(Release 1.00)
1. About this program.
This program, lcut_s, is the tool for extracting text lines in text
block image. A relatively simple algorithm is used.
Note: There is no technical article or report about the algorithm,
because I think the algorithm is not so difficult and is
not worth writing about.
The strong point of this program is that the existence of ruby
characters in Japanese document, underlines and overlines are taken
into account. Ruby characters here are the printed Japanese
characters alongside Kanji characters. These characters and
lines are the attendant elements on main text lines, and they
should be analyzed and recognized together. This program can
combine these elements with main text lines.
Here is an example.
Another advantage of this program is that the program is tolerant
of slight skew of text lines. This property comes from the algo-
rithm of text line extraction based on local projection profile.
This program is originally one of my private tools for Document
Analysis and Recognition. Some redundant and experimental codes
may be included.
There is a copyright notice in every file of the source code. I
have granted that you have the right to use and distribute the code
in this package without paying me, as long as you do it on a non-
commercial act. Please remember to check the copyright notes.
2. Requirements
Under every Operating System:
1) Following packages provided with this package are required.
O2-libsrc-1.xx.tar.gz (Release 1.03 or later)
O2-dlibsrc-1.xx.tar.gz (Release 1.00 or later)
O2-dtools-1.xx.tar.gz (Release 1.01 or later)
The following package is not required. But, I recommend you
to have it.
O2-tools-1.xx.tar.gz (Release 1.04 or later)
Under UNIX or UNIX-like Operating System:
1) ANSI C and C++ compilers. GNU's gcc and g++ are desirable.
2) "make" command.
3) "xmkmf", "imake" and "makedepend" commands in X Window System
Version 11 Release 5 or later.
Under Windows NT Operating System:
1) ANSI C and C++ compilers. (Never use 16bit-compilers!)
Microsoft Visual C++ 2.0 or later is desirable.
3. Testing Environments
Machine: Sun Ultra 1 Creator Model 170E (compatible)
OS: Solaris2.5.1 (JP)
X Window System: X11R6.1 PL-1
Compilers: gcc,g++-2.7.2.2 with libg++-2.7.2
Machine: Sun SPARCstation 10 Model50 (compatible)
OS: SunOS4.1.3-JLE1.1.3_U1 (Solaris1.1.1)
X Window System: X11R6 PL-26
Compilers: gcc,g++-2.6.3
Machine: IBM RS6000/SP
OS: AIX4.1.4
X Window System: X11R5 (IBM)
Compilers: gcc, g++-2.6.3
Note: I have not tested the program under Windows NT.