Re: [Jocr-devels] Image Spam detection using gocr

[Stephen T. <st...@ne...>]

On 2006-11-16, Todd Lyons wrote:
> On Thu, Nov 16, 2006 at 05:46:27PM +1000, Stephen Thorne wrote:
> 
> >> > I'm working on some techniques to detect image spam by using gocr on
> >> > images attached to emails. You may or may not be familiar with this
> >> > style of spam, they often advertise stockscams or medications. 
> >> Before you go any further and reinvent the wheel, did you look at 
> >> FuzzyOCR ? http://fuzzyocr.own-hero.net/
> >Yep, I used quite a few of the ideas in the FuzzyOCR spamassassin
> >plugin, but because of constraints of the environment where my code must
> >run, I cannot use FuzzyOCR directly.
> 
> What are those constraints?

Linux mail server environment with a python based mail scanner. I'm
extending the email scanner to do the ocr based filtering.

> >I read the code thoroughly, but I'm not a perl programmer, I don't
> >believe it did any cleaning of the image.
> 
> What do you mean when you say "clean" the image?

Despeckling and removal of colour noise. 

-- 
Regards,
Stephen Thorne
Development Engineer

Scanned by the NetBox from NetBox Blue
(http://netboxblue.com/)