Geography Reference
In-Depth Information
locate, compute and map the spatial distribution of each change-type. The results
were defined by twentieth detailed combinations of the five classified general
classes, in addition to the class of (no change), i.e., it presents 21-classes of
changes. Figure 6.14 visualizes the results of the changes.
Evaluation of changes/accuracy. Some 15-99 testing points for each change
combination were distributed, i.e., 1,163 in total (Table 6.9 ), automatically, for the
twentieth resulted change combinations classes, over the resulted thematic map
(Fig. 6.14 ). The overall accuracy was 83 %, i.e., lesser than those resulted using
the pre-classification approach (86 %) (see Sect. 6.3.1 ). There were two major
reasons. The first was because of the misclassification of the five general LULC-
classes of interest, especially between the artificial surfaces and the bare areas, and
the second was due to the pre-classification approach being limited to only three
wide general change possibilities in contrast to the post-classification approach,
that had twentieth one change possibilities to be tested.
References
Braimoh, A. K. (2004). Modeling land-use change in the Volta Basin of Ghana. Doctoral
dissertation, Center for Research Development (ZEF), University of Bonn, Bonn, p. 159.
Congalton, R. G. (1991). A review of assessing the accuracy of classifications of remotely sensed
data. Remote Sensing of Environment, 37(1), 35-46.
Jensen, J. R. (2005). Introductory digital image processing: A remote sensing perspective (3rd
ed.). New Jersey: Prentice Hall Series in Geographic Information Science.
MacLeod, R. D., & Congalton, R. G. (1998). A quantitative comparison of change detection
algorithms for monitoring eelgrass from remotely sensed data. Photogrammetric Engineering
& Remote Sensing, 64(3), 207-216.
Search WWH ::




Custom Search