% fields, Tools for spatial data
% Copyright 2004-2011, Institute for Mathematics Applied Geosciences
% University Corporation for Atmospheric Research
% Licensed under the GPL -- www.gpl.org/licenses/gpl.html

\name{image.cov}
\alias{stationary.image.cov}
\alias{Exp.image.cov}
\alias{Rad.image.cov}
\alias{matern.image.cov}
\title{
Exponential, Matern and general covariance functions for 2-d 
gridded locations. 
}
\description{
Given two sets of locations defined on a 2-d grid efficiently multiplies a 
cross covariance with a vector.  Exp.image.cov and matern.image.cov will be  
depreciated functions and are replaced by stationary.image.cov.
}
\usage{
stationary.image.cov(ind1, ind2, Y, cov.obj = NULL, setup = FALSE, 
grid, M=NULL,N=NULL, Covariance="Matern", Distance="rdist",...) 

Exp.image.cov(ind1, ind2, Y, cov.obj = NULL, setup = FALSE, grid, ...)

Rad.image.cov(ind1, ind2, Y, cov.obj = NULL, setup = FALSE, grid, ...)

matern.image.cov(ind1, ind2, Y, cov.obj = NULL, setup = FALSE, grid,
M=NULL,N=NULL,...)

}
\arguments{
\item{ind1}{
Matrix of indices for first set of locations this is a two column matrix
where each row is the row/column index of the image element. If missing
the default is to use all grid locations.
}
\item{ind2}{
Matrix of indices for second set of locations. If missing this is taken to
be ind2. If ind1 is missing ind2 is coerced to be all grid locations.
}
\item{Y}{
Vector to multiply by the cross covariance matrix. Y must be the same
locations as those referred to by ind2.
}
\item{cov.obj}{
A list with the information needed to do the multiplication by
convolutions. This is usually found by using the returned list when
setup=T.
}
\item{setup}{
If true do not do the multiplication but just return the covariance object
required by this function. 
}
\item{grid}{
A grid list giving the X and Y grids for the image. (See example below.)
This is only required if setup is true. 
}
\item{M}{
Size of x-grid used to compute multiplication (see notes on image.smooth 
for details) by the FFT. If NULL, the default for M is the largest power 
of 2 
greater than or equal to 2*m where m= length( grid\$x). 
This  will give an exact 
result but  smaller values of M will yield an approximate, faster result. 
}

\item{N}{Size of y-grid used to compute multiplication by the FFT.}
\item{Covariance}{Covariance function that is apllied to the distance 
between locations. (see stationary.cov) Default is the Matern model.
(with smoothness=.5) the expontential. }
\item{Distance}{
Distance function applied to locations. Default is Euclidean distance. 
Another choice is "rdist.earth", great circle distnace for lon/lat 
coordinates. }
\item{\dots}{
Any arguments to pass to the covariance function in setting up the
covariance object. This is only required if setup is TRUE.
For the "Matern" the arguments are theta ( the range default=1), 
and smoothness (default=.5 giving the exponential).
theta can be a matrix reflecting a rotation and scaling of 
coordinates. See stationary.cov for details. }
}
\value{
A vector that is the multiplication of the cross covariance matrix with
the vector Y.
}
\details{
This function was provided to do fast computations for large numbers of
spatial locations and supports the conjugate gradient solution in
krig.image. In doing so the observations can be irregular spaced
but their coordinates must be 2-dimensional and be restricted to grid
points.
(The function as.image will take irregular, continuous coordinates and
overlay a grid on them.) 

Returned value: If ind1 and ind2 are matrices where nrow(ind1)=m and
nrow(ind2)=n then the cross covariance matrix, Sigma is an mXn matrix
(i,j) element is the covariance between the grid locations indexed at
ind1[i,] and ind2[j,]. The returned result is Sigma\%*\%Y. Note that one can
always recover the coordinates themselves by evaluating the grid list at
the indices.  e.g.  cbind( grid\$x[ ind1[,1]], grid\$y[ind1[,2])) will give
the coordinates associated with ind1. Clearly it is better just to work
with ind1!
  
