Reconstruction of Vegetation Index Time Series Based on Self-Weighting Function Fitting from Curve Features

Vegetation index (VI) time series derived from satellite sensors have been widely used in the estimation of vegetation parameters, but quality VI is easily affected by clouds and poor atmospheric conditions. The function fitting method a effective noise reduction technique for series, it vulnerable to noise. Thus, ancillary data about are utilized alleviate interference However, this approach limited availability, accuracy, application rules data. In paper, we aimed develop new reconstruction that does not require Based on assumptions follow gradual growth decline pattern dynamics, or conditions usually depress values, proposed based self-weighting curve features (SWCF). SWCF consists two major procedures: (1) determining weight each point (2) implementing weighted reconstruct series. double logistic function, Gaussian polynomial were tested simulated dataset. results indicate with outperformed corresponding unweighted root-mean-square error (RMSE) significantly reduced 26.82–52.44% (p < 0.05), also Savitzky–Golay filtering RMSE 13.98–54.04% 0.05) 270 sample points selected mid-high latitudes Northern Hemisphere. Moreover, showed excellent robustness applicability regional applications.