2 General techniques
2.1 Assessing the accuracy of a derived composite map
The quality of source data and derived products is of
paramount importance to the reliability of decisions based on
the integrated analysis of multiple spatial variables. Some effective
methods have been developed to test the accuracy of
the original data layers commonly used in GIS analysis. The
most widely used method is statistical sampling. First,a set of
sample points are carefully selected from the map
products using some sampling schemes such as
random, systematic, or cluster sampling. The position
and attribute at each of the sample points are
verified by field survey or using measurements from
sources of known and higher accuracy such as
large-scale aerial photographs or maps. Then the
results are used to calculate some statistical measures
such as the root, mean, square, mean variance,
and standard deviation. Obviously, these
techniques can also be used in assessing the positional
accuracy of maps derived from integrating
multiple data layers of various scales. However,
these methods are generally costly and time consuming.
The results of accuracy assessment also
vary with the number of sample points and the
adopted sampling scheme. In this paper, we propose
a new approach for evaluating the positional
accuracy of spatial objects in composite maps derived
from overlay operations. The approach involves
the following six steps:
Step ].Classifying geographic features on the
derived composite maps. The process is illustrated
in Fig. 1.