRating |
class | ShapeClassifier |
struct | UnicharAndFonts |
class | Shape |
class | ShapeTable |
class | TessClassifier |
class | TrainingSample |
class | TrainingSampleSet |
class | AltList |
class | BeamSearch |
class | Bmp8 |
class | CachedFile |
class | CharAltList |
struct | Bigram |
struct | CharBigram |
struct | CharBigramTable |
class | CharBigrams |
class | CharSamp |
class | CharSampEnum |
class | CharSampSet |
class | CharSet |
class | CharClassifier |
class | CharClassifierFactory |
class | ConCompPt |
class | ConComp |
class | ConvNetCharClassifier |
class | CubeLineObject |
class | CubeLineSegmenter |
class | CubeObject |
class | CubeSearchObject |
class | CubeTuningParams |
class | CubeUtils |
class | FeatureBase |
class | FeatureBmp |
class | FeatureChebyshev |
class | FeatureHybrid |
class | HybridNeuralNetCharClassifier |
class | LangModEdge |
class | LangModel |
class | SearchColumn |
class | SearchNode |
class | SearchNodeHashTable |
class | SearchObject |
class | TessLangModEdge |
class | TessLangModel |
class | TuningParams |
class | WordAltList |
class | WordListLangModel |
struct | PairSizeInfo |
struct | FontPairSizeInfo |
class | WordSizeModel |
class | WordUnigrams |
class | CUtil |
struct | NodeChild |
class | Dawg |
struct | DawgInfo |
class | DawgInfoVector |
class | SquishedDawg |
struct | DawgArgs |
class | Dict |
class | PermuterState |
class | Trie |
class | Image |
class | InputFileBuffer |
class | NeuralNet |
class | Neuron |
struct | AlignedBlobParams |
class | AlignedBlob |
class | GridBase |
class | IntGrid |
class | BBGrid |
class | GridSearch |
class | TabEventHandler |
class | BlobGrid |
class | CCNonTextDetect |
class | ColumnFinder |
class | ColPartition |
class | ColPartitionGrid |
class | ColPartitionSet |
class | PixelHistogram |
class | ShiroRekhaSplitter |
class | EquationDetectBase |
class | ImageFind |
class | LineFinder |
class | StrokeWidth |
class | TabFind |
class | ColSegment |
class | TableFinder |
class | StructuredTable |
class | TableRecognizer |
class | TabConstraint |
class | TabVector |
class | TextlineProjection |
class | Textord |
class | WorkingPartSet |
struct | AssociateStats |
class | AssociateUtils |
struct | LanguageModelConsistencyInfo |
struct | LanguageModelDawgInfo |
struct | LanguageModelNgramInfo |
struct | ViterbiStateEntry |
struct | LanguageModelState |
struct | BestChoiceBundle |
struct | BestPathByColumn |
class | LanguageModel |
struct | MATCH |
class | BlobMatchTable |
class | FRAGMENT |
class | Wordrec |
Typedefs |
typedef int(Dict::* | DictFunc )(void *void_dawg_args, UNICHAR_ID unichar_id, bool word_end) const |
typedef double(Dict::* | ProbabilityInContextFunc )(const char *lang, const char *context, int context_bytes, const char *character, int character_bytes) |
typedef void(Wordrec::* | FillLatticeFunc )(const MATRIX &ratings, const LIST &best_choices, const UNICHARSET &unicharset, BlamerBundle *blamer_bundle) |
typedef TessCallback3< const
UNICHARSET &, int, PAGE_RES * > | TruthCallback |
typedef GenericVectorEqEq
< const ParagraphModel * > | SetOfModels |
typedef void(Tesseract::* | WordRecognizer )(BLOCK *block, ROW *row, WERD_RES *word) |
typedef GenericVector
< ParamsTrainingHypothesis > | ParamsTrainingHypothesisList |
typedef GenericVector< UNICHAR_ID > | UnicharIdVector |
typedef GenericVector
< AmbigSpec_LIST * > | UnicharAmbigsVector |
typedef signed int | char_32 |
typedef basic_string< char_32 > | string_32 |
typedef GenericVector< NodeChild > | NodeChildVector |
typedef GenericVector< int > | SuccessorList |
