[tehaaron]

This is really neat. I was playing with it on pictures of street signs and buildings and realized that if I select some text and then do ctrl+a it tried to select everything it thought was text...Then I used right click > translate > reprint to see what it thought each thing was.

Here is the picture: http://thesuperslice.com/wp-content/uploads/2012/04/downtown...

And the text outcome - found it most interesting what symbols it thought it recognized:

lam

on-0'0

s.

Ic 0on

§-i-

I-*-

-unm

-$3.»;

o

G %T1

00-O

. o C-‘7' H ' .-.”-." «'~3;

.35

$16 O-O

‘D Q-=¢1

‘-M

km“

‘MIMI

DOW:

TLDR

D001”

'."'IIu

ff"

)0‘

\\

,¢-.5 ,:~L.

r/J