The overall accuracies of the supervised classification of the satellite images accounted for 94 % (for year 2000), 91 % (for year 2005), and 88 % (for year 2010), respectively (Table 13.3). The kappa statistics were 0.90, 0.80, and 0.79, respectively for the years 2000, 2005, and 2010 classifications.
The KHAT statistic was calculated to determine the statistical significance or classification accuracy between the years. The computed KHAT statistic is a better indicator of percentage because the values approach 1. The overall image accuracies for each date, following verification from field data, were well above the
1 = water
2 
= tea
3 = secondary forest shmb
 |
4= mixed agriculture/residential/ road 5= primary forest
|
Table 13.3 Accuracy assessment and kappa statistic for land cover classification (years 2000, 2005, and 2010)
|
generally accepted 85 % standard of accuracy for image classifications (Foody 2003), with kappa statistics above 0.75 (Nagendra et al. 2003). The generally accepted overall accuracy level for land use maps is 85 %, which is approximately equal in accuracy for most categories (Jensen 1986; Mohd Hasmadi and Kamaruzaman 2008).
|
Land cover type
|
CA class area, PLAND percent of landscape, NP number of patches, PD patch density, AREA_MN mean patch area, LPI largest patch index