typedef GenericVector
< SuccessorList * > | SuccessorListsVector |
typedef GenericVector< Dawg * > | DawgVector |
typedef GridSearch< BLOBNBOX,
BLOBNBOX_CLIST, BLOBNBOX_C_IT > | BlobGridSearch |
typedef GridSearch
< ColPartition,
ColPartition_CLIST,
ColPartition_C_IT > | ColPartitionGridSearch |
typedef GenericVector
< ColPartitionSet * > | PartSetVector |
typedef TessResultCallback1
< bool, int > | WidthCallback |
typedef BBGrid< ColSegment,
ColSegment_CLIST,
ColSegment_C_IT > | ColSegmentGrid |
typedef GridSearch< ColSegment,
ColSegment_CLIST,
ColSegment_C_IT > | ColSegmentGridSearch |
typedef unsigned char | LanguageModelFlagsType |
Enumerations |
enum | LineType { LT_START = 'S',
LT_BODY = 'C',
LT_UNKNOWN = 'U',
LT_MULTIPLE = 'M'
} |
enum | CMD_EVENTS { ACTION_1_CMD_EVENT,
RECOG_WERDS,
RECOG_PSEUDO,
ACTION_2_CMD_EVENT
} |
enum | ScriptPos { SP_NORMAL,
SP_SUBSCRIPT,
SP_SUPERSCRIPT,
SP_DROPCAP
} |
enum | NormalizationMode { NM_BASELINE = -3,
NM_CHAR_ISOTROPIC = -2,
NM_CHAR_ANISOTROPIC = -1
} |
enum | ParamsTrainingRawFeatureType {
PTRAIN_RAW_FEATURE_DICT_MATCH_TYPE,
PTRAIN_RAW_FEATURE_UNAMBIG_DICT_MATCH,
PTRAIN_RAW_FEATURE_SHAPE_COST,
PTRAIN_RAW_FEATURE_NGRAM_PROB,
PTRAIN_RAW_FEATURE_NUM_BAD_PUNC,
PTRAIN_RAW_FEATURE_NUM_BAD_CASE,
PTRAIN_RAW_FEATURE_NUM_BAD_CHAR_TYPE,
PTRAIN_RAW_FEATURE_NUM_BAD_SPACING,
PTRAIN_RAW_FEATURE_NUM_BAD_SCRIPT,
PTRAIN_RAW_FEATURE_NUM_BAD_FONT,
PTRAIN_RAW_FEATURE_WORST_CERT,
PTRAIN_RAW_FEATURE_RATING,
PTRAIN_RAW_FEATURE_ADAPTED,
PTRAIN_RAW_FEATURE_NUM_UNICHARS,
PTRAIN_RAW_FEATURE_OUTLINE_LEN,
PTRAIN_NUM_RAW_FEATURE_TYPES
} |
enum | Orientation { ORIENTATION_PAGE_UP = 0,
ORIENTATION_PAGE_RIGHT = 1,
ORIENTATION_PAGE_DOWN = 2,
ORIENTATION_PAGE_LEFT = 3
} |
enum | WritingDirection { WRITING_DIRECTION_LEFT_TO_RIGHT = 0,
WRITING_DIRECTION_RIGHT_TO_LEFT = 1,
WRITING_DIRECTION_TOP_TO_BOTTOM = 2
} |
enum | TextlineOrder { TEXTLINE_ORDER_LEFT_TO_RIGHT = 0,
TEXTLINE_ORDER_RIGHT_TO_LEFT = 1,
TEXTLINE_ORDER_TOP_TO_BOTTOM = 2
} |
enum | PageSegMode {
PSM_OSD_ONLY,
PSM_AUTO_OSD,
PSM_AUTO_ONLY,
PSM_AUTO,
PSM_SINGLE_COLUMN,
PSM_SINGLE_BLOCK_VERT_TEXT,
PSM_SINGLE_BLOCK,
PSM_SINGLE_LINE,
PSM_SINGLE_WORD,
PSM_CIRCLE_WORD,
PSM_SINGLE_CHAR,
PSM_COUNT
} |
enum | PageIteratorLevel {
RIL_BLOCK,
RIL_PARA,
RIL_TEXTLINE,
RIL_WORD,
RIL_SYMBOL
} |
enum | ParagraphJustification { JUSTIFICATION_UNKNOWN,
JUSTIFICATION_LEFT,
JUSTIFICATION_CENTER,
JUSTIFICATION_RIGHT
} |
enum | OcrEngineMode { OEM_TESSERACT_ONLY,
OEM_CUBE_ONLY,
OEM_TESSERACT_CUBE_COMBINED,
OEM_DEFAULT
} |
enum | AmbigType {
NOT_AMBIG,
REPLACE_AMBIG,
DEFINITE_AMBIG,
SIMILAR_AMBIG,
CASE_AMBIG,
AMBIG_TYPE_COUNT
} |
enum | SetParamConstraint { SET_PARAM_CONSTRAINT_NONE,
SET_PARAM_CONSTRAINT_DEBUG_ONLY,
SET_PARAM_CONSTRAINT_NON_DEBUG_ONLY,
SET_PARAM_CONSTRAINT_NON_INIT_ONLY
} |
enum | TessdataType {
TESSDATA_LANG_CONFIG,
TESSDATA_UNICHARSET,
TESSDATA_AMBIGS,
TESSDATA_INTTEMP,
TESSDATA_PFFMTABLE,
TESSDATA_NORMPROTO,
TESSDATA_PUNC_DAWG,
TESSDATA_SYSTEM_DAWG,
TESSDATA_NUMBER_DAWG,
TESSDATA_FREQ_DAWG,
TESSDATA_FIXED_LENGTH_DAWGS,
TESSDATA_CUBE_UNICHARSET,
TESSDATA_CUBE_SYSTEM_DAWG,
TESSDATA_SHAPE_TABLE,
TESSDATA_BIGRAM_DAWG,
TESSDATA_UNAMBIG_DAWG,
TESSDATA_PARAMS_TRAINING_MODEL,
TESSDATA_NUM_ENTRIES
} |
enum | CharSegmentationType { CST_FRAGMENT,
CST_WHOLE,
CST_IMPROPER,
CST_NGRAM
} |
enum | CountTypes {
CT_SHAPE_TOP_CORRECT,
CT_SHAPE_TOP_ERR,
CT_FONT_ATTR_ERR,
CT_UNICHAR_TOP1_ERR,
CT_UNICHAR_TOP2_ERR,
CT_UNICHAR_TOPN_ERR,
CT_OK_MULTI_UNICHAR,
CT_REJECT,
CT_NUM_RESULTS,
CT_RANK,
CT_REJECTED_JUNK,
CT_ACCEPTED_JUNK,
CT_SIZE
} |
enum | DawgType {
