Interpretation of land cover using spectral modulation pattern- an example with Landsat 8 OLIimage
Keywords:Spectral signature, Spectral reflectance curve, Landsat 8 OLI, modulation pattern, Visual interpretation.
Spectral signature, in general, can be defined as characteristics of surface objects of transmission, absorption and reflection of electromagnetic radiation. Spectral signature is expected stable and unique for given surface material. Spectral signature can be graphically represented in two-dimensional space in the form of spectral curves. Spectral signature has been used long time for object detection and classification mostly by spectral matching methods. Spectral matching is time-consuming and requires reference spectra that is not always available in hand. Spectral modulation pattern is a simplified form of spectral curve shapes. We define this pattern by pairwise comparison of reflectance between two spectral bands. The aim of this research is to point out the use of spectral modulation patterns for land cover mapping. The experiment has been carried out with Landsat 8 OLI image data, which has six reflective bands of 30 m spatial resolution. The Landsat 8 OLI is an excellent data source for land cover mapping in both local and global scale. Due to very huge data volume, automated analysis is crucial when we need accomplish land cover maps in a short time. In this paper, the authors introduce the concept to use spectral modulation patterns of spectral signatures for automated interpretation of land cover. The first step in this concept is to understand correctly meaning of each modulation pattern of spectral signatures and their relation to land cover classes. In this study, we use three Landsat 8 OLI images of Vietnam in 2013 and 2014. Ground GPS field photos were collected to support interpretation of land cover.
Canty, M.J., Nielsen, A.A., Schmidt, M., 2004: Automatic radiometric normalization of multitemporal satellite imagery. Remote Sensing of Environment 91.3, 441-451.
Duong, N. D.,1997: Graphical analysis of spectral reflectance curve.In Proceedings of the 18th Asian Conference on Remote Sensing, Oct 1997, pp. 20-24.
Laborte, Alice G., Maunahan, A.A., Hijmans, R.J., 2010: Spectral signature generalization and expansion can improve the accuracy of satellite image classification. PloS one 5.5, e10516.
Olthof, I., Butson, C., Fraser, R., 2005: Signature extension through space for northern landcover classification: A comparison of radiometric correction methods. Remote Sensing of Environment 95.3, 290-302.
Ruby, J.G., Fischer, R.L., 2002: Spectral signatures database for remote sensing applications. International Symposium onOptical Science and TechnologyInternationalSociety for Optics and Photonics, 2002.
Shaw, G.A., Burke, H-h.K, 2003: Spectral imaging for remote sensing". Lincoln Laboratory Journal 14.1, 3-28.
Thenkabail, P., et al., 2007: Spectral matching techniques to determine historical land-use/land-cover (LULC) and irrigated areas using time-series 0.1-degree AVHRR Pathfinder datasets. Photogrammetric Engineering & Remote Sensing73.10, 1029-1040.