Abstract:
Environmental noise in the form of influences from cloud cover and/or
smoke is often a critigal problem faced when deploying analyses using
rernotely sensed infoimation. Such 'noise' will impact on the quality of
the analysis. Hencg, it becomes necessary to remove such 'noise' in the
pre-processing stages of the GIS-remote sensing analysis. whilst various
noir"-r"n1oval methods have been developed by the purist image
processing communiry a common problem faced by general practitioners
in environmental science is that these methods are complex and very
hard to comprehend and apply off-the-hand. Considering these difficulties,
a simple yeicomprehensive methodolory is presented in this paper, where
a tutorial spatial model using MS Excel is applied to explain the principle(s)
behind spatial noise removal. using digitized data based on Landsat
images for the Vavuniya district, the paper provides a detailed step-bystep
discussion on how environmental noise could be removed using
Boblean logic operators in"the MS Excel platform. Further, a detailed
description of how this 'noise removal concept' can be transferred into
the BtiOAS Imagine software platform for remotely sensed analysis is
also given using the (original) Landsat image (ry) for Vavuniya district
u, un
"*urnple.
Whilst this prototyping method is established as an
excellent .\r"nu" for validation, refinement and improven;'ent of the
analytical procedure, it is also stressed that future research in-these
l."gu.d, should focus on developing software interfaces in advanced
mathematical programming languages for enabling development of
efficient and user-friendly tools for environmental noise removal'
