odd question: how to "read" letters from an image?

[VMI]

Is it possible to "look" at an image and "read" the letters inside of that
image? For example, if I have several images of exactly the same size and
each has two lines of letters (Name, telephone) that always start and end in
the same exact position, is it possible to analyze that part of the image
and distinguish between the "light" pixels (the image background) and the
"dark" pixels (the pixels that compose the letters)? Then I can create a
string of those dark pixels.
I'm not looking for a definitive answer; I just want to explore some way of
doing this, even if it's very elemntary.

Thanks.

[Indie Pop]

VMI wrote:

Is it possible to "look" at an image and "read" the letters inside of that
image? For example, if I have several images of exactly the same size and
each has two lines of letters (Name, telephone) that always start and end
in the same exact position, is it possible to analyze that part of the
image and distinguish between the "light" pixels (the image background)
and the "dark" pixels (the pixels that compose the letters)? Then I can
create a string of those dark pixels.
I'm not looking for a definitive answer; I just want to explore some way
of doing this, even if it's very elemntary.

Thanks.

That's OCR ( Optical Character Recognition ).

Here's a .net/c# article on it:

http://www.codeproject.com/dotnet/simple_ocr.asp
--
Texeme
http://texeme.com

[Chad Z. Hower aka Kudzu]

"VMI" <vo******@yahoo.com> wrote in
news:#1**************@TK2MSFTNGP09.phx.gbl:

Is it possible to "look" at an image and "read" the letters inside of
that image? For example, if I have several images of exactly the same
size and each has two lines of letters (Name, telephone) that always
start and end in the same exact position, is it possible to analyze that
part of the image and distinguish between the "light" pixels (the image

Yes. You need to find an OCR lib.

--
Chad Z. Hower (a.k.a. Kudzu) - http://www.hower.org/Kudzu/
"Programming is an art form that fights back"

Make your ASP.NET applications run faster
http://www.atozed.com/IntraWeb/

[Landi]

Isn't that how Alcida people were communicating with each other?

"VMI" <vo******@yahoo.com> wrote in message
news:#1**************@TK2MSFTNGP09.phx.gbl...

Is it possible to "look" at an image and "read" the letters inside of that
image? For example, if I have several images of exactly the same size and
each has two lines of letters (Name, telephone) that always start and end in the same exact position, is it possible to analyze that part of the image
and distinguish between the "light" pixels (the image background) and the
"dark" pixels (the pixels that compose the letters)? Then I can create a
string of those dark pixels.
I'm not looking for a definitive answer; I just want to explore some way of doing this, even if it's very elemntary.

Thanks.

[Chad Taylor]

Landi wrote:

Isn't that how Alcida people were communicating with each other?

I thought that was steganography...shoving encrypted information into
JPEGs or other compressed picture formats. I could be wrong, though.

[Landi]

Thanks Cad,
I really wasn't expecting an answer but thanks for the info.
I don't know anything about it so I will take your word for it.

-- Landi

"Chad Taylor" <ca****@gmail.com> wrote in message
news:LN*******************@fe40.usenetserver.com.. .

Landi wrote:

Isn't that how Alcida people were communicating with each other?

I thought that was steganography...shoving encrypted information into
JPEGs or other compressed picture formats. I could be wrong, though.

[Predictor]

VMI wrote:

Is it possible to "look" at an image and "read" the letters inside of that image? For example, if I have several images of exactly the same size and each has two lines of letters (Name, telephone) that always start and end in the same exact position, is it possible to analyze that part of the image and distinguish between the "light" pixels (the image background) and the "dark" pixels (the pixels that compose the letters)? Then I can create a string of those dark pixels.
I'm not looking for a definitive answer; I just want to explore some way of doing this, even if it's very elemntary.

This is sometimes referred to as "OCR" (optical character recognition).
Your options are to either use an existing OCR libray or build your
own recognizer.
-Will Dwinnell
http://will.dwinnell.com