Gaussian Markov random field based improved texture descriptor for image segmentation

作者：

Highlights：

• Distributions of GMRF local parameter estimates as improved texture descriptors.

• Spatially localized parameter estimation using local linear regression.

• Approaches to overcome inconsistencies in localized parameter estimation.

• Proposed descriptors capture both spatial dependencies and distributions of texture.

摘要

•Distributions of GMRF local parameter estimates as improved texture descriptors.•Spatially localized parameter estimation using local linear regression.•Approaches to overcome inconsistencies in localized parameter estimation.•Proposed descriptors capture both spatial dependencies and distributions of texture.

论文关键词：Gaussian Markov random field,Texture feature extraction,Local feature distributions,Local linear regression,Texture segmentation,Natural image analysis

论文评审过程：Received 30 July 2013, Revised 23 March 2014, Accepted 21 July 2014, Available online 27 July 2014.

论文官网地址：https://doi.org/10.1016/j.imavis.2014.07.002