home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ShareWare OnLine 2
/
ShareWare OnLine Volume 2 (CMS Software)(1993).iso
/
demos
/
russoft1.zip
/
INTERREC.ZIP
/
README.ENG
< prev
next >
Wrap
Text File
|
1992-07-12
|
19KB
|
392 lines
Optical character recognition system InterRec
Version 1.0
User's Manual
1. Purpose of the InterRec system
The InterRec system of optical character recognition is
intended for the recognition of texts entered into a personal
computer with the aid of any kind of equipment for the input
of optical information.
The source material for InterRec's work is a graphics
file of the text aquired with the aid of a device for the
input of optical information and stored on a hard disk or a
diskette in PCX format (two colors).
The result of the system's work is a file containing the
ASCII representation of the source graphics image of the text.
2. General information
InterRec is an omnifont system. This means that it does
not need any kind of setup or teaching for the recognition of
a specific font.
The text being processed must me written in a graphics
file with a resolution not less that 200 pixels per inch. A
resolution of 300 dots per inch is recommended, which
increases the accuracy of recognition (with an insignificant
increase in processing time).
The character height must be within the following limits:
for a resolution of 200 pixels per inch: 6 to 50 points
(2.2 to 19 mm);
for a resolution of 300 pixels per inch: 4 to 30 points
(1.5 to 11 mm).
The system has modes of operation that are
user-selectable: ENGLISH and RUSSIAN for the recognition of
texts in the English and Russian languages respectively.
When working in the ENGLISH mode InterRec recognizes the
capital and small letters of the English alphabet, numbers and
the following symbols: ! ? . , : ; " / $ % & * ( ) - + = in
vertical and italics, in normal and wide lines.
When working in the RUSSIAN mode InterRec recognizes the
capital and small letters of the Russian alphabet, numbers and
the following symbols: ! ? . , : ; " / % * ( ) - + = in
vertical, in normal and wide lines.
In both modes InterRec recognizes letters, numbers and
the above listed special symbols printed in proportional fonts
on a laser printer, on a dot matrix printer (in
near-letter-quality mode), on a typewriter or on a typesetting
machine.
To guarantee that the system works correctly, there must
not be any characters other than those listed above, or any
underlines. The presence of any tables or such graphic
elements such as drawings, sketches or graphs may lead to the
rejection of the entire page for character recognition.
The accuracy of recognition depends on the quality of the
graphics copy of the text and with high-quality of the source
material attains 99%. The accuracy of recognition decreases
with the presence of a large number of joined characters, and
also with the presence of breaks in the representation of
characters.
The time for recognizing one page (2000 characters) does
not exceed 1.5 minutes (with a resolution of 300 pixels per
inch and using an IBM PC AT with a clock speed of 12 Mhz).
The requirements for a personal computer are any machine
that is compatible with an IBM PC XT/AT with not less than 512
Kbytes of real memory and an EGA or VGA video adapter. A mouse
makes the work with the system more comfortable, though it is
not a required device. The operating system must be MS-DOS
version 3.3 or higher.
3. The complete set
A 5.25" DS/DD diskette containing:
the program interrec.exe (intrdemo.exe for demo version)
and the files needed to work with it: eng.trr and
rus.trr; the instruction files for the user in English
(readme.eng) and in Russian (readme.rus); files that
let you read the files containing recognized Russian
texts on the screen (egarfont.com, vgarfont.com,
egarfont.fnt, vgarfont.fnt).
The demo version of InterRec system does not write the
recognized text into the file. Therefore you do not need the
files egarfont.com, vgarfont.com, egarfont.fnt, vgarfont.fnt.
4. How to work with InterRec
The EXE file used to load and start the program is called
interrec.exe.
After startup, a window containing some brief information
about the system is displayed on the screen. To enter the main
menu you must press any key on the keyboard or any mouse
button.
The main menu displays the system control functions which
are divided into four groups corresponding to the following
items of the main menu: File, Screen, Language, Start. To
invoke a function of the main menu you should either
(a) place the highlighted bar on the selected function
using the left and right arrow keys, and press the
Enter key;
(b) move the mouse cursor to the necessary item and press
twice the left button of the mouse (after the first
click the highlighted bar will move to the necessary
item).
Submenu items, falling out of the main menu, are selected
in the same way but using the up and down arrow keys. To
return from submenu items to the main menu you should either:
(a) press Esc;
(b) press the right button of the mouse if its cursor is
within the submenu, or press the left button if the
cursor is outside this area.
Now, let us discuss one by one the functions of the main
menu.
4.1. File
The purpose of this function is to select both a name for
the file which stores the graphic image of the text being
recognized, and a name of the file which will store the ASCII
representation of the text. Here the exit control is also
located. Accordingly, the submenu has three items: Load.PCX,
Load.TXT, Exit.
4.1.1. Load.PCX.
This item is used to select the name of the file
containing graphics image of the text being recognized. This
file must have the .PCX extension.
When you select this option, a dialog box appears in the
center on the screen in which you can see: an input line; a
scroll window, where the file to be recognized is searched;
and buttons labeled Ok and Cancel.
