The analysis of East Dongting lake water change based on time series of remote sensing data
Caihong Ma, Qin Dai, Xinpeng Li, Shibin Liu · 2014
Long time series of remote sensing data (such as MODIS, NOAA/AVHRR, SPOT/VEGETATION data) as an important source of research data, has been widely applied on vegetation growth monitoring, phenology information extraction and land use monitoring and other fields. In this paper, we take MODIS time-series data in the East-Dongting lake area as experimental remote sensing data. Firstly, long time series of remote sensing data will be gotten through features exaction. Then, the time series data will be analyzed by the hierarchical clustering method. Lastly, this paper presented a new hierarchical clustering method with class number automatic calculation, and according to the characteristics of the time series data, the dynamic time distance (DTW) instead of traditional Euclidean distance were used in the clustering method. The experimental results show that the proposed algorithm can well solve the problems of time axis stretching, bending, linear drifting and etc. And, its clustering effect is more significant.