Integration of Landsat Digital Data and Ancillary Data for Improving Automated Classification of Forest Lands

Abstract

The use of Landsat data for classifying forest lands have, often resulted in insufficiently high classification accuracy being obtained. Since topographic variables such as slope and aspect have a significant effect on spectral response as recorded on Landsat CCT, these variables can be incorporated to increase classification accuracy. A number of approaches have been developed to improve cJassificatory performance. In this paper various methods using ancillary data along with digital Landsat data for improving classification accuracy and problems associated with them have been discussed. Since digital remote sensing data is increasingly becoming part of geographic information systems, the way in which ancillary data are combined with spectral data and algorithms used in making decisions about class allocations need to be carefully considered and better developed than at present.