tesseract  5.0.0-alpha-619-ge9db
associate.h
Go to the documentation of this file.
1 // File: associate.h
3 // Description: Structs, classes, typedefs useful for the segmentation
4 // search. Functions for scoring segmentation paths according
5 // to their character widths, gap widths and seam cuts.
6 // Author: Daria Antonova
7 // Created: Mon Mar 8 11:26:43 PDT 2010
8 //
9 // (C) Copyright 2010, Google Inc.
10 // Licensed under the Apache License, Version 2.0 (the "License");
11 // you may not use this file except in compliance with the License.
12 // You may obtain a copy of the License at
13 // http://www.apache.org/licenses/LICENSE-2.0
14 // Unless required by applicable law or agreed to in writing, software
15 // distributed under the License is distributed on an "AS IS" BASIS,
16 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
17 // See the License for the specific language governing permissions and
18 // limitations under the License.
19 //
21 
22 #ifndef ASSOCIATE_H
23 #define ASSOCIATE_H
24 
25 #include "blobs.h"
26 #include "elst.h"
27 #include "ratngs.h"
28 #include "seam.h"
29 #include "split.h"
30 
31 class WERD_RES;
32 
33 namespace tesseract {
34 
35 // Statisitcs about character widths, gaps and seams.
38 
39  void Clear() {
40  shape_cost = 0.0f;
41  bad_shape = false;
42  full_wh_ratio = 0.0f;
43  full_wh_ratio_total = 0.0f;
44  full_wh_ratio_var = 0.0f;
47  gap_sum = 0;
48  }
49 
50  void Print() { tprintf("AssociateStats: s(%g %d)\n", shape_cost, bad_shape); }
51 
52  float shape_cost; // cost of blob shape
53  bool bad_shape; // true if the shape of the blob is unacceptable
54  float full_wh_ratio; // width-to-hight ratio + gap on the right
55  float full_wh_ratio_total; // sum of width-to-hight ratios
56  // on the path terminating at this blob
57  float full_wh_ratio_var; // variance of full_wh_ratios on the path
58  bool bad_fixed_pitch_right_gap; // true if there is no gap before
59  // the blob on the right
60  bool bad_fixed_pitch_wh_ratio; // true if the blobs has width-to-hight
61  // ratio > kMaxFixedPitchCharAspectRatio
62  int gap_sum; // sum of gaps within the blob
63 };
64 
65 // Utility functions for scoring segmentation paths according to their
66 // character widths, gap widths, seam characteristics.
68  public:
69  static const float kMaxFixedPitchCharAspectRatio;
70  static const float kMinGap;
71 
72  // Returns outline length of the given blob is computed as:
73  // rating_cert_scale * rating / certainty
74  // Since from Wordrec::SegSearch() in segsearch.cpp
75  // rating_cert_scale = -1.0 * getDict().certainty_scale / rating_scale
76  // And from Classify::ConvertMatchesToChoices() in adaptmatch.cpp
77  // Rating = Certainty = next.rating
78  // Rating *= rating_scale * Results->BlobLength
79  // Certainty *= -(getDict().certainty_scale)
80  static inline float ComputeOutlineLength(float rating_cert_scale,
81  const BLOB_CHOICE &b) {
82  return rating_cert_scale * b.rating() / b.certainty();
83  }
84  static inline float ComputeRating(float rating_cert_scale,
85  float cert, int width) {
86  return static_cast<float>(width) * cert / rating_cert_scale;
87  }
88 
89  // Computes character widths, gaps and seams stats given the
90  // AssociateStats of the path so far, col, row of the blob that
91  // is being added to the path, and WERD_RES containing information
92  // about character widths, gaps and seams.
93  // Fills associate_cost with the combined shape, gap and seam cost
94  // of adding a unichar from (col, row) to the path (note that since
95  // this function could be used to compute the prioritization for
96  // pain points, (col, row) entry might not be classified yet; thus
97  // information in the (col, row) entry of the ratings matrix is not used).
