Minimising Positional Errors in Line Simplification Using Adaptive Tolerance Values

Line simplification is an important function in GIS and cartography and is widely used in commercial GIS software packages. Most line simplification algorithms require the user to supply a tolerance value, which is used to determine the extent to which simplification is to be applied.. All simplification algorithms induce positional errors in the data set, because they produce a discrepancy between the original line and its simplified version. The amount of this error depends on both the tolerance value and the shape of the line. This is the reason that many researchers have focused on measuring the geometric characteristics (or complexity) of lines. Using one tolerance value for all lines in the data set results in different positional errors for different lines. What is usually important for the user, is to maintain a specific level of quality, and not the tolerance value itself. The question is, 'how does one specify a tolerance value for each line based on the user specified level of accuracy'? This paper presents a solution to solve this problem. In this approach, the user supplies the target level for desired accuracy and the simplification tolerance value is calculated accordingly.