Inheritimage and Maf: Difference between pages

From Eigenvector Research Documentation Wiki
(Difference between pages)
Jump to navigation Jump to search
imported>Scott
(New page: ===Purpose=== Converts data DSO to image DSO if source contains image data. ===Synopsis=== :idata = inheritimage(data,source) :idata = inheritimage(data,source,options) ===Description=== C...)
 
imported>Scott
(New page: ===Purpose=== Maximum Autocorrelative Factors. ===Synopsis=== :[scores,loads,mn,ssq] = maf(x) ===Description=== Similar to PCA, MAF uses singular value decomposition to provide a model of ...)
 
Line 1: Line 1:
===Purpose===
===Purpose===
Converts data DSO to image DSO if source contains image data.
Maximum Autocorrelative Factors.
===Synopsis===
===Synopsis===
:idata = inheritimage(data,source)
:[scores,loads,mn,ssq] = maf(x)
:idata = inheritimage(data,source,options)
===Description===
===Description===
Check 'source' for datasource image associated fields. If they are found then change 'data' into an image DSO. To be used with plotscores and other image enabled plotting.
Similar to PCA, MAF uses singular value decomposition to provide a model of data (x) which captures maximum spatially-correlated variance. The resulting scores (scores), loadings (loads), and mean spectrum (mn) can be used to reconstruct the mean-centered data matrix X_mn:
====INPUTS====
X_mn = scores\*loads
*    '''data''': DataSet Object.
The difference between PCA and MAF is that MAF extracts loadings which are highly correlated in the spatial dimension of an image. In addition, MAF always returns the entire set of components up to the rank of the data matrix.
*    '''source''':  Either a DataSet Object, a standard model structure, or a cell''' '''containing one or more standard model structures.
Input X is either an Image DataSet object or a three-way double matrix (first two dimensions are spatial, last is variable).
OUPUT:
Output ssq is an experimental sum of squares captured table but because the nature of the decomposition, this table is only approximate. It includes the component number (column 1), estimated eigenvalue (column 2), and the estimated captured per variable and total (columns 3 and 4, respectively).
* '''idata''':  Image DataSet Object.
===Options===
''options'' is a structure is a structure array with the following fields:
* '''datamode''':  ['none'|{'final'}]  governs level of plotting,
* '''sourcemode''':  [{'distslct'}| 'classmean'] algorithm for determining classes
* '''minare''': {1} minimum area (in %) that a class must account for to be retained as a unique class
===See Also===
===See Also===
[[buildimage ]]
[[mcr]], [[parafac]], [[pca]]

Revision as of 10:36, 29 September 2009

Purpose

Maximum Autocorrelative Factors.

Synopsis

[scores,loads,mn,ssq] = maf(x)

Description

Similar to PCA, MAF uses singular value decomposition to provide a model of data (x) which captures maximum spatially-correlated variance. The resulting scores (scores), loadings (loads), and mean spectrum (mn) can be used to reconstruct the mean-centered data matrix X_mn:

X_mn = scores\*loads

The difference between PCA and MAF is that MAF extracts loadings which are highly correlated in the spatial dimension of an image. In addition, MAF always returns the entire set of components up to the rank of the data matrix. Input X is either an Image DataSet object or a three-way double matrix (first two dimensions are spatial, last is variable). Output ssq is an experimental sum of squares captured table but because the nature of the decomposition, this table is only approximate. It includes the component number (column 1), estimated eigenvalue (column 2), and the estimated captured per variable and total (columns 3 and 4, respectively).

See Also

mcr, parafac, pca