Geoscience Reference
In-Depth Information
Fig. 10
Accuracy comparison between D16 (a) and D8 (b) algorithms result
generalisation process respectively. The generalisation technique in D16 will
reduce the number of stagnant areas while increasing the number of sink areas.
On the other side, D8 algorithm will reduce the number of sink area when
undergo the process of generalisation. As a result, D8 only produces several sink
areas as shown in Figs.
13
and
14
, after four (4) times scale generalisation from
actual resolution. The percentage difference of D8 generalisation process that still
maintains the sink area is about 15.38 % compare to D16 result around 125.0 %.
Thus, D16 algorithm is capable of producing 8.13 times greater accuracy details in
the generalisation process compared to D8 algorithm using the constant values in
stagnant and sink formulae as discussed in
Sect. 3.1.1
.
