Various methods and techniques developed by researchers worldwide for
enhancement and processing ATM, MSSĀ· and TM remotely sensed data are
tested. on LANDSAT 5 Thematic Mapper data from a part of the Barberton
Greenstone Belt straddling the border between the Republic of South Africa and
the Kingdom of Swaziland.
Various enhancement techniques employed to facilitate the extraction of
structural features and lineaments, and the findings Of the ensuing
photogeologlcal interpretation are compared with existing geological maps~
Methods for the detection of zones of hydrothermal alteration. are also
considered.
The reflectance from vegetation, both natural and cultivated, and the possible
reduction of the interference caused by this reflectance, are considered in detail.
Partial unmixing of reflectances through the use of various methods and
techniques, some of which are readily available from the literature, are
performed and its effectiveness tested. Since large areas within the study area
are covered by plantations, the interfereiice from the two types of vegetation
present (i.e. natural and cultivated), were initially considered separately. In an
attempt to isolate the forested areas from the natural vegetation, masks derived
through image classification were used to differentially enhance the various
features.
Results indicate that the use of any particular method to the exclusion of all
others will seriously limit the scope of conclusions possible through interpretation
of the information present. Enhancement of information in one domain will
inadvertently lead to the suppression of information from one or more of the coexisting
domains. A series of results from a sequence of procedures interpreted
in parallel will in every case produce information of a higher decision making
quality. / AC2017
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/22827 |
Date | January 1993 |
Creators | Cloete, Derik |
Source Sets | South African National ETD Portal |
Language | English |
Detected Language | English |
Type | Thesis |
Format | application/pdf |
Page generated in 0.0644 seconds