I always knew I'd find a use for this £5k worth of software
From the Office Art challenge. See all 328 entries (closed)
( , Tue 5 Jun 2007, 9:53, archived)
From the Office Art challenge. See all 328 entries (closed)
( , Tue 5 Jun 2007, 9:53, archived)
Cor!
High quality this morning!
It simply cannot last
*2nd click of the day*
( ,
Tue 5 Jun 2007, 9:54,
archived)
It simply cannot last
*2nd click of the day*
Blimey.
That's woo.
I would click it, but it might get above mine. Arses. The dilemna.
*closes eyes*
*clicks*
( ,
Tue 5 Jun 2007, 9:56,
archived)
I would click it, but it might get above mine. Arses. The dilemna.
*closes eyes*
*clicks*
that will be getting clicked
because I like it, and the facility exists to express this 'like' through the click of a link. Isn't technology dull!?
( ,
Tue 5 Jun 2007, 9:57,
archived)
Correctamundo
It is *Maths*, it was Maths and it always will be *Maths*. Math - pah.
( ,
Tue 5 Jun 2007, 10:00,
archived)
it's going to get pretty messy in here if we
all start spitting en masse
( ,
Tue 5 Jun 2007, 10:04,
archived)
I'm standing on a chair
and have am using a broomstick to type with
( ,
Tue 5 Jun 2007, 10:17,
archived)
I just love stuff
That's more better than other stuff and this is more better than some other stuff.
( ,
Tue 5 Jun 2007, 10:00,
archived)
Wow that's a bit good
I'm still struggling to plot a CDC. I'm sure I had it saved somewhere but seem to have deleted it.
( ,
Tue 5 Jun 2007, 11:11,
archived)
I considered doing something similar using Surfer
but I'm glad I didn't because it wouldn't have been as good as yours!
( ,
Tue 5 Jun 2007, 11:38,
archived)
matlab abuse
this is brilliant! so good infact im interested as to how you did it, could you post the code?
( ,
Tue 5 Jun 2007, 17:57,
archived)
Cheers, I'll post it soon.
It's fairly simple really 'cos Matlab does all the hard work. Basicly, make a monochrome low-res (eg 128pix long) image in paint/Gimp, etc - i.e. black text (bold works good) white BG. Save it as BMP or similar.
You can then open in Matlab using imread(). There's a bit of jiggery-pokery to convert the numbers before plotting with mesh(), then it's just fiddling around with view angles. Matlab is ace! And 20yrs ago I failed maths at A'level with a 'U' grade. It'll be tomorrowish B4 I can post.
( ,
Tue 5 Jun 2007, 18:16,
archived)
You can then open in Matlab using imread(). There's a bit of jiggery-pokery to convert the numbers before plotting with mesh(), then it's just fiddling around with view angles. Matlab is ace! And 20yrs ago I failed maths at A'level with a 'U' grade. It'll be tomorrowish B4 I can post.
Hell's Bells. I'd better get back to work now.
Original image: (after editing so that lightest = highest points)
Super happy enhanced fourier transform of Robbie Williams's song: "Angels".
% Matlab code for useful happy work (JuanTheMan)
[X,m]=imread('domo.bmp'); % load image
% next, flip upsidedown (matlab quirk) and take every second point in image
x=flipud(double(X(1:2:141,1:2:141,1)));
% convert char to double (we're just using single colour channel (1 = red)
% 141 is image size. you could skip 3 or more places for big images
n=randn(size(x))*10; % generate some noise
f=find(x < 30); % find places where current image is low brightness
x(f)=x(f)+n(f); % add jaggedy background noise
m=mesh(x); % plot
colormap hot % copper is good also
shading interp % looks good
set(m,'facecolor',[0.68 0.45 0]); % colour inbetween mesh
view(-12,65) % set view angle
axis off % get rid of nasty axes
set(gca,'projection','perspective'); % looks better than orthographic
( ,
Wed 6 Jun 2007, 10:13,
archived)
Super happy enhanced fourier transform of Robbie Williams's song: "Angels".
% Matlab code for useful happy work (JuanTheMan)
[X,m]=imread('domo.bmp'); % load image
% next, flip upsidedown (matlab quirk) and take every second point in image
x=flipud(double(X(1:2:141,1:2:141,1)));
% convert char to double (we're just using single colour channel (1 = red)
% 141 is image size. you could skip 3 or more places for big images
n=randn(size(x))*10; % generate some noise
f=find(x < 30); % find places where current image is low brightness
x(f)=x(f)+n(f); % add jaggedy background noise
m=mesh(x); % plot
colormap hot % copper is good also
shading interp % looks good
set(m,'facecolor',[0.68 0.45 0]); % colour inbetween mesh
view(-12,65) % set view angle
axis off % get rid of nasty axes
set(gca,'projection','perspective'); % looks better than orthographic