s-news
[Top] [All Lists]

Re: Gaussian smoothing

To: "Liaw, Andy" <andy_liaw@merck.com>, S-news list <s-news@lists.biostat.wustl.edu>
Subject: Re: Gaussian smoothing
From: Catalina Mesina <cmesina@cs.vu.nl>
Date: Mon, 25 Nov 2002 17:10:18 +0100
References: <51F9C42DA15CD311BD220008C707D81906FFC90A@usrymx10.merck.com>
In fact my  image is not only noise, but a matrix with low noise levels and one 
small area with higher values. However due to the noise in data is difficult to 
detect the peak. Therefore I need to smooth. 

Catalina

"Liaw, Andy" wrote:
> 
> If I understood you correctly, what you want to try is set the "span"
> argument in loess() to something fairly small (e.g., 0.1 or 0.05).  The
> default (0.75) is probably too large for smoothing images.
> 
> I'm not sure what you're expecting to see in your simulated example.  You
> simulated pure noise, so you should see (nearly) nothing after the smooth.
> 
> Andy
> 
> -----Original Message-----
> From: Catalina Mesina [mailto:cmesina@cs.vu.nl]
> Sent: Monday, November 25, 2002 10:54 AM
> To: S-news list
> Subject: [S] Gaussian smoothing
> 
> Dear all,
> 
> I have tried a 3D smoothing of my data, using "loess", or smooth but I did
> not get the right smoothed image.
> Suppose that my data is a matrix  "myim" 50x50 and at each point (x,y) in
> the grid I have an intensity value.
> 
> myim<-matrix(rnorm(2500),50,50)
> 
> x<-seq(1,50,1)
> y<-seq(1,50,1)
> grid<-expand.grid(x,y)
> myd<-data.frame(grid[,1],grid[,2],as.vector(myim))
> sm.myd<-fitted(loess(myd$X3~myd$X1*myd$X2,data=myd))
> image(matrix(myd$X3,50,50))
> 
> Then what I would like to have is a gaussian smoothed image (matrix) of
> myim.
> 
> I will appreciate every help answer from you.
> 
> Catalina
> --------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu.  To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> the BODY of the message:  unsubscribe s-news
> 
> ------------------------------------------------------------------------------
> Notice:  This e-mail message, together with any attachments, contains 
> information of Merck & Co., Inc. (Whitehouse Station, New Jersey, USA) that 
> may be confidential, proprietary copyrighted and/or legally privileged, and 
> is intended solely for the use of the individual or entity named in this 
> message.  If you are not the intended recipient, and have received this 
> message in error, please immediately return this by e-mail and then delete it.
> 
> ==============================================================================

<Prev in Thread] Current Thread [Next in Thread>