An algorithm for treating uncertainties in the visualization of pipeline sensors' datasets

Abstract

Researchers have seen visualization as a tool in presenting data based on available datasets. Its usage is however undermined by its inability to acknowledge the associated uncertainties in real world measurements. Visualization results are said to be "too generous", providing us with visual assumptions that though, may not be too far from reality, but the associated inaccuracies could become significant when dealing with life dependant datasets. Uncertainty reality is now becoming a significant research interest. In most cases accuracy is a neglected issue. Two wrong assumptions are believed; the first is that the data visualized is accurate, and the second is that the visualization process is exempt from errors. The objectives of this paper are to present the implications of inaccuracies and propose a treatment algorithm for the visualizations of pipeline sensors' datasets. The paper also features attributes that gives a user an idea of sensors' datasets inaccuracies. © 2009 Springer-Verlag.