Short-Interval Monitoring of Land Use and Land Cover Change Using a Time Series of RADARSAT-2 Polarimetric SAR Images - Space-Time Integration in Geography and GIScience - page 359

Fig. 19.6 Overall accuracy and kappa value for determining types of LULC change

Illegal land development is commonly related to LULC changes from barren
land to urban/built-up areas, that from vegetation to barren land, and that from
vegetation to urban/built-up areas. Accurate identification of these types of LULC
change are important for detecting potential illegal land use development. The
user's and producer's accuracies for determining these types of LULC change
were calculated. As shown in Fig. 19.7 , the user's and producer's accuracies for
changes from vegetation to urban/built-up areas were unstable and very low at some
times, and the user's and producer's accuracies for changes from barren land to
urban/built-up areas also changed with time. The study found that this was mainly
because of the confusion between these two types of LULC change. It is commonly
easy to distinguish between barren land and vegetation in PolSAR images because
they have different scattering mechanisms. Vegetation has strong volume scattering,
while barren land has strong single-bounce scattering because of their relatively flat
surface. However, in the study area, barren lands in many construction sites had
rough surface because they were dug or covered with mounds. These barren lands
were prone to be confused with vegetation because of their rough surface, which
causes strong volume scattering. Fortunately, both changes from barren land to