Abstract:
The processing of digital imagery is an important aspect of the analysis of Landsat multispectral data. Classification of the Landsat data into representable land use cover types is an integral part of the digital process. Cartographers and planners must receive the most accurate information from the digital processing of Landsat data, to enable them to portray that data in map form and integrate the products of computer aided image analysis to their application. The purpose of this paper is to present a program which analyzes Landsat classifications in order to determine the accuracy of sample area classifications, individual cover types, and pixel representations.