DAWG_TYPE_PUNCTUATION,
DAWG_TYPE_WORD,
DAWG_TYPE_NUMBER,
DAWG_TYPE_PATTERN,
DAWG_TYPE_COUNT
} |
enum | ColumnSpanningType {
CST_NOISE,
CST_FLOWING,
CST_HEADING,
CST_PULLOUT,
CST_COUNT
} |
enum | NeighbourPartitionType {
NPT_HTEXT,
NPT_VTEXT,
NPT_WEAK_HTEXT,
NPT_WEAK_VTEXT,
NPT_IMAGE,
NPT_COUNT
} |
enum | LeftOrRight { LR_LEFT,
LR_RIGHT
} |
enum | ColSegType {
COL_UNKNOWN,
COL_TEXT,
COL_TABLE,
COL_MIXED,
COL_COUNT
} |
enum | TabAlignment {
TA_LEFT_ALIGNED,
TA_LEFT_RAGGED,
TA_CENTER_JUSTIFIED,
TA_RIGHT_ALIGNED,
TA_RIGHT_RAGGED,
TA_SEPARATOR,
TA_COUNT
} |
Functions |
int | CubeAPITest (Boxa *boxa_blocks, Pixa *pixa_blocks, Boxa *boxa_words, Pixa *pixa_words, const FCOORD &reskew, Pix *page_pix, PAGE_RES *page_res) |
TBLOB * | make_tesseract_blob (float baseline, float xheight, float descender, float ascender, bool numeric_mode, Pix *pix) |
TBOX | char_box_to_tbox (Box *char_box, TBOX word_box, int x_offset) |
bool | IsTextOrEquationType (PolyBlockType type) |
bool | IsLeftIndented (const EquationDetect::IndentType type) |
bool | IsRightIndented (const EquationDetect::IndentType type) |
template<typename T > |
void | SimpleSwap (T &a, T &b) |
STRING | RtlEmbed (const STRING &word, bool rtlify) |
bool | IsLatinLetter (int ch) |
bool | IsDigitLike (int ch) |
bool | IsOpeningPunct (int ch) |
bool | IsTerminalPunct (int ch) |
const char * | SkipChars (const char *str, const char *toskip) |
const char * | SkipChars (const char *str, bool(*skip)(int)) |
const char * | SkipOne (const char *str, const char *toskip) |
bool | LikelyListNumeral (const STRING &word) |
bool | LikelyListMark (const STRING &word) |
bool | AsciiLikelyListItem (const STRING &word) |
int | UnicodeFor (const UNICHARSET *u, const WERD_CHOICE *werd, int pos) |
bool | LikelyListMarkUnicode (int ch) |
bool | UniLikelyListItem (const UNICHARSET *u, const WERD_CHOICE *werd) |
void | LeftWordAttributes (const UNICHARSET *unicharset, const WERD_CHOICE *werd, const STRING &utf8, bool *is_list, bool *starts_idea, bool *ends_idea) |
void | RightWordAttributes (const UNICHARSET *unicharset, const WERD_CHOICE *werd, const STRING &utf8, bool *is_list, bool *starts_idea, bool *ends_idea) |
int | ClosestCluster (const GenericVector< Cluster > &clusters, int value) |
void | CalculateTabStops (GenericVector< RowScratchRegisters > *rows, int row_start, int row_end, int tolerance, GenericVector< Cluster > *left_tabs, GenericVector< Cluster > *right_tabs) |
void | MarkRowsWithModel (GenericVector< RowScratchRegisters > *rows, int row_start, int row_end, const ParagraphModel *model, bool ltr, int eop_threshold) |
void | GeometricClassifyThreeTabStopTextBlock (int debug_level, GeometricClassifierState &s, ParagraphTheory *theory) |
void | GeometricClassify (int debug_level, GenericVector< RowScratchRegisters > *rows, int row_start, int row_end, ParagraphTheory *theory) |
bool | ValidFirstLine (const GenericVector< RowScratchRegisters > *rows, int row, const ParagraphModel *model) |
bool | ValidBodyLine (const GenericVector< RowScratchRegisters > *rows, int row, const ParagraphModel *model) |
bool | CrownCompatible (const GenericVector< RowScratchRegisters > *rows, int a, int b, const ParagraphModel *model) |
void | DiscardUnusedModels (const GenericVector< RowScratchRegisters > &rows, ParagraphTheory *theory) |
void | DowngradeWeakestToCrowns (int debug_level, ParagraphTheory *theory, GenericVector< RowScratchRegisters > *rows) |
void | RecomputeMarginsAndClearHypotheses (GenericVector< RowScratchRegisters > *rows, int start, int end, int percentile) |
