Re: Clsuter analysis wiht IDL

[Gonzalo Rojas <grojasy(at)puc.cl>]

Dorthe Wildenschild wrote:

 Benno,

Thanks for the input.
You suggested:

use something like
ref_img=REFORM(image,1,658*658)

Dear Dorthe:

  the error is that 658*658 is greater than 65536 (max integer)... you must put 658l*658 in reform procedure (the "l" letter is necessary to convert to long type numbers)...

  best regards...

Gonzalo Rojas
Medicine School
Catholic University of Chile

 
weights = clust_wts(ref_img, n_clusters=3)
cls=cluster(ref_img,weights,n_cluster=3)
tvscl,reform(cls,658,658)

but when computing the weights it goes bad
Program caused arithmetic error: Floating underflow
and I get an array that is [1,3] and the values are
5.60519e-045
5.60519e-045
5.60519e-045

tried converting to float before calculating weights, but same result. Any suggestions anyone? realize per Craig's post that cluster analysis isn't what IDL has been used most for...
D
 

begin:vcard 
n:Rojas;Gonzalo
tel;fax:(562) 6333266
tel;home:(562) 2292145
tel;work:(562) 6863276
x-mozilla-html:FALSE
org:Catholic University of Chile;Medicine School, Nuclear Medicine Service
adr:;;Marcoleta 367;Santiago;-;6510260;CHILE
version:2.1
email;internet:grojas@computer.org
fn:Gonzalo Rojas
end:vcard