If using a keyboard, you must continuously press the Tab
key to transfer control between the input line, Ok and Cancel
buttons.
While entering the system, the input line displays the
name of the directory where InterRec is located, and a .PCX
searching mask.
The input line has 7 lines. A line designated as .. \
lets you call up the directory of the higher level. The
remaining lines contain filenames of the current directory
corresponding to the mask , i. e. filenames with . PCX
extension, and also subdirectory names. If these filenames and
subdirectory names are rather numerous and cannot be displayed
simultaneously in the given lines, use the scrolling function
to view them all. You should either press the down arrow key
or move the mouse cursor to the down arrow of the scroll bar
and press the left button of the mouse.
Reverse scrolling is performed either with the up arrow
key or by moving the mouse cursor to the up arrow of the
scroll bar and pressing the left button of the mouse.
You may also enter the directory you need (and change the
current drive if necessary) by direct input of the pathname in
the input line. To open this line for writing down, you can
use the following ways:
(a) if using a keyboard, press the Tab key until a cursor
in the form of an underline appears in this line;
(b) if using a mouse, move the mouse cursor to this line
and press the left button of the mouse.
The underline cursor points to the position in the line
where you will start typing. You can move the cursor along the
line using the left and right arrow keys. You can type both in
replace and insert modes. The modes are switched with the
Insert key. Finally, you must press the Enter key.
There are two ways to select a filename for recognition:
(a) move the highlighted bar using the up and down arrow
keys and press the Enter key;
(b) move the mouse cursor to the appropriate item and
press the left button of the mouse twice (after the
first click the highlighted bar will move to the
appropriate item).
The selected filename will be displayed on the input
line. To proceed, you have to confirm the correctness of your
choice. This is performed in the following ways:
(a) if using a keyboard, activate the Ok button with the
aid of the Tab key, as stated above (doing this, the
letters "Ok" will change to red), and finally press
the Enter key;
(b) if using a mouse, move its cursor to the Ok button
and press the left button of the mouse.
After this the dialog box will disappear, and the
following record in the upper line of the screen will appear:
Recognition from full pathname and name of the file
selected for recognition to full pathname and filename for
storing ASCII code of the recognized text.
The system itself prompts you to save the recognized text
in a file in the same directory as the original file, and
which retains the name of the original but with a new
extension: .TXT. If the full pathnames are too long to be
contained within one line, only the head parts of the
pathnames (that is, the name of the current drive) as well as
the tail parts of it (that is, the filenames) are displayed.
In addition, in the window in the lower part of the
screen where there is a scroll bar, the image of the graphics
file selected for recognition (or its fragment) will appear.
Shading in horizontal and/or vertical scales indicates
that the graphics file is too large to be fully displayed in
the window. Nevertheless, the system lets you view the
remaining parts of the graphics file. The displayed area is
moved along the file in the following ways:
(a) if using a keyboard, press the left, right, up and
down keys;
(b) if using a mouse, move the pointer to the desired
scroll bar arrows and pressing the left button.
To switch control from the mouse to the keyboard and vice
versa you should use the Tab key.
Unshaded areas on the scroll bar show the location of the
displayed fragment in relation to the whole page.
If you are satisfied with the name and location of the
ASCII file which have been proposed to you by the system, your
work with the File item of main menu is completed. If the
choice proposed to you by the system does not suit you, you
should select the Load.TXT second option.
Job in the dialog box can be cancelled prior to selecting
a file for recognition. It can be done in the following ways:
(a) if using a keyboard, activate the Cancel key using
the Tab key until the word "Cancel" turns red,
and then press Enter;
(b) if using a mouse, move its arrow to the Cancel button
and press the left button of the mouse.
4.1.2. Load.TXT
This item is used to select the filename which will
contain the recognized text.
The selection of the appropriate file for the ASCII
representation is done following the same sequence of
operations as that for the graphics file. Extension .TXT
should be used instead as a search mask , and the current
directory must be the one where the file to be recognized was
selected. The mask can be changed by inputting the required
extension in the input line as in 4.1.1. After you have
approved the selection, the pathname and filename will be
displayed in the upper line of the screen.
ASCII representation can be stored in a newly opened
file. To do so you must type its name in the input line, press
the Enter key and confirm your choice (Ok) as stated above.
The pathname and the filename will be displayed in the upper
line of the screen.
After selecting the files for recognition and saving the
ASCII representation, you should check the correctness of your
choice. If you are satisfied, your work with the File function
is completed; otherwise, you need to return to the appropriate
function (Load. PCX or Load. TXT) and make a new selection
following the above sequence of operations.
4.1.3. Exit
Entering this item results in exiting from the InterRec
system.
4.2. Screen
This item is used to select the form in which information
is displayed on the screen in the recognition process. Thus
the submenu has three items: Graphics, All, ASCII.
4.2.1. Graphics
When item Graphics is selected, the window for viewing
graphics file increases in size. In this enlarged window a
graphics file can be viewed in the same way as in 4.1.1.
