Mangrove ecosystem study is one of the main beneficiaries of the application of hyperspectral data and spectral matching techniques. Diversity and density of mangrove species leads to complexity of the ecosystem. Hence, species level mapping becomes difficult. Though hyperspectral images are appropriate for such a mapping, different mangrove species with closely matching spectra pose a challenge. This paper proposes a novel hyperspectral matching algorithm by integrating the stochastic Jeffries-Matusita measure (JM) and deterministic Spectral Angle Mapper (SAM) to accurately map most species of the mangrove ecosystem. The JM-SAM algorithm signifies the combination of an quantitative angle measure (SAM) and an qualitative distance measure (JM). The spectral capabilities of both the measures are orthogonally projected using tangent and sine functions to result in the combined algorithm. The developed JM-SAM algorithm is implemented to discriminate the mangrove species and the landcover classes of Pichavaram and Muthupet mangrove forests of southern India using the Hyperion datasets. The developed algorithm is extended in a supervised framework for improved classification of the Hyperion image. The reference spectra of the mangrove species and other cover types are extracted from the Hyperion image. From the values of relative spectral discriminatory probability and relative discriminatory entropy value, it can be inferred that hybrid JM-SAM matching measure results in improved discriminability than the individual SAM and JM algorithms. This performance is reflected in the classification results where the JM-SAM (TAN) and JM-SAM (SIN) matching algorithms yielded an improved accuracy of (86.25%,85%) and (88.10%, 86.96) for both the study sites.

Date Published: 18 November 2014
PDF: 13 pages
Proc. SPIE 9263, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications V, 926317 (18 November 2014); doi: 10.1117/12.2068938

Published in SPIE Proceedings Vol. 9263:
Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications V
Allen M. Larar; Makoto Suzuki; Jianyu Wang, Editor(s)