Thanks for the info and tip. I'll see if I can get a break for New
Years...grin
Scott
On Jan 3, 2006, at 9:30 AM, Jacob Schmude wrote:
Hi
The trick with Read Iris is to go into the settings menu and check
the "detect image orientation" option. Without this, it won't re-
orient the image if it's the wrong way. This improves recognition
results drastically and puts Read Iris on the level with
Textbridge, at least that's what I'd compare it to, which isn't
that bad at all. It costs $129.99, but I got it for $59.99 as they
were having a special discount over Christmas to anyone who
downloaded the trial for Mac within the month of December. That's
what really attracted me to it over Omnipage, $449 was the cheapest
I could find Omnipage for and I really didn't want to pay that much
for an OCR program that I couldn't try first--I've used Omnipage
for windows and like it but have no idea what the Mac version is
like. Someone here said they found it for $149, but I could only
find the upgrade for that price and not the full version.
If you were able to read the letters and numbers without an OCR
program, they must have set the alt or title tags of the image to
those letters and numbers, which is cool. It does compromise the
security of the verification somewhat but though.
On Jan 3, 2006, at 2:37 AM, Scott Howell wrote:
Funny, I was on the Roxio site trying to put a bug in their ear
about accessibility of Popcorn and there was a graphic with
letters/numbers in it. I didn't try running it through ocr, but it
did read the letter/numbers to me. Only problem was I couldn't
interact with it so had to slow speech way down and repeat it a
number of times.
While we're on the topic, I've been messing about with Read Iris,
but haven't gotten super fra yet. I'm curious what attracted you
or anyone to Read Iris over lets say Omni Page. At least with Read
Iris you can try before you buy and that does not appear possible
with Onipage.
Just curious and also I did not see the pricing on he Read Iris
site so am curious what it costs. Well I saw some pricing, but
assumed it was not for individuals.
tnx
Scott
On Jan 2, 2006, at 8:09 PM, Jacob Schmude wrote:
Hi guys
I thought I'd share a rather cool experience I had with Safari
today. Many web sites now require that you type the word, or
string of characters, in a picture to complete a sign up
procedure. I've found that, with Safari and VO, I can locate that
image and, by ctrl+click, pull it down to my PC and run it
through Read Iris and perform OCR on it. This doesn't always
work, but many times it does, and I'm able, through OCR, to find
the code in the image with no assistance. Note that the "only
navigate images with a description" option in VO must be turned
off for this to work, unless the image has an alt tag associated
with it. I have found no way whatsoever to even attempt this on a
windows system. I'll stress again--this does not always work, but
it's worth trying for those who have OCR programs on their Mac.
It seems to work about 50% of the time, and I'll be experimenting
further. This is certainly a step in the right direction in
dealing with these, and just one more reason why Mac rules!
Later