Functional Form: Following the same form as Exp.cov stationary.cov for 
irregular locations, the covariance is defined as phi( D.ij) where D.ij is
the Euclidean distance between x1[i,] and x2[j,] but having first been
scaled by theta. Specifically,

D.ij = sqrt(  sum.k (( x1[i,k] - x2[j,k]) /theta[k])**2 ).

See \code{Matern} for the version of phi for the Matern family. 
 
Note that if theta is a scalar then this defines an isotropic covariance
function.  

Implementation: This function does the multiplication on the full
grid efficiently by a 2-d FFT. The irregular pattern in Y is handled by
padding with zeroes and once that multiplication is done only the
appropriate subset is returned. 

As an example assume that the grid is 100X100 let big.Sigma denote the big
covariance matrix among all grid points ( If the parent grid is 100x100
then big.Sigma is 10K by 10K !) Here are the computing steps:

temp<- matrix( 0, 100,100)

temp[ ind2] <- Y

temp2<- big.Sigma\%*\% temp

temp2[ind1]

Notice how much we pad with zeroes or at the end throw away!
Here the matrix multiplication is effected through convolution/FFT tricks
to avoid creating and multiplying big.Sigma explicitly. It is often faster
to multiply the regular grid and throw away the parts we do not need then
to deal directly with the irregular set of locations.

Note: In this entire discussion Y is treated as vector. However if
one has complete data then Y can also be interpreted as a image matrix
conformed to correspond to spatial locations. See the last example for
this distinction. 
  
}
\seealso{
  smooth.2d, as.image, krig.image, stationary.cov 
}
\examples{
# multiply 2-d isotropic exponential with theta=4 by a random vector 

junk<- matrix(rnorm(100*100), 100,100)

cov.obj<- stationary.image.cov( setup=TRUE, 
             grid=list(x=1:100,y=1:100),theta=8) 
result<-  stationary.image.cov(Y=junk,cov.obj=cov.obj)

image( matrix( result, 100,100)) # NOTE that is also a smoother!

# to do it again, no setup is needed 
#  e.g. 
#  junk2<- matrix(rnorm(100**2, 100,10))
#  result2<-  stationary.image.cov(Y=junk2, cov.obj=cov.obj)

# generate a grid and set of indices based on discretizing the locations
# in the precip dataset

 out<-as.image( RMprecip$y, x= RMprecip$x)
 ind1<- out$ind
 grid<- list( x= out$x, y=out$y)

#
# discretized x locations  to use for comparison
  xd<- cbind( out$x[ out$ind[,1]], out$y[ out$ind[,2]] )

# setup to create cov.obj for exponential covariance with range= 1.25

 cov.obj<- stationary.image.cov( setup=TRUE, grid=grid, theta=1.25) 

# multiply covariance matrix by an arbitrary vector
 junk<-  rnorm(nrow( ind1))
 result<- stationary.image.cov( ind1, ind1, Y= junk,cov.obj=cov.obj)

# The brute force way would be  
#   result<- stationary.cov( xd, xd, theta=1.25, C=junk)
# or 
#   result<- stationary.cov( xd, xd, theta=1.25) %*% junk
# both of these take much longer 


# evaluate the covariance between all grid points and the center grid point
 Y<- matrix(0,cov.obj$m, cov.obj$n)
 Y[32,32]<- 1
 result<- stationary.image.cov( Y= Y,cov.obj=cov.obj)
# covariance surface with respect to the grid point at (32,32)
# 
# reshape "vector" as an image
 temp<-  matrix( result, cov.obj$m,cov.obj$n)
 image.plot(cov.obj$grid$x,cov.obj$grid$y, temp)
# or persp( cov.obj$grid$x,cov.obj$grid$y, temp) 

# check out the Matern 
 cov.obj<- stationary.image.cov( 
             setup=TRUE, grid=grid, theta=1.25, smoothness=2)
 Y<- matrix(0,64,64)
 Y[16,16]<- 1

 result<- stationary.image.cov( Y= Y,cov.obj=cov.obj)
  temp<-  matrix( result, cov.obj$m,cov.obj$n)
 image.plot( cov.obj$grid$x,cov.obj$grid$y, temp)

# Note we have centered at the location (16,16) for this case

}

\keyword{spatial}
% docclass is function
% Converted by Sd2Rd version 1.21.