Final published version, 67 MB, PDF document
Available under license: CC BY: Creative Commons Attribution 4.0 International License
Research output: Thesis › Doctoral Thesis
Research output: Thesis › Doctoral Thesis
}
TY - BOOK
T1 - Wavelet methods for the statistical analysis of image texture
AU - Taylor, Sarah L.
PY - 2013/7
Y1 - 2013/7
N2 - This thesis considers the application of locally stationary wavelet-based stochastic modelsto the analysis of image texture. In the first part we propose a test of stationarity for spatialdata on a regular grid. This test is then incorporated into a segmentation framework inorder to determine the number of textures contained within an image, a key feature to manytexture segmentation approaches. These novel methods are subsequently applied to varioustexture analysis problems arising from work with an industrial collaborator. The secondpart of this thesis considers the modelling of the spectral structure of a non-stationarymultivariate image, i.e. an image containing different colour channels. We propose a multivariatelocally stationary wavelet-based modelling framework which permits a measure ofdependence between pairs of channels. The performance of this modelling approach is thenassessed using various colour texture examples encountered by an industrial collaborator.
AB - This thesis considers the application of locally stationary wavelet-based stochastic modelsto the analysis of image texture. In the first part we propose a test of stationarity for spatialdata on a regular grid. This test is then incorporated into a segmentation framework inorder to determine the number of textures contained within an image, a key feature to manytexture segmentation approaches. These novel methods are subsequently applied to varioustexture analysis problems arising from work with an industrial collaborator. The secondpart of this thesis considers the modelling of the spectral structure of a non-stationarymultivariate image, i.e. an image containing different colour channels. We propose a multivariatelocally stationary wavelet-based modelling framework which permits a measure ofdependence between pairs of channels. The performance of this modelling approach is thenassessed using various colour texture examples encountered by an industrial collaborator.
M3 - Doctoral Thesis
ER -