Svd texture

From Eigenvector Research Documentation Wiki
Revision as of 10:34, 29 October 2009 by imported>Scott (→‎See Also)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Purpose

Calculates SVD spectrum of images for texture analysis.

Synopsis

y = svd_texture(img,nn,options)
y = svd_texture(img,nn,incr)

Description

SVD_TEXTURE can be used to characterize texture for the input (img) which is a m by n matrix (e.g. an image layer) for a given sample size (nn) which specifies the size of the square submatrices samples from the original matrix. The optional third input (options) may contain any of the following fields:

  • incr {[1]} Defines the increment used to move the submatrix samples.Default is 1 (one) indicating that ALL possible nn by nn submatrices in the original matrix are used.

If a single scalar value is given instead of an options structure, it is assumed to be the value for options.incr. The output (y) is a 2x1 cell array. The first cell contains a vector of the mean SVD spectrum, and the second cell contains a vector with the standard deviation of the SVD spectrum.

See Also

autocor_img, detrend_img, fft_texture, semivar, texture