LandCover/LandUse Classification from Landsat 8 OLI Images (River Tana Watershed)

Author: Mariam Tsitsagi
Keywords: RS, LC/LU classification, Landsat 8
Annotation:

Remote sensing has long been an important and effective means for monitoring land cover with its ability to quickly provide broad, precise, impartial and easily available information regarding the spatial variability of the land surface. The successful launch of Landsat 8 provides a new data source for monitoring land cover, which has the potential to significantly improve the characterization of the earth’s surface. To assess data performance, Landsat 8 Operational Land Imager (OLI) data were first compared with Landsat 7 ETM + data using texture features as the indicators. Compared to Landsat 7, Landsat 8 has several new features: (1) two sensors, i.e. OLI and TIRS, (2) two spectral bands that include coastal aerosol and cirrus bands, (3) one thermal band that is split into two bands, (4) a refined spectral range for some bands, improving the spectral responses across the channels, e.g. near-infrared (NIR) and panchromatic bands, (5) improved radiometric resolution from 8 bits to 12 bits and (6) the change in instrument design has resulted in significant improvements in signal to noise ratios (SNR), almost twice as good as Landsat 7. The narrowing of the NIR band avoids the effect of water vapour absorption at 0.825 μm, similar to that of MODIS, and helps acquire accurate surface reflectance. The enhanced radiometric resolution improves the spectral record precision and avoids spectral saturation compared to previous Landsat data. Therefore, the Landsat 8 OLI data are expected to perform better for in land cover mapping.