COST-EFFECTIVE COMPUTER-PROCESSING OF AIRBORNE SCANNER DATA FOR REGIONAL LEVEL MAPPING

Resource surveys through the use of an aircraft-based scanner involve the generation of enormous quantities of spectral data in numerical format, requiring machine processing for expeditious information extraction. Computer analysis of the airborne 11-channel multispectral scanner data, to be cost effective, requires that a minimum number of channels of data be used commensurate with information requirements. The results of an investigation carried out on the Multispectral Data Analysis System (MDAS), and interactive computer, towards reduction of computer time through the use of fewer spectral channels, but without decreasing categorization accuracy, are described. The results of categorization of a test area of 15 km by 5 km indicate that, for a typical MDAS analysis with 36 land cover/use groups, there can be a reduction of about 130 minutes in CPU time when four channels are used compared to that involving all 11 bands.