Advanced Very High Resolution Radiometer (AVHRR) data have been extensively
used for global land-cover classification, but few studies have taken dire
ct and full advantage of the multi-year properties of AVHRR data. This stud
y focused on generating effective classification features from multi-year A
VHRR data to improve classification accuracy. Three types of features were
derived from 12-year monthly composite normalized difference vegetation ind
ex (NDVI) and channel 4 brightness temperature from the NOAA/NASA Pathfinde
r AVHRR Land data for land-cover classification. The first is based on the
shape of the annual average NDVI or brightness-temperature profile, which w
as then approximated by a Fourier series. The coefficients estimated by the
weighted least-squares method were used for classification. The second and
third features were based on the raw periodogram of the time series and th
e auto-regressive modelling. A global land-cover training database created
from Landsat Thematic Mapper and Multi-spectral Scanner imagery was used fo
r training and validation. Both quadrature discriminate analysis (QDA) and
linear discriminate analysis (LDA) were explored for classification, and re
sults indicate that QDA performs much better than LDA. The first feature, b
ased on the mean annual shape, produced much better results than the other
two features. It was also found that NDVI signals worked better than bright
ness-temperature signals. That is probably because top-of-atmosphere signal
s were used, and atmospheric contaminations significantly disturb the therm
al signals and correlation structures of different cover types. Further val
idations are needed.