During recognition process the graphics file will be scrolled
in the window. The recognition results will not be displayed.
4.2.2. All
With choosing this item the windows do not change their
positions on the screen. In the recognition process the lower
window displays graphics file being scrolled, and the upper -
the resulting text file.
This mode is set by default if the Screen item of the
main menu is not invoked.
4.2.3. ASCII
Upon entering this item the whole screen will be used for
a large window for demonstration of recognition results, i. e.
the resulting ASCII file. This mode is faster than the
previous one.
4.3. Language
This item lets you switch operating modes of the
recognition program with regards to the language of the source
text. The selected mode is marked with a tick. English is the
default language.
The text editor must have an appropriate screen driver to
let you read the files containing the recognized Russian text.
If your text editor has no such driver, use the enclosed
drivers that come with the InterRec system. To use them, you
must start egarfont.com (with an EGA video adapter) or
vgarfont.com (with a VGA video adapter) before you work with
the editor. The character coding used by the system is given
in Appendix.
4.4. Start
This item of the main menu has no submenu and is used to
start the recognition program.
If a file selected for ASCII representation already
exists, the following message will appear on the screen:
ASCII file already exists
New Add Cancel
While using a keyboard, the required item is selected
with the left and right arrow keys followed by pressing the
Enter key. The selected item changes to red.
If you chose the New option, the previous contents of the
text file will be lost. If the Add item is selected, the
recognition result is added to the contents of the file. This
mode of operation can be effectively used if you want to
combine into a single file the results of processing multiple
sheets. As soon as you select an item, recognition will begin.
Information will be displayed on the screen as it was
described in 4.2. The unshaded area in the vertical scroll bar
corresponds to the page fragment being recognized.
Recognition will not start if the Cancel item was chosen.
Program execution can be cancelled by pressing Esc.
Upon the program termination a message will appear on the
screen: Successful recognition.
5. Possible User Errors
The system provides you with several of messages which
indicate possible incorrect actions of the user.
5.1. File can't be opened
This message appears when a non-existing file is selected
for recognition.
5.2. Invalid drive or directory
The message indicates that while selecting the file, the
user has typed wrong name of the drive or directory.
5.3. This disk can't be set
The message indicates that, when trying to open a file,
the user has typed the name of a non-existing drive.
5.4. Data reading error
This message appears when an error occurs in reading from
the disk.
5.5. Invalid PCX format
The message indicates that the file selected for
recognition does not meet the requirements of PCX format,
version 2.8 and lower. In this case you need to form the file
for recognition anew.
5.6. Not compatible format
The message testifies to the fact, that a file with color
or gray-scale image, not supported by current system version,
was selected for recognition.
5.7. Graphic file not loaded
This message appears at the attempt to start recognition
when the graphics file is not yet selected.
5.8. File .TRR can't open
This message appears during recognition when there are no
rus.trr or eng.trr files in the current directory.
5.9. Insufficient memory
The message warns you that the RAM space for the program
is not enough. In this event, you need to delete, if possible,
some resident routines the system can do without.
5.10. Recognition is completed
The message is possible in the following cases: the file
contains some non-text elements of great vertical size; the
size of the character is too great; two or more adjacent lines
have merged or overlapped; the lines of the inputted image and
the higher (lower) edge of the sheet of paper are far from
being parallel.
In addition to the above mentioned messages, some other
error messages may appear on the screen (however, this is
highly improbable). In such a case, you need to consult the
designers of the system or their representative in your
country.
Appendix
Alternative ASCII code used by InterRec system
┌───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┐
│ Ç │ ü │ é │ â │ ä │ à │ å │ ç │ ê │ ë │ è │ ï │ î │ ì │ Ä │ Å │
│128│129│130│131│132│133│134│135│136│137│138│139│140│141│142│143│
├───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┤
│ É │ æ │ Æ │ ô │ ö │ ò │ û │ ù │ ÿ │ Ö │ Ü │ ¢ │ £ │ ¥ │ ₧ │ ƒ │
│144│145│146│147│148│149│150│151│152│153│154│155│156│157│158│159│
├───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┤
│ á │ í │ ó │ ú │ ñ │ Ñ │ ª │ º │ ¿ │ ⌐ │ ¬ │ ½ │ ¼ │ ¡ │ « │ » │
│160│161│162│163│164│165│166│167│168│169│170│171│172│173│174│175│
├───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┤
│ α │ ß │ Γ │ π │ Σ │ σ │ µ │ τ │ Φ │ Θ │ Ω │ δ │ ∞ │ φ │ ε │ ∩ │
│224│225│226│227│228│229│230│231│232│233│234│235│236│237│238│239│
├───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┼───┤
│ ≡ │ ± │ ≥ │ ≤ │ ⌠ │ ⌡ │ ÷ │ ≈ │ ° │ ∙ │ · │ √ │ ⁿ │ ² │ ■ │ │
│240│241│242│243│244│245│246│247│248│249│250│251│252│253│254│255│
└───┴───┴───┴───┴───┴───┴───┴───┴───┴───┴───┴───┴───┴───┴───┴───┘