-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathCHANGES
129 lines (71 loc) · 3.37 KB
/
CHANGES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
nhocr-0.22.tar.gz : (Aug. 29, 2014)
* Some code fixes have been made.
* A part of the image manipulation library O2-tools has been
included in the source tree.
* Vertical writing has been supported. However, the accompanying
dictionaries in this release lack vertical fonts, especially in
symbols. Many characters are compatible and could be recognized
well also in vertical documents.
The dictionaries with vertical fonts will be available in future
releases.
nhocr-0.21.tar.gz : (May 3, 2013)
* Language post-processing support has been added.
(Server "gramd" by Tomasz Mikolajewski is required to
enable the feature.)
This feature is experimental, and some side effects may exist.
See also README-gramd.
* Minor bug-fixes.
nhocr-0.20.tar.gz : (Feb. 12, 2010)
* Automatic image polarity detection has been added in the
multi-character recognition mode.
* makechar has been fixed. (Now it excludes SPACE character.)
* Command line parsing has been fixed.
* The location of makechar and makevec has been changed to
sbin/ from bin/.
* The output format in the multi-character recognition mode
has been revised so that it complies with the WeOCR API.
* makedic has been added.
nhocr-0.19.tar.gz : (Dec. 31, 2009)
* Internal handling for SPACE character has been changed.
* Multi-character recognition mode (-mchar option) has been
added.
nhocr-0.18.tar.gz : (Oct. 15, 2009)
* Multi-dictionary support has been added.
* Dictionaries have been revised.
The current character dictionaries are made from Ume-fonts
0.408 (http://sourceforge.jp/projects/ume-font/).
* Force-alphabet mode has been added. (experimental)
* Dictionary development tools (makechar, makevec) have
been added.
nhocr-0.17.tar.gz : (May 30, 2009)
* Text block recognition mode has been added.
* Some fixes in configure script.
nhocr-0.16.tar.gz : (May 15, 2009)
* The first source code release.
* Some code refinements for source code release.
* The license has been changed to Apache License, Version 2.0
from MIT-X.
* The native character code has been changed to UTF-8 from EUC-JP.
* Command name has been changed to nhocr from rec_line.
* NHocr-OCRopus bridge (Lua script) has been updated.
* Environment variable NHOCR_LIBDIR has been renamed to NHOCR_DICDIR.
* Autoconf has been introduced.
nhocr-1.5e-32-i686-linux.tar.gz : (Dec. 31, 2008)
* A feature of reading multiple input files is supported on
rec_line. This feature is useful for improving the system
performance by avoiding multiple OCR initializations.
* Some code refinements for forthcoming source code release.
nhocr-1.4e-32-i686-linux.tar.gz : (Oct. 5, 2008)
* An ad-hoc tune has been made in character segmentation
to improve performance with some horizontally split Japanese
characters.
* An ad-hoc Wide/ASCII detection mechanism has been introduced.
nhocr-1.3e-36-i686-linux.tar.gz : (Sep. 26, 2008)
* rec_line does not print output text on the screen now.
A new option "-o -" will enable that.
* Automatic foreground/background detector and image
inverter have been added.
* A quick and dirty script of NHocr-OCRopus(0.2) bridge.
* A minor enhancement for reducing noisy output.
nhocr-1.2e-37-i686-linux.tar.gz : (Sep. 15, 2008)
* The initial Linux i686 binary distribution of NHocr.