Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/tesseract/src/classify/shapeclassifier.h @ 2:b50eed0cc0ef upstream

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
ADD: MuPDF v1.26.7: the MuPDF source as downloaded by a default build of PyMuPDF 1.26.4. The directory name has changed: no version number in the expanded directory now.
author Franz Glasner <fzglas.hg@dom66.de>
date Mon, 15 Sep 2025 11:43:07 +0200
parents
children
comparison
equal deleted inserted replaced
1:1d09e1dec1d9 2:b50eed0cc0ef
1 ///////////////////////////////////////////////////////////////////////
2 // File: shapeclassifier.h
3 // Description: Base interface class for classifiers that return a
4 // shape index.
5 // Author: Ray Smith
6 //
7 // (C) Copyright 2011, Google Inc.
8 // Licensed under the Apache License, Version 2.0 (the "License");
9 // you may not use this file except in compliance with the License.
10 // You may obtain a copy of the License at
11 // http://www.apache.org/licenses/LICENSE-2.0
12 // Unless required by applicable law or agreed to in writing, software
13 // distributed under the License is distributed on an "AS IS" BASIS,
14 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15 // See the License for the specific language governing permissions and
16 // limitations under the License.
17 //
18 ///////////////////////////////////////////////////////////////////////
19
20 #ifndef TESSERACT_CLASSIFY_SHAPECLASSIFIER_H_
21 #define TESSERACT_CLASSIFY_SHAPECLASSIFIER_H_
22
23 #include "image.h"
24
25 #include <tesseract/unichar.h>
26
27 #include <vector>
28
29 struct Pix;
30
31 namespace tesseract {
32
33 class ScrollView;
34 class UNICHARSET;
35
36 struct ShapeRating;
37 class ShapeTable;
38 class TrainingSample;
39 class TrainingSampleSet;
40 struct UnicharRating;
41
42 // Interface base class for classifiers that produce ShapeRating results.
43 class TESS_API ShapeClassifier {
44 public:
45 virtual ~ShapeClassifier() = default;
46
47 // Classifies the given [training] sample, writing to results.
48 // If page_pix is not nullptr, the overriding function may call
49 // sample.GetSamplePix(padding, page_pix) to get an image of the sample
50 // padded (with real image data) by the given padding to extract features
51 // from the image of the character. Other members of TrainingSample:
52 // features(), micro_features(), cn_feature(), geo_feature() may be used
53 // to get the appropriate tesseract features.
54 // If debug is non-zero, then various degrees of classifier dependent debug
55 // information is provided.
56 // If keep_this (a UNICHAR_ID) is >= 0, then the results should always
57 // contain keep_this, and (if possible) anything of intermediate confidence.
58 // (Used for answering "Why didn't it get that right?" questions.) It must
59 // be a UNICHAR_ID as the callers have no clue how to choose the best shape
60 // that may contain a desired answer.
61 // The return value is the number of classes saved in results.
62 // NOTE that overriding functions MUST clear and sort the results by
63 // descending rating unless the classifier is working with a team of such
64 // classifiers.
65 // NOTE: Neither overload of ClassifySample is pure, but at least one must
66 // be overridden by a classifier in order for it to do anything.
67 virtual int UnicharClassifySample(const TrainingSample &sample, Image page_pix, int debug,
68 UNICHAR_ID keep_this, std::vector<UnicharRating> *results);
69
70 protected:
71 virtual int ClassifySample(const TrainingSample &sample, Image page_pix, int debug,
72 UNICHAR_ID keep_this, std::vector<ShapeRating> *results);
73
74 public:
75 // Returns the shape that contains unichar_id that has the best result.
76 // If result is not nullptr, it is set with the shape_id and rating.
77 // Returns -1 if ClassifySample fails to provide any result containing
78 // unichar_id. BestShapeForUnichar does not need to be overridden if
79 // ClassifySample respects the keep_this rule.
80 virtual int BestShapeForUnichar(const TrainingSample &sample, Image page_pix,
81 UNICHAR_ID unichar_id, ShapeRating *result);
82
83 // Provides access to the ShapeTable that this classifier works with.
84 virtual const ShapeTable *GetShapeTable() const = 0;
85 // Provides access to the UNICHARSET that this classifier works with.
86 // Must be overridden IFF GetShapeTable() returns nullptr.
87 virtual const UNICHARSET &GetUnicharset() const;
88
89 // Visual debugger classifies the given sample, displays the results and
90 // solicits user input to display other classifications. Returns when
91 // the user has finished with debugging the sample.
92 // Probably doesn't need to be overridden if the subclass provides
93 // DisplayClassifyAs.
94 void DebugDisplay(const TrainingSample &sample, Image page_pix, UNICHAR_ID unichar_id);
95
96 // Displays classification as the given unichar_id. Creates as many windows
97 // as it feels fit, using index as a guide for placement. Adds any created
98 // windows to the windows output and returns a new index that may be used
99 // by any subsequent classifiers. Caller waits for the user to view and
100 // then destroys the windows by clearing the vector.
101 virtual int DisplayClassifyAs(const TrainingSample &sample, Image page_pix, UNICHAR_ID unichar_id,
102 int index, std::vector<ScrollView *> &windows);
103
104 // Prints debug information on the results. context is some introductory/title
105 // message.
106 virtual void UnicharPrintResults(const char *context,
107 const std::vector<UnicharRating> &results) const;
108 virtual void PrintResults(const char *context, const std::vector<ShapeRating> &results) const;
109
110 protected:
111 // Removes any result that has all its unichars covered by a better choice,
112 // regardless of font.
113 void FilterDuplicateUnichars(std::vector<ShapeRating> *results) const;
114 };
115
116 } // namespace tesseract.
117
118 #endif // TESSERACT_CLASSIFY_SHAPECLASSIFIER_H_