int | InterwordSpace (const GenericVector< RowScratchRegisters > &rows, int row_start, int row_end) |
bool | FirstWordWouldHaveFit (const RowScratchRegisters &before, const RowScratchRegisters &after, tesseract::ParagraphJustification justification) |
bool | FirstWordWouldHaveFit (const RowScratchRegisters &before, const RowScratchRegisters &after) |
bool | TextSupportsBreak (const RowScratchRegisters &before, const RowScratchRegisters &after) |
bool | LikelyParagraphStart (const RowScratchRegisters &before, const RowScratchRegisters &after) |
bool | LikelyParagraphStart (const RowScratchRegisters &before, const RowScratchRegisters &after, tesseract::ParagraphJustification j) |
ParagraphModel | InternalParagraphModelByOutline (const GenericVector< RowScratchRegisters > *rows, int start, int end, int tolerance, bool *consistent) |
ParagraphModel | ParagraphModelByOutline (int debug_level, const GenericVector< RowScratchRegisters > *rows, int start, int end, int tolerance) |
bool | RowsFitModel (const GenericVector< RowScratchRegisters > *rows, int start, int end, const ParagraphModel *model) |
void | MarkStrongEvidence (GenericVector< RowScratchRegisters > *rows, int row_start, int row_end) |
void | ModelStrongEvidence (int debug_level, GenericVector< RowScratchRegisters > *rows, int row_start, int row_end, bool allow_flush_models, ParagraphTheory *theory) |
void | StrongEvidenceClassify (int debug_level, GenericVector< RowScratchRegisters > *rows, int row_start, int row_end, ParagraphTheory *theory) |
void | SeparateSimpleLeaderLines (GenericVector< RowScratchRegisters > *rows, int row_start, int row_end, ParagraphTheory *theory) |
void | ConvertHypothesizedModelRunsToParagraphs (int debug_level, const GenericVector< RowScratchRegisters > &rows, GenericVector< PARA * > *row_owners, ParagraphTheory *theory) |
bool | RowIsStranded (const GenericVector< RowScratchRegisters > &rows, int row) |
void | LeftoverSegments (const GenericVector< RowScratchRegisters > &rows, GenericVector< Interval > *to_fix, int row_start, int row_end) |
void | CanonicalizeDetectionResults (GenericVector< PARA * > *row_owners, PARA_LIST *paragraphs) |
void | DetectParagraphs (int debug_level, GenericVector< RowInfo > *row_infos, GenericVector< PARA * > *row_owners, PARA_LIST *paragraphs, GenericVector< ParagraphModel * > *models) |
void | InitializeRowInfo (const MutableIterator &it, RowInfo *info) |
void | DetectParagraphs (int debug_level, const MutableIterator *block_start, GenericVector< ParagraphModel * > *models) |
bool | StrongModel (const ParagraphModel *model) |
bool | read_t (PAGE_RES_IT *page_res_it, TBOX *tbox) |
ICOORD | ComputeEndFromGradient (const ICOORD &start, double m) |
bool | CompareFontInfo (const FontInfo &fi1, const FontInfo &fi2) |
bool | CompareFontSet (const FontSet &fs1, const FontSet &fs2) |
void | FontInfoDeleteCallback (FontInfo f) |
void | FontSetDeleteCallback (FontSet fs) |
bool | read_info (FILE *f, FontInfo *fi, bool swap) |
bool | write_info (FILE *f, const FontInfo &fi) |
bool | read_spacing_info (FILE *f, FontInfo *fi, bool swap) |
bool | write_spacing_info (FILE *f, const FontInfo &fi) |
bool | read_set (FILE *f, FontSet *fs, bool swap) |
bool | write_set (FILE *f, const FontSet &fs) |
void | OtsuThreshold (const unsigned char *imagedata, int bytes_per_pixel, int bytes_per_line, int left, int top, int width, int height, int **thresholds, int **hi_values) |
void | HistogramRect (const unsigned char *imagedata, int bytes_per_pixel, int bytes_per_line, int left, int top, int width, int height, int *histogram) |
