Abstract

Hyperspectral imagery (HSI) is an effective technology for quantitative monitoring of shallow water coastal environments. Hyperspectral sensors collect hundreds of narrow and contiguously spaced spectral bands of data organized in the so-called hyperspectral cube. Two challenges arise in the use of hyperspectral sensors for benthic habitat mapping: (1) low spatial resolution and (2) small signal coming out of the water compared to atmospheric clutter which requires good signal to noise ratio (SNR) sensors for remote sensing of benthic habitats. To deal with low spatial resolution, unmixing of the hyperspectral signature is performed. Spectral unmixing is used to retrieve subpixel information. Spectral unmixing is the process of decomposing the measured spectrum into a collection of constituent spectra, or endmembers, and a set of corresponding fractions or abundances. Unmixing algorithms are used to estimate bottom coverage of different species for benthic habitat mapping. Two algorithms are applied to HYPERION and AVIRIS imagery as pre-processing to improve its SNR:resolution enhancement filtering and principal component filtering. A study of how the performance of the unmixing algorithm is affected by the different pre-processing algorithms is performed. AVIRIS imagery from Enrique Reef and Kaneohe Bay is used in the experiments.

Notes

Poster presented at the 2007 Thrust R2C Multi Spectral Discrimination Methods Conference

Keywords

Spectral Signature Enhancement, HIS, coastal environments

Subject Categories

Remote sensing

Disciplines

Engineering

Publisher

Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS)

Publication Date

2007

Rights Holder

Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS)

Click button above to open, or right-click to save.

Included in

Engineering Commons

Share

COinS