98  //
99  // Note: the function assumes that word_res, stats and
100  // associate_cost pointers are not nullptr.
101  static void ComputeStats(int col, int row,
102  const AssociateStats *parent_stats,
103  int parent_path_length,
104  bool fixed_pitch,
105  float max_char_wh_ratio,
106  WERD_RES *word_res,
107  bool debug,
108  AssociateStats *stats);
109 
110  // Returns the width cost for fixed-pitch text.
111  static float FixedPitchWidthCost(float norm_width, float right_gap,
112  bool end_pos, float max_char_wh_ratio);
113 
114  // Returns the gap cost for fixed-pitch text (penalizes vertically
115  // overlapping components).
116  static inline float FixedPitchGapCost(float norm_gap, bool end_pos) {
117  return (norm_gap < 0.05 && !end_pos) ? 5.0f : 0.0f;
118  }
119 };
120 
121 } // namespace tesseract
122 
123 #endif
elst.h
split.h
tesseract::AssociateUtils::FixedPitchGapCost
static float FixedPitchGapCost(float norm_gap, bool end_pos)
Definition: associate.h:116
BLOB_CHOICE::certainty
float certainty() const
Definition: ratngs.h:81
tesseract::AssociateStats::Clear
void Clear()
Definition: associate.h:39
tesseract::AssociateUtils::ComputeRating
static float ComputeRating(float rating_cert_scale, float cert, int width)
Definition: associate.h:84
tesseract::AssociateStats::bad_shape
bool bad_shape
Definition: associate.h:53
tesseract::AssociateUtils::kMaxFixedPitchCharAspectRatio
static const float kMaxFixedPitchCharAspectRatio
Definition: associate.h:69
WERD_RES
Definition: pageres.h:160
tesseract::AssociateStats::full_wh_ratio_total
float full_wh_ratio_total
Definition: associate.h:55
tesseract::AssociateStats::bad_fixed_pitch_right_gap
bool bad_fixed_pitch_right_gap
Definition: associate.h:58
blobs.h
tesseract::AssociateStats
Definition: associate.h:36
ratngs.h
tesseract::AssociateUtils::FixedPitchWidthCost
static float FixedPitchWidthCost(float norm_width, float right_gap, bool end_pos, float max_char_wh_ratio)
Definition: associate.cpp:156
tesseract::AssociateStats::full_wh_ratio_var
float full_wh_ratio_var
Definition: associate.h:57
tesseract
Definition: baseapi.h:65
tesseract::AssociateStats::full_wh_ratio
float full_wh_ratio
Definition: associate.h:54
BLOB_CHOICE::rating
float rating() const
Definition: ratngs.h:78
tesseract::AssociateUtils::ComputeOutlineLength
static float ComputeOutlineLength(float rating_cert_scale, const BLOB_CHOICE &b)
Definition: associate.h:80
BLOB_CHOICE
Definition: ratngs.h:49
tesseract::AssociateStats::gap_sum
int gap_sum
Definition: associate.h:62
tesseract::AssociateUtils
Definition: associate.h:67
tprintf
DLLSYM void tprintf(const char *format,...)
Definition: tprintf.cpp:34
tesseract::AssociateUtils::ComputeStats
static void ComputeStats(int col, int row, const AssociateStats *parent_stats, int parent_path_length, bool fixed_pitch, float max_char_wh_ratio, WERD_RES *word_res, bool debug, AssociateStats *stats)
Definition: associate.cpp:34
seam.h
tesseract::AssociateStats::Print
void Print()
Definition: associate.h:50
tesseract::AssociateStats::bad_fixed_pitch_wh_ratio
bool bad_fixed_pitch_wh_ratio
Definition: associate.h:60
tesseract::AssociateUtils::kMinGap
static const float kMinGap
Definition: associate.h:70
tesseract::AssociateStats::AssociateStats
AssociateStats()
Definition: associate.h:37
tesseract::AssociateStats::shape_cost
float shape_cost
Definition: associate.h:52