int | OtsuStats (const int *histogram, int *H_out, int *omega0_out) |
| ELISTIZE (AmbigSpec) |
| ELISTIZEH (AmbigSpec) |
template<typename T > |
bool | cmp_eq (T const &t1, T const &t2) |
template<typename T > |
int | sort_cmp (const void *t1, const void *t2) |
template<typename T > |
int | sort_ptr_cmp (const void *t1, const void *t2) |
void | ClearFeatureSpaceWindow (NORM_METHOD norm_method, ScrollView *window) |
WERD_CHOICE * | get_best_delete_other (WERD_CHOICE *choice1, WERD_CHOICE *choice2) |
BLOB_CHOICE * | get_nth_choice (BLOB_CHOICE_LIST *blob_list, int n) |
UNICHAR_ID | get_top_choice_uid (BLOB_CHOICE_LIST *blob_list) |
int | find_choice_by_uid (BLOB_CHOICE_LIST *blob_list, UNICHAR_ID target_uid) |
WERD_CHOICE * | get_choice_from_posstr (const UNICHARSET *unicharset, const BLOB_CHOICE_LIST_VECTOR &char_choices, int start_pos, const char *pos_str, float *certainties) |
void | get_posstr_from_choice (const BLOB_CHOICE_LIST_VECTOR &char_choices, WERD_CHOICE *word_choice, int start_pos, char *pos_str) |
BLOB_CHOICE * | find_choice_by_type (BLOB_CHOICE_LIST *blob_choices, char target_type, const UNICHARSET &unicharset) |
BLOB_CHOICE * | find_choice_by_script (BLOB_CHOICE_LIST *blob_choices, int target_sid, int backup_sid, int secondary_sid) |
Pix * | GridReducedPix (const TBOX &box, int gridsize, ICOORD bleft, int *left, int *bottom) |
Pix * | TraceOutlineOnReducedPix (C_OUTLINE *outline, int gridsize, ICOORD bleft, int *left, int *bottom) |
Pix * | TraceBlockOnReducedPix (BLOCK *block, int gridsize, ICOORD bleft, int *left, int *bottom) |
template<class BBC > |
int | SortByBoxLeft (const void *void1, const void *void2) |
template<class BBC > |
int | SortRightToLeft (const void *void1, const void *void2) |
template<class BBC > |
int | SortByBoxBottom (const void *void1, const void *void2) |
template<typename T > |
void | DeleteObject (T *object) |
ShapeTable * | LoadShapeTable (const STRING &file_prefix) |
void | WriteShapeTable (const STRING &file_prefix, const ShapeTable &shape_table) |
MasterTrainer * | LoadTrainingData (int argc, const char *const *argv, bool replication, ShapeTable **shape_table, STRING *file_prefix) |
| ELISTIZE (ViterbiStateEntry) |
| ELISTIZEH (ViterbiStateEntry) |
template<class BLOB_CHOICE > |
int | SortByUnicharID (const void *void1, const void *void2) |
template<class BLOB_CHOICE > |
int | SortByRating (const void *void1, const void *void2) |
Variables |
const int | kMinRectSize = 10 |
const char | kTesseractReject = '~' |
const char | kUNLVReject = '~' |
const char | kUNLVSuspect = '^' |
const char * | kInputFile = "noname.tif" |
const char * | kOldVarsFile = "failed_vars.txt" |
const int | kMaxIntSize = 22 |
const int | kMinCredibleResolution = 70 |
| Minimum believable resolution.
|
const int | kMaxCredibleResolution = 2400 |
const int | kNumbersPerBlob = 5 |
const int | kBytesPerNumber = 5 |
const int | kBytesPerBlob = kNumbersPerBlob * (kBytesPerNumber + 1) + 1 |
const int | kBytesPerBoxFileLine = (kBytesPerNumber + 1) * kNumbersPerBlob + 1 |
const int | kBytesPer64BitNumber = 20 |
const int | kMaxBytesPerLine |
const int | kUniChs [] |
const int | kLatinChs [] |
const float | kMathDigitDensityTh1 = 0.25 |
const float | kMathDigitDensityTh2 = 0.1 |
const float | kMathItalicDensityTh = 0.5 |
const float | kUnclearDensityTh = 0.25 |
const int | kSeedBlobsCountTh = 10 |
const int | kLeftIndentAlignmentCountTh = 1 |
const int | kMaxCharTopRange = 48 |
const int | kDefaultResolution = 300 |
| Default resolution used if input in not believable.
