Wavelet-Based Semantic Features for Hyperspectral Signature Discrimination

• DOI: 10.48550/arxiv.1602.03903
• Publication Date: 2016-01-01
• Authors: Siwei Feng; Yuki Itoh; M. Parente; Marco F. Duarte
• Keywords: Signature (topology); Spectral signature

Hyperspectral signature classification is a quantitative analysis approach for hyperspectral imagery which performs detection and of the constituent materials at pixel level in scene. The procedure can be operated directly on data or performed by using some features extracted from corresponding signatures containing information like signature's energy shape. In this paper, we describe technique that applies non-homogeneous hidden Markov chain (NHMC) models to classification. basic idea use statistical (such as NHMC) characterize wavelet coefficients capture spectrum semantics (i.e., structural information) multiple levels. Experimental results show based NHMC outperform existing approaches relevant tasks.