Land-cover classi cation methods for multi-year AVHRR data

AdvancedVery HighResolutionRadiometer (AVHRR) data have been extensively used for global land-cover classiŽ cation, but few studies have taken direct and full advantage of the multi-year properties of AVHRR data. This study focused on generating eVective classiŽ cation features from multi-year AVHRR data to improve classiŽ cation accuracy.Three types of features were derived from 12-year monthly composite normalized diVerence vegetation index (NDVI) and channel 4 brightness temperature from the NOAA/NASA PathŽ nder AVHRR Land data for land-cover classiŽ cation. The Ž rst is based on the shape of the annual averageNDVI or brightness-temperatureproŽ le, which was then approximated by a Fourier series. The coeYcients estimated by the weighted least-squares method were used for classiŽ cation. The second and third features were based on the raw periodogram of the time series and the auto-regressive modelling. A global land-cover training database created from Landsat Thematic Mapper and Multi-spectral Scanner imagery was used for training and validation.Both quadrature discriminate analysis (QDA) and linear discriminate analysis (LDA) were explored for classiŽ cation, and results indicate that QDA performs much better than LDA. The Ž rst feature, based on the mean annual shape, produced much better results than the other two features. It was also found that NDVI signals worked better than brightness-temperaturesignals. That is probably because topof-atmosphere signals were used, and atmospheric contaminations signiŽ cantly disturb the thermal signals and correlation structures of diVerent cover types. Further validations are needed.