Analysis of satellite VEGETATION NDVI time series for estimating Post Fire vegetation recovery

摘要: 1. The Hurst exponent for a data set provides measure of whether, the is pure random walk or has underlying trends. was computed using aggregate variance which time domain method useful non-stationary series. It obtains multi-scale analysis with aggregation adjacent points and measures similarity in terms variance. If H=0.5, signal uncorrelated; if H>0.5 correlations are persistent, where persistence means that large (small) value (compared to average) more likely be followed by value; H<0.5 antipersistent, indicates most not value.