tesseract  5.0.0-alpha-619-ge9db
wordrec.cpp
Go to the documentation of this file.
1 // File: wordrec.cpp
3 // Description: wordrec class.
4 // Author: Samuel Charron
5 //
6 // (C) Copyright 2006, Google Inc.
7 // Licensed under the Apache License, Version 2.0 (the "License");
8 // you may not use this file except in compliance with the License.
9 // You may obtain a copy of the License at
10 // http://www.apache.org/licenses/LICENSE-2.0
11 // Unless required by applicable law or agreed to in writing, software
12 // distributed under the License is distributed on an "AS IS" BASIS,
13 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 // See the License for the specific language governing permissions and
15 // limitations under the License.
16 //
18 
19 #include "wordrec.h"
20 
21 #ifdef DISABLED_LEGACY_ENGINE
22 
23 #include "params.h"
24 
25 
26 namespace tesseract {
28  // control parameters
29 
30  BOOL_MEMBER(wordrec_debug_blamer, false,
31  "Print blamer debug messages", params()),
32 
33  BOOL_MEMBER(wordrec_run_blamer, false,
34  "Try to set the blame for errors", params()) {
35  prev_word_best_choice_ = nullptr;
36 }
37 
38 } // namespace tesseract
39 
40 #else // DISABLED_LEGACY_ENGINE not defined
41 
42 #include "language_model.h"
43 #include "params.h"
44 
45 
46 namespace tesseract {
48  // control parameters
49  BOOL_MEMBER(merge_fragments_in_matrix, true,
50  "Merge the fragments in the ratings matrix and delete them"
51  " after merging", params()),
52  BOOL_MEMBER(wordrec_enable_assoc, true, "Associator Enable",
53  params()),
54  BOOL_MEMBER(force_word_assoc, false,
55  "force associator to run regardless of what enable_assoc is."
56  " This is used for CJK where component grouping is necessary.",
57  CCUtil::params()),
58  INT_MEMBER(repair_unchopped_blobs, 1, "Fix blobs that aren't chopped",
59  params()),
60  double_MEMBER(tessedit_certainty_threshold, -2.25, "Good blob limit",
61  params()),
62  INT_MEMBER(chop_debug, 0, "Chop debug",
63  params()),
64  BOOL_MEMBER(chop_enable, 1, "Chop enable",
65  params()),
66  BOOL_MEMBER(chop_vertical_creep, 0, "Vertical creep",
67  params()),
68  INT_MEMBER(chop_split_length, 10000, "Split Length",
69  params()),
70  INT_MEMBER(chop_same_distance, 2, "Same distance",
71  params()),
72  INT_MEMBER(chop_min_outline_points, 6, "Min Number of Points on Outline",
73  params()),
74  INT_MEMBER(chop_seam_pile_size, 150, "Max number of seams in seam_pile",
75  params()),
76  BOOL_MEMBER(chop_new_seam_pile, 1, "Use new seam_pile", params()),
77  INT_MEMBER(chop_inside_angle, -50, "Min Inside Angle Bend",
78  params()),
79  INT_MEMBER(chop_min_outline_area, 2000, "Min Outline Area",
80  params()),
81  double_MEMBER(chop_split_dist_knob, 0.5, "Split length adjustment",
82  params()),
83  double_MEMBER(chop_overlap_knob, 0.9, "Split overlap adjustment",
84  params()),
85  double_MEMBER(chop_center_knob, 0.15, "Split center adjustment",
86  params()),
87  INT_MEMBER(chop_centered_maxwidth, 90, "Width of (smaller) chopped blobs "
88  "above which we don't care that a chop is not near the center.",
89  params()),
90  double_MEMBER(chop_sharpness_knob, 0.06, "Split sharpness adjustment",
91  params()),
92  double_MEMBER(chop_width_change_knob, 5.0, "Width change adjustment",
93  params()),
94  double_MEMBER(chop_ok_split, 100.0, "OK split limit",
95  params()),
96  double_MEMBER(chop_good_split, 50.0, "Good split limit",
97  params()),
98  INT_MEMBER(chop_x_y_weight, 3, "X / Y length weight",
99  params()),
100  BOOL_MEMBER(assume_fixed_pitch_char_segment, false,
101  "include fixed-pitch heuristics in char segmentation",
102  params()),
103  INT_MEMBER(wordrec_debug_level, 0,
104  "Debug level for wordrec", params()),
105  INT_MEMBER(wordrec_max_join_chunks, 4,
106  "Max number of broken pieces to associate", params()),
107  BOOL_MEMBER(wordrec_skip_no_truth_words, false,
108  "Only run OCR for words that had truth recorded in BlamerBundle",
109  params()),
110  BOOL_MEMBER(wordrec_debug_blamer, false,
111  "Print blamer debug messages", params()),
112  BOOL_MEMBER(wordrec_run_blamer, false,
113  "Try to set the blame for errors", params()),
114  INT_MEMBER(segsearch_debug_level, 0,
115  "SegSearch debug level", params()),
116  INT_MEMBER(segsearch_max_pain_points, 2000,
117  "Maximum number of pain points stored in the queue",
118  params()),
119  INT_MEMBER(segsearch_max_futile_classifications, 20,
120  "Maximum number of pain point classifications per chunk that"
121  " did not result in finding a better word choice.",
122  params()),
123  double_MEMBER(segsearch_max_char_wh_ratio, 2.0,
124  "Maximum character width-to-height ratio", params()),
125  BOOL_MEMBER(save_alt_choices, true,
126  "Save alternative paths found during chopping"
127  " and segmentation search",
128  params()),
129  pass2_ok_split(0.0f) {
130  prev_word_best_choice_ = nullptr;
132  &(getDict())));
133  fill_lattice_ = nullptr;
134 }
135 
136 } // namespace tesseract
137 
138 #endif // DISABLED_LEGACY_ENGINE
tesseract::Wordrec::Wordrec
Wordrec()
Definition: wordrec.cpp:47
language_model.h
INT_MEMBER
#define INT_MEMBER(name, val, comment, vec)
Definition: params.h:312
params.h
wordrec.h
tesseract::Wordrec::prev_word_best_choice_
WERD_CHOICE * prev_word_best_choice_
Definition: wordrec.h:476
tesseract::LanguageModel
Definition: language_model.h:50
tesseract::Classify::getDict
virtual Dict & getDict()
Definition: classify.h:107
tesseract::Classify::get_fontinfo_table
UnicityTable< FontInfo > & get_fontinfo_table()
Definition: classify.h:386
tesseract::Wordrec::fill_lattice_
void(Wordrec::* fill_lattice_)(const MATRIX &ratings, const WERD_CHOICE_LIST &best_choices, const UNICHARSET &unicharset, BlamerBundle *blamer_bundle)
Definition: wordrec.h:480
double_MEMBER
#define double_MEMBER(name, val, comment, vec)
Definition: params.h:321
tesseract
Definition: baseapi.h:65
tesseract::Wordrec::language_model_
std::unique_ptr< LanguageModel > language_model_
Definition: wordrec.h:471
BOOL_MEMBER
#define BOOL_MEMBER(name, val, comment, vec)
Definition: params.h:315
tesseract::CCUtil
Definition: ccutil.h:40