Sequential Maximum a Posterior (SMAP) Algorithm for Classification of Urban Area using Multi-resolution Spatial Data with Derived Geographical Layers

Effective conservation and management of natural resources requires up-to-date information on the land cover (LC) types and their dynamics. Multi-resolution remote sensing (RS) data with appropriate classification strategies have been used to categorise land use and land cover (LULC) of a landscape. RS data coupled with other important environmental layers (both remotely acquired or derived from ground measurements) would be more effective in capturing the LC dynamics and changes associated with the natural resources, as ancillary layers provide additional information that would make the decision boundaries between the LC classes more widely separable, enabling classification with higher accuracy compared to conventional methods of RS data classification.The objective of this paper is to ascertain the possibility of improved classification accuracy of high spatiallow spectral resolution IKONOS data and low spatial-high spectral resolution Landsat ETM+ data of an urban area with the addition of ancillary and derived geographical layers such as vegetation index, temperature, digital elevation model (DEM), aspect, slope and texture. Results showed that texture played a major role in discriminating individual classes which were rather difficult to distinguish using only original high spatial resolution IKONOS Multispectral (MS) bands. DEM plays a role when the terrain is undulating, however, due to limited vegetation cover, vegetation index was not useful in classification.With ETM+ MS data, inclusion of temperature, NDVI (Normalised Difference Vegetation Index), EVI (Enhanced Vegetation Index), elevation, slope, aspect, Panchromaticband and texture increased the overall accuracy by 7.6% for the same urban area.Thestudy helped in the selection of appropriate ancillary layers for improved classification of RS data depending on the terrain.