Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/tesseract/src/classify/intfeaturespace.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 // Copyright 2010 Google Inc. All Rights Reserved.
2 // Author: rays@google.com (Ray Smith)
3 ///////////////////////////////////////////////////////////////////////
4 // File: intfeaturespace.h
5 // Description: Indexed feature space based on INT_FEATURE_STRUCT.
6 //
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 //
17 ///////////////////////////////////////////////////////////////////////
18
19 #ifndef TESSERACT_CLASSIFY_INTFEATURESPACE_H_
20 #define TESSERACT_CLASSIFY_INTFEATURESPACE_H_
21
22 #include "intproto.h"
23
24 // Extent of x,y,theta in the input feature space. [0,255].
25 const int kIntFeatureExtent = 256;
26 // Extent of x,y,theta dimensions in the quantized feature space.
27 const int kBoostXYBuckets = 16;
28 const int kBoostDirBuckets = 16;
29
30 namespace tesseract {
31
32 class IndexMap;
33
34 // Down-sampling quantization of the INT_FEATURE_STRUCT feature space and
35 // conversion to a single scalar index value, used as a binary feature space.
36 class TESS_API IntFeatureSpace {
37 public:
38 IntFeatureSpace();
39 // Default copy constructors and assignment OK!
40
41 // Setup the feature space with the given dimensions.
42 void Init(uint8_t xbuckets, uint8_t ybuckets, uint8_t thetabuckets);
43
44 // Serializes the feature space definition to the given file.
45 // Returns false on error.
46 bool Serialize(FILE *fp) const;
47
48 // Returns the total size of the feature space.
49 int Size() const {
50 return static_cast<int>(x_buckets_) * y_buckets_ * theta_buckets_;
51 }
52 // Returns an INT_FEATURE_STRUCT corresponding to the given index.
53 // This is the inverse of the Index member.
54 INT_FEATURE_STRUCT PositionFromIndex(int index) const;
55
56 // Returns a 1-dimensional index corresponding to the given feature value.
57 // Range is [0, Size()-1]. Inverse of PositionFromIndex member.
58 int Index(const INT_FEATURE_STRUCT &f) const {
59 return (XBucket(f.X) * y_buckets_ + YBucket(f.Y)) * theta_buckets_ + ThetaBucket(f.Theta);
60 }
61 // Bulk calls to Index. Maps the given array of features to a vector of
62 // int32_t indices in the same order as the input.
63 void IndexFeatures(const INT_FEATURE_STRUCT *features, int num_features,
64 std::vector<int> *mapped_features) const;
65 // Bulk calls to Index. Maps the given array of features to a vector of
66 // sorted int32_t indices.
67 void IndexAndSortFeatures(const INT_FEATURE_STRUCT *features, int num_features,
68 std::vector<int> *sorted_features) const;
69 // Returns a feature space index for the given x,y position in a display
70 // window, or -1 if the feature is a miss.
71 int XYToFeatureIndex(int x, int y) const;
72
73 protected:
74 // Converters to generate indices for individual feature dimensions.
75 int XBucket(int x) const {
76 int bucket = x * x_buckets_ / kIntFeatureExtent;
77 return ClipToRange(bucket, 0, static_cast<int>(x_buckets_) - 1);
78 }
79 int YBucket(int y) const {
80 int bucket = y * y_buckets_ / kIntFeatureExtent;
81 return ClipToRange(bucket, 0, static_cast<int>(y_buckets_) - 1);
82 }
83 // Use DivRounded for theta so that exactly vertical and horizontal are in
84 // the middle of a bucket. The Modulo takes care of the wrap-around.
85 int ThetaBucket(int theta) const {
86 int bucket = DivRounded(theta * theta_buckets_, kIntFeatureExtent);
87 return Modulo(bucket, theta_buckets_);
88 }
89 // Returns an INT_FEATURE_STRUCT corresponding to the given buckets.
90 INT_FEATURE_STRUCT PositionFromBuckets(int x, int y, int theta) const;
91
92 // Feature space definition - serialized.
93 uint8_t x_buckets_;
94 uint8_t y_buckets_;
95 uint8_t theta_buckets_;
96 };
97
98 } // namespace tesseract.
99
100 #endif // TESSERACT_CLASSIFY_INTFEATURESPACE_H_