CSDL Home IEEE Transactions on Pattern Analysis & Machine Intelligence 2004 vol.26 Issue No.07 - July
Issue No.07 - July (2004 vol.26)
Huawu Deng , IEEE
David A. Clausi , IEEE
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TPAMI.2004.30
<p><b>Abstract</b>—Features based on Markov random field (MRF) models are sensitive to texture rotation. This paper develops an anisotropic circular Gaussian MRF (ACGMRF) model for retrieving rotation-invariant texture features. To overcome the singularity problem of the least squares estimate method, an approximate least squares estimate method is designed and implemented. Rotation-invariant features are obtained from the ACGMRF model parameters using the discrete Fourier transform. The ACGMRF model is demonstrated to be a statistical improvement over three published methods. The three methods include a Laplacian pyramid, an isotropic circular GMRF (ICGMRF), and gray level cooccurrence probability features.</p>
Markov random field (MRF), Gaussian MRF (GMRF) model, isotropic, anisotropic, least squares estimate (LSE), discrete Fourier transform (DFT), rotational invariance, texture analysis, classification.
Huawu Deng, David A. Clausi, "Gaussian MRF Rotation-Invariant Features for Image Classification", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.26, no. 7, pp. 951-955, July 2004, doi:10.1109/TPAMI.2004.30