Fax configuration

"Robert J. Brown" (rj@eli.elilabs.com)
Mon, 29 Dec 1997 22:14:01 +0100


>>>>> "Gert" == Gert Doering <gert@greenie.muc.de> writes:

 Gert> BTW: Does anybody know some kind of OCR software in the
 Gert> "free software pool"?

Take a look at:

 http://www.dlr.de/fresh/unix/src/misc/xocr_4_2.deutsch.tar.gz

This is a package I looked at over a year ago, but never did anything
with it -- no time. :-(

*From the Xocr distribution:

 WSA-algorithm: (Degree-cut-analysis)
 ===================================

 Every character is zoomed to a fixed size. (here: 16x16 pixel)
 Parallel lines are layerd over the picture. (here: 16 lines)
 Now, all Pixels which are set in the picture and placed on a line are counted.
 (here: 0..24 Points)
 After that the lines will be turned by a fixed degree-value and again calculated like above.
 All lines will be turned step by step until 180 degrees are reached. 
 We have 128 values calculated.
 These values are coresponding with a 128 dimensional space. 

 Now , all trained characters are points in this space. The lowest 
 distance between the character we want to know and all of the trained characters will be calculated.
 If this distance is very small the character will be accepted as well-recognized, otherwise
 the user is consulted if it was right detected !


 REASON FOR DOING THIS STUFF:
 ============================

 1. Learning by doing - training myself
 2. Supporting the LINUX-operating system
 3. Just for fun !

You may also like to see a simple example of a neural network that
does the kind of trained recognizing spoken of above. This may be
seen in the April 1987 Dr. Dobb's Journal. On of these days, I will
have to get around to putting that paper on-line.

-- 
-------- "And there came a writing to him from Elijah" [2Ch 21:12] --------
Robert Jay Brown III rj@eli.elilabs.com http://www.elilabs.com 1 847 705-0424
Elijah Laboratories Inc.; 37 South Greenwood Avenue; Palatine, IL 60067-6328
----- M o d e l i n g t h e M e t h o d s o f t h e M i n d ------
.