text
stringlengths
0
4k
With a photocopier, enlarge or reduce the picture so that it is
close to the best size (as stated above). Then, use a wide black
marker to "highlight" the lines and contours (this may not be
necessary if the picture came from a colouring book). If there is
a background image that you want to remove, just cut and tape white
paper over those areas or just cut out the undesired areas. This
is important since the "junk" you would otherwise leave in would
eventually interfere with the grid.
When the lines are all really dark and just about all the
background has been removed you can photocopy the new picture on
the grid. You may have to experiment a few times to get the
picture to come out the same as the original on an upside up grid.
When you have such an image you can now bring up the electronic
grid. Work at it from left to right and from top to bottom. Put
your word processor on typeover and space over the characters of
the first line until you get to a character that is partially or
completely covered by a dark line on the paper grid. Use the 1 to
5 and the A to P to locate the correct position. When you get to
such a character, try to find a character that resembles the shape
of the line and type that in. Ex: '8' for a character that is
totally covered, 'a' for a character that has the bottom half
covered, etc. This is actually starting the smoothing process at
the same time you are converting from paper to screen. Remember
that shading complicates things unnecessarily, so ignore it
completely.
Repeat this process for all the lines of the grid until the picture
is completely transferred to electronic form. You don't need to be
an artist do this. You just need to have some time on your hands.
Of course at this point the ascii picture is still slightly rough,
but at least you have the right proportions, etc. Now you could
try to smooth it out by using the "weight distribution" technique
described above and perhaps a printed version of one of the ascii
pictures I have smoothed (for concrete examples).
It may sound very complicated at first, but just to give you an
example, the ying yang I did (a few people informed me that I had
forgotten the eyes):
.,ad88888888baa,
,d8P""" ""9888ba.
.a8" ,ad88888888888a
aP' ,88888888888888888a
,8" ,88888888888888888888,
,8' (888888888( )888888888,
,8' `8888888888888888888888
8) `888888888888888888888,
8 "8888888888888888888)
8 `888888888888888888)
8) "8888888888888888
(b "88888888888888'
`8, (8) 8888888888888)
"8a ,888888888888)
V8, d88888888888"
`8b, ,d8888888888P'
`V8a, ,ad8888888888P' Normand
""88888888888888888P" Veilleux
""""""""""""
took just over half an hour using that technique. I hope a few of
you will be interested enough to try it out. If you do, let us
know how it goes.
EXPERIMENT: Would someone with access to a scanner and the
GIF2ASC program (or an OCR) try the following:
take a picture from a colouring book, scan it and
then convert it to ascii. Alternatively, tell your
scanner that what you are scanning is text, if you
have an OCR. I am under the impression that the
results would be quite presentable. Of course,
sometimes theory and practice differ considerably.
Please post your findings.
From: Normand Veilleux <nveilleu@emr1.emr.ca>
Date: 27 Jun 1994 20:26:35 -0500
I have seen several ascii versions of the Mona Lisa circulating in
alt.ascii-art but I was not happy with any of them. They all
appeared to be converted GIFs and even the best one, which was
distinctly superior to all the others, only looked decent when I
stood back at least 5 feet from it. It also had the drawback of
being stretched vertically. So I decided to make my own.
I also wanted to take this opportunity to give a concrete example
of how the special grid technique that I use, accelerates and
simplifies the creation of many ASCII drawings. I wanted to
explain, as much as possible, what was involved in drawing by
"hand" the ASCII version of a relatively complex picture while
starting from scratch.
I have described the grid technique in a previous post (in Jan.
'94), but a recapitulation is necessary before going any further.
Basically, the grid was nothing more than a series of characters
that filled up the area taken by the drawing. I realized that the
characters composing the grid would be more useful if they were
chosen to help pinpoint where a particular portion of the drawing