Analysis of the Effectiveness of Spectral Mixture Analysis and Markov Random Field Based Super Resolution Mapping in the Context of Urban Composition

  • D . R . Welikanna
  • Published 2007


Disclaimer This document describes work undertaken as part of a programme of study at the International Institute for Geo-information Science and Earth Observation. All views and opinions expressed therein remain the sole responsibility of the author, and do not necessarily represent those of the institute. Abstract The information in a pixel of satellite data within the instantaneous Field of View (IFOV) of the sensor is a mixture of different land cover types, and the individual land cover components can be estimated using soft classification techniques. However these techniques do not account for the spatial distribution of the class proportions, the information itself has a great relevance. Over the years techniques have been developed in attempting to provide an improved spatial distribution of land cover. Few studies have been tested on the difficult task of mapping the land cover from real satellite images, and more over a very few attempts have been done in a heterogeneous urban environment. Over the years the Markov Random Field (MRF) based Super Resolution Mapping (SRM) technique has been used for land cover classifications, and found to be an effective tool for the generation of land cover maps from remotely sensed images and it consider the spatial distribution of class proportions with in and between the pixels. In this study MRF based SRM with certain modifications have been analysed for its performance with respect to the linear unmixing technique applied on hyperspectral data. To standardize the application of these techniques the urban environment was defined by the Vegetation, Impervious surface and soil (V-IS) model which has been used as an accepted alternative in characterising the urban land cover components. Linear unmixing technique with a hyperspectral remote sensing image (Hyperion) has been used to generate fractions according to the spectral variability of the V-IS classes. Modified MRF based SRM technique was applied on IKONOS, ASTER MSS and Landsat images with markedly different spatial and spectral resolutions. Those are the 3 band with 15m spatial resolution and 6 bands with 30m spatial resolution of the ASTER image, 4m spatial resolution of IKONOS image and the 6 bands with 30m spatial resolution of the Landsat image. Reference maps for the validation were created from the IKONOS MSS image using hard Maximum likelihood classification. And the super resolution maps which contain the spatial information were again turned in to fractions representing each class (V-IS). The results of MRF based SRM …

45 Figures and Tables

Cite this paper

@inproceedings{Welikanna2007AnalysisOT, title={Analysis of the Effectiveness of Spectral Mixture Analysis and Markov Random Field Based Super Resolution Mapping in the Context of Urban Composition}, author={D . R . Welikanna}, year={2007} }