SPA304 Remote Sensing Algorithms (8)

• Have an understanding of the sources and characteristics of remote sensed data;
• Have an awareness of the radiometric and geometric corrections that are required to correct remote sensed imagery for atmospheric and view geometry distortions;
• Have an understanding of the mathematical and statistical formulation of image processing techniques commonly used in remote sensing;
• Be able select and use appropriate image processing techniques to ehance and classify land cover classes using multispectral image data.

Sources and characteristics of remote sensed data; Geometric and radiometric correction imagery; Radiometric enhancement techniques including look up tables, linear contrast enhancement, histogram equalisation, histogram matching, density slicing and pseudocolour techniques; Geometric enhancement techniques including smoothing, edge detection and enhancement, spatial derivatives and general convolution filtering; Multispectral transformations of image data including principal component analysis, band ratios, vegetation indicies, tasseled cap and Taylor transformations; Spatial filtering using Fourier transformation techniques; Supervised clasification of multispectral imagery including maximum likelihood, box, Euclidean distance and Mahalanobis classifiers; Advanced classification techniques including contextural, neural network and expert system classifiers; Unsupervised classification techniques; Feature reduction and land cover discrimination using cannonical variate analysis techniques; Analysis of multispectral remote sensed image data - case studies.