|
const int | kMaxCircleErosions = 8 |
const int | kStrayLinePer = 6 |
const ParagraphModel * | kCrownLeft = reinterpret_cast<ParagraphModel *>(0xDEAD111F) |
const ParagraphModel * | kCrownRight = reinterpret_cast<ParagraphModel *>(0xDEAD888F) |
const inT16 | kMaxBoxEdgeDiff = 2 |
const int | kBoxClipTolerance = 2 |
const int | kMinSubscriptOffset = 20 |
const int | kMinSuperscriptOffset = 20 |
const int | kMaxDropCapBottom = -128 |
const int | kNumEndPoints = 3 |
const int | kHistogramSize = 256 |
CCUtilMutex | tprintfMutex |
const char * | kUTF8LineSeparator = "\u2028" |
const char * | kUTF8ParagraphSeparator = "\u2029" |
const char * | kLRM = "\u200E" |
const char * | kRLM = "\u200F" |
const char * | kRLE = "\u202A" |
const char * | kPDF = "\u202C" |
const char * | kHyphenLikeUTF8 [] |
const char * | kApostropheLikeUTF8 [] |
const int | kMaxOffsetDist = 32 |
const double | kMinPCLengthIncrease = 1.0 / 1024 |
const int | kMinClusteredShapes = 1 |
const int | kMaxUnicharsPerCluster = 2000 |
const float | kFontMergeDistance = 0.025 |
const float | kInfiniteDist = 999.0f |
const int | kRandomizingCenter = 128 |
const int | kTestChar = -1 |
const int | kSquareLimit = 25 |
const int | kPrime1 = 17 |
const int | kPrime2 = 13 |
const int | kMinOutlierSamples = 5 |
const int | kStateCnt = 4 |
const int | kNumLiteralCnt = 5 |
const int | case_state_table [6][4] |
const char | kDoNotReverse [] = "RRP_DO_NO_REVERSE" |
const char | kReverseIfHasRTL [] = "RRP_REVERSE_IF_HAS_RTL" |
const char | kForceReverse [] = "RRP_FORCE_REVERSE" |
const char *const | RTLReversePolicyNames [] |
const double | kAlignedFraction = 0.03125 |
const double | kRaggedFraction = 2.5 |
const double | kAlignedGapFraction = 0.75 |
const double | kRaggedGapFraction = 1.0 |
const int | kVLineAlignment = 3 |
const int | kVLineGutter = 1 |
const int | kVLineSearchSize = 150 |
const int | kMinRaggedTabs = 5 |
const int | kMinAlignedTabs = 4 |
const int | kVLineMinLength = 500 |
const double | kMinTabGradient = 4.0 |
const int | kMaxSkewFactor = 15 |
const char * | kTextordDebugPix = "psdebug_pix" |
const double | kMaxSmallNeighboursPerPix = 1.0 / 32 |
const int | kMaxLargeOverlapsWithSmall = 3 |
const int | kMaxMediumOverlapsWithSmall = 12 |
const int | kMaxLargeOverlapsWithMedium = 12 |
const int | kOriginalNoiseMultiple = 8 |
const int | kNoisePadding = 4 |
const double | kPhotoOffsetFraction = 0.375 |
const double | kMinGoodTextPARatio = 1.5 |
const int | kMinColumnWidth = 100 |
const int | kMaxIncompatibleColumnCount = 2 |
const double | kMarginOverlapFraction = 0.25 |
const double | kHorizontalGapMergeFraction = 0.5 |
const double | kMinNonNoiseFraction = 0.5 |
const double | kMinGutterWidthGrid = 0.5 |
const double | kMaxDistToPartSizeRatio = 1.5 |
bool | textord_tabfind_show_initial_partitions = false |
bool | textord_tabfind_show_reject_blobs = false |
int | textord_tabfind_show_partitions = 0 |
bool | textord_tabfind_show_columns = false |
bool | textord_tabfind_show_blocks = false |
bool | textord_tabfind_find_tables = true |
const int | kMaxPartnerDepth = 4 |
const double | kMaxSpacingDrift = 1.0 / 72 |
const double | kMaxTopSpacingFraction = 0.25 |
const double | kMaxSameBlockLineSpacing = 3 |
const double | kMaxSizeRatio = 1.5 |
const double | kMaxLeaderGapFractionOfMax = 0.25 |
const double | kMaxLeaderGapFractionOfMin = 0.5 |
const int | kMinLeaderCount = 5 |
const int | kLeaderCutCost = 8 |
const int | kMinStrongTextValue = 6 |
const int | kMinChainTextValue = 3 |
const int | kHorzStrongTextlineCount = 8 |
const int | kHorzStrongTextlineHeight = 10 |
const int | kHorzStrongTextlineAspect = 5 |
const double | kMaxBaselineError = 0.4375 |
const double | kMinBaselineCoverage = 0.5 |
const int | kMaxRMSColorNoise = 128 |
const int | kMaxColorDistance = 900 |
const int | kRGBRMSColors = 4 |
bool | textord_tabfind_show_color_fit = false |
