simple thresholding
in [Matlab]
|
From: Mike on 3 May 2010 23:17 hi just a simple question. >> a=magic(4) a = 16 2 3 13 5 11 10 8 9 7 6 12 4 14 15 1 >> [r,c]=find(a>13) r = 1 4 4 c = 1 2 3 Then I'd like only those data in r and c set to 255. How to do this please? Mike
From: Steven Lord on 4 May 2010 00:11 "Mike" <sulfateion(a)gmail.com> wrote in message news:b2ca8e16-9816-4f2d-ba64-3bead1381948(a)s13g2000prc.googlegroups.com... > hi > > just a simple question. >>> a=magic(4) > > a = > > 16 2 3 13 > 5 11 10 8 > 9 7 6 12 > 4 14 15 1 > >>> [r,c]=find(a>13) If you want to use this information to modify the locations in a, don't use the two output form of FIND -- in fact, don't use FIND at all. a = magic(4); a(a > 13) = -1; -- Steve Lord slord(a)mathworks.com comp.soft-sys.matlab (CSSM) FAQ: http://matlabwiki.mathworks.com/MATLAB_FAQ
From: Mike on 4 May 2010 01:37 On May 4, 12:11 pm, "Steven Lord" <sl...(a)mathworks.com> wrote: > "Mike" <sulfate...(a)gmail.com> wrote in message > > news:b2ca8e16-9816-4f2d-ba64-3bead1381948(a)s13g2000prc.googlegroups.com... > > > hi > > > just a simple question. > >>> a=magic(4) > > > a = > > > 16 2 3 13 > > 5 11 10 8 > > 9 7 6 12 > > 4 14 15 1 > > >>> [r,c]=find(a>13) > > If you want to use this information to modify the locations in a, don't use > the two output form of FIND -- in fact, don't use FIND at all. > > a = magic(4); > a(a > 13) = -1; > > -- > Steve Lord > sl...(a)mathworks.com > comp.soft-sys.matlab (CSSM) FAQ:http://matlabwiki.mathworks.com/MATLAB_FAQ Thank you very much. Another similar question but with more dimension: >> a=magic(3);b=a+1;c=a+2; >> d=cat(3,a,b,c) d(:,:,1) = 8 1 6 3 5 7 4 9 2 d(:,:,2) = 9 2 7 4 6 8 5 10 3 d(:,:,3) = 10 3 8 5 7 9 6 11 4 >> d(:,:,1)(d(:,:,1)>8)=999 ??? Error: ()-indexing must appear last in an index expression. >> dd=d(:,:,1);dd(dd>8)=999 dd = 8 1 6 3 5 7 4 999 2 Is there a way not to create a temporary "dd"? Mike
From: Walter Roberson on 4 May 2010 02:19 Mike wrote: >>> d(:,:,1)(d(:,:,1)>8)=999 > ??? Error: ()-indexing must appear last in an index expression. > >>> dd=d(:,:,1);dd(dd>8)=999 > > dd = > > 8 1 6 > 3 5 7 > 4 999 2 > > Is there a way not to create a temporary "dd"? Yup. d(vertcat(d(:,:,1)>8, false(size(d,1)*size(d,2),1))) = 999; Didn't say it would be short or easy to understand.
From: Walter Roberson on 4 May 2010 03:01
Walter Roberson wrote: > Mike wrote: > >>>> d(:,:,1)(d(:,:,1)>8)=999 >> ??? Error: ()-indexing must appear last in an index expression. >> >>>> dd=d(:,:,1);dd(dd>8)=999 >> >> dd = >> >> 8 1 6 >> 3 5 7 >> 4 999 2 >> >> Is there a way not to create a temporary "dd"? > > Yup. > > d(vertcat(d(:,:,1)>8, false(size(d,1)*size(d,2),1))) = 999; > > Didn't say it would be short or easy to understand. Or error-proof either. I should have written, d(vertcat(d(:,:,1)>8, false(prod(size(d)-[0 0 1]),1))) = 999; Though now that I think of it, d(find(d(:,:,1)>8)) = 999; should work as well. Both versions will break if fixed index being searched over does not happen to be 1, but I suspect more people would get a hint on changing the code from the longer version; the reason the second shorter one works for the first dimension only is more obscure... which is probably why I used the more general version of it in a posting a few years ago ;-) |