Fisher tensors for classifying human epithelial cells

作者：

Highlights：

• Exploring region covariance descriptors for representing HEp-2 cells.

• Introducing bag of Riemannian words for efficient classification of HEp-2 cells.

• Proposing Fisher tensors as an extension of bag of Riemannian words for classifying HEp-2 cells.

• Thorough experiments show proposed approach outperforms state-of-the-art methods.

摘要

Highlights•Exploring region covariance descriptors for representing HEp-2 cells.•Introducing bag of Riemannian words for efficient classification of HEp-2 cells.•Proposing Fisher tensors as an extension of bag of Riemannian words for classifying HEp-2 cells.•Thorough experiments show proposed approach outperforms state-of-the-art methods.

论文关键词：Riemannian manifolds,Region covariance descriptor,Bag of visual words,Fisher vectors,Human Epithelial Cell Type 2 (HEp-2),Texture classification

论文评审过程：Available online 18 October 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.10.011