const int | kMaxPadFactor = 6 |
const int | kMaxNeighbourDistFactor = 4 |
const int | kMaxCaptionLines = 7 |
const double | kMinCaptionGapRatio = 2.0 |
const double | kMinCaptionGapHeightRatio = 0.5 |
const double | kBigPartSizeRatio = 1.75 |
const double | kStrokeWidthFractionTolerance = 0.25 |
const double | kStrokeWidthConstantTolerance = 2.0 |
const double | kTinyEnoughTextlineOverlapFraction = 0.25 |
const double | kMaxPartitionSpacing = 1.75 |
const int | kSmoothDecisionMargin = 4 |
const double | kMinRectangularFraction = 0.125 |
const double | kMaxRectangularFraction = 0.75 |
const double | kMaxRectangularGradient = 0.1 |
const int | kMinImageFindSize = 100 |
const double | kRMSFitScaling = 8.0 |
const int | kMinColorDifference = 16 |
const int | kThinLineFraction = 20 |
| Denominator of resolution makes max pixel width to allow thin lines.
|
const int | kMinLineLengthFraction = 4 |
| Denominator of resolution makes min pixels to demand line lengths to be.
|
const int | kCrackSpacing = 100 |
| Spacing of cracks across the page to break up tall vertical lines.
|
const int | kLineFindGridSize = 50 |
| Grid size used by line finder. Not very critical.
|
const int | kMinThickLineWidth = 12 |
const int | kMaxLineResidue = 6 |
const double | kThickLengthMultiple = 0.75 |
const double | kMaxNonLineDensity = 0.25 |
const double | kMaxStaveHeight = 1.0 |
const double | kMinMusicPixelFraction = 0.75 |
int | textord_tabfind_show_strokewidths = 0 |
bool | textord_tabfind_only_strokewidths = false |
bool | textord_tabfind_vertical_text = true |
bool | textord_tabfind_force_vertical_text = false |
bool | textord_tabfind_vertical_horizontal_mix = true |
double | textord_tabfind_vertical_text_ratio = 0.5 |
const double | kStrokeWidthTolerance = 1.5 |
const double | kStrokeWidthFractionCJK = 0.25 |
const double | kStrokeWidthCJK = 2.0 |
const int | kCJKRadius = 2 |
const double | kCJKBrokenDistanceFraction = 0.25 |
const int | kCJKMaxComponents = 8 |
const double | kCJKAspectRatio = 1.25 |
const double | kCJKAspectRatioIncrease = 1.0625 |
const int | kMaxCJKSizeRatio = 5 |
const double | kBrokenCJKIterationFraction = 0.125 |
const double | kDiacriticXPadRatio = 7.0 |
const double | kDiacriticYPadRatio = 1.75 |
const double | kMinDiacriticSizeRatio = 1.0625 |
const double | kMaxDiacriticDistanceRatio = 1.25 |
const double | kMaxDiacriticGapToBaseCharHeight = 1.0 |
const int | kSearchRadius = 2 |
const int | kLineTrapLongest = 4 |
const int | kLineTrapShortest = 2 |
const int | kMostlyOneDirRatio = 3 |
const double | kLineResidueAspectRatio = 8.0 |
const int | kLineResiduePadRatio = 3 |
const double | kLineResidueSizeRatio = 1.75 |
const float | kSizeRatioToReject = 2.0 |
const int | kMaxLargeOverlaps = 3 |
const double | kNeighbourSearchFactor = 2.5 |
const int | kTabRadiusFactor = 5 |
const int | kMinVerticalSearch = 3 |
const int | kMaxVerticalSearch = 12 |
const int | kMaxRaggedSearch = 25 |
const int | kMinLinesInColumn = 10 |
const double | kMinFractionalLinesInColumn = 0.125 |
const double | kMinGutterWidthAbsolute = 0.02 |
const double | kMaxGutterWidthAbsolute = 2.00 |
const int | kRaggedGutterMultiple = 5 |
const double | kLineFragmentAspectRatio = 10.0 |
const double | kSmoothFactor = 0.25 |
const double | kCharVerticalOverlapFraction = 0.375 |
const double | kMaxHorizontalGap = 3.0 |
const int | kMinEvaluatedTabs = 3 |
const int | kMaxTextLineBlobRatio = 5 |
const int | kMinTextLineBlobRatio = 3 |
const double | kMinImageArea = 0.5 |
const double | kCosMaxSkewAngle = 0.866025 |
bool | textord_tabfind_show_initialtabs = false |
bool | textord_tabfind_show_finaltabs = false |
double | textord_tabfind_aligned_gap_fraction = 0.75 |
const int | kColumnWidthFactor = 20 |
const int | kMaxVerticalSpacing = 500 |
const int | kMaxBlobWidth = 500 |
const double | kSplitPartitionSize = 2.0 |
const double | kAllowTextHeight = 0.5 |
const double | kAllowTextWidth = 0.6 |
const double | kAllowTextArea = 0.8 |
const double | kAllowBlobHeight = 0.3 |
const double | kAllowBlobWidth = 0.4 |
const double | kAllowBlobArea = 0.05 |
const int | kMinBoxesInTextPartition = 10 |
const int | kMaxBoxesInDataPartition = 20 |
const double | kMaxGapInTextPartition = 4.0 |
const double | kMinMaxGapInTextPartition = 0.5 |
const double | kMaxBlobOverlapFactor = 4.0 |
const double | kMaxTableCellXheight = 2.0 |
const int | kMaxColumnHeaderDistance = 4 |
const double | kTableColumnThreshold = 3.0 |
const int | kRulingVerticalMargin = 3 |
const double | kMinOverlapWithTable = 0.6 |
const int | kSideSpaceMargin = 10 |
const double | kSmallTableProjectionThreshold = 0.35 |
const double | kLargeTableProjectionThreshold = 0.45 |
const int | kLargeTableRowCount = 6 |
const int | kMinRowsInTable = 3 |
const double | kRequiredFullJustifiedSpacing = 4.0 |
const int | kAdjacentLeaderSearchPadding = 2 |
const double | kParagraphEndingPreviousLineRatio = 1.3 |
const double | kMaxParagraphEndingLeftSpaceMultiple = 3.0 |
const double | kMinParagraphEndingTextToWhitespaceRatio = 3.0 |
const double | kMaxXProjectionGapFactor = 2.0 |
const double | kStrokeWidthFractionalTolerance = 0.25 |
bool | textord_dump_table_images = false |
bool | textord_show_tables = false |
bool | textord_tablefind_show_mark = false |
bool | textord_tablefind_show_stats = false |
bool | textord_tablefind_recognize_tables = false |
const double | kHorizontalSpacing = 0.30 |
const double | kVerticalSpacing = -0.2 |
const int | kCellSplitRowThreshold = 0 |
const int | kCellSplitColumnThreshold = 0 |
const int | kLinedTableMinVerticalLines = 3 |
const int | kLinedTableMinHorizontalLines = 3 |
const double | kRequiredColumns = 0.7 |
const double | kMarginFactor = 1.1 |
const double | kMaxRowSize = 2.5 |
const double | kGoodRowNumberOfColumnsSmall [] = { 2, 2, 2, 2, 2, 3, 3 } |
const int | kGoodRowNumberOfColumnsSmallSize |
const double | kGoodRowNumberOfColumnsLarge = 0.7 |
const double | kMinFilledArea = 0.35 |
const int | kGutterMultiple = 4 |
const int | kGutterToNeighbourRatio = 3 |
const int | kSimilarVectorDist = 10 |
const int | kSimilarRaggedDist = 50 |
const int | kMaxFillinMultiple = 11 |
const double | kMinGutterFraction = 0.5 |
const double | kLineCountReciprocal = 4.0 |
const double | kMinAlignedGutter = 0.25 |
const double | kMinRaggedGutter = 1.5 |
double | textord_tabvector_vertical_gap_fraction = 0.5 |
double | textord_tabvector_vertical_box_ratio = 0.5 |
const char * | kAlignmentNames [] |
Make a word from the selected blobs and run Tess on them.
fp_eval_word_spacing() Evaluation function for fixed pitch word lists.
Basically, count the number of "nice" characters - those which are in tess acceptable words or in dict words and are not rejected. Penalise any potential noise chars
Walk the current block list applying the specified word processor function to each word that overlaps the selection_box.
Process a command returned from the command window (Just call the appropriate command handler)
Blank display of word then redisplay word according to current display mode settings
---------------------------------------------------------------------------- Public Code ----------------------------------------------------------------------------
---------------------------------------------------------------------------- Include Files and Type Defines ----------------------------------------------------------------------------
---------------------------------------------------------------------------- Include Files and Type Defines ---------------------------------------------------------------------------- ---------------------------------------------------------------------------- Public Code ----------------------------------------------------------------------------