Abstract
Abstract. Due to the influence of cloud cover, atmospheric disturbance and many other factors, normalized difference vegetation index (NDVI) corrupted by noises has a negative effect on the downstream applications. To this end, researchers have developed a large number of methods to reconstruct NDVI time series. The harmonic analysis of time series (HANTS) is one of the most widely used approaches of NDVI reconstruction. In this paper, HANTS algorithm was improved by the utilization of spatio-temporal information of NDVI time series with spatial filling and filtering, which makes up the deficiency of HANTS that only uses temporal information of NDVI time series. The simulation experiments on moderate resolution imaging spectroradiometer (MODIS) NDVI time series have proved that our method has effectively improved the quantitative and qualitative reconstruction performance of HANTS algorithm.
Highlights
Normalized difference vegetation index (NDVI) time series data derived from numerous satellite sensors have been widely applied in assessing the global ecological environment (Pettorelli et al, 2005), monitoring and information extraction of vegetation phenology (Atkinson et al, 2012; Zhang et al, 2003) and dynamic change of land cover (De Beurs et al, 2004; De Fries et al, 1998)
The data used in this study is MOD13A1, one of the moderate resolution imaging spectroradiometer (MODIS) products of EOS/Terra satellite, including the 500m resolution normalized difference vegetation index (NDVI) time series synthesized by maximum value composite (MVC) method in 16 days and its quality control data VI_Quality from 2008 to 2018
2.2.2 Image Reorganization To overcome the limitation of harmonic analysis of time series (HANTS) algorithm which performs poorly in the reconstruction of long-time gaps, we proposed an image reorganization method to obtain a 2dimensional image, in which the adjacent pixels had a strong positive correlation
Summary
Normalized difference vegetation index (NDVI) time series data derived from numerous satellite sensors have been widely applied in assessing the global ecological environment (Pettorelli et al, 2005), monitoring and information extraction of vegetation phenology (Atkinson et al, 2012; Zhang et al, 2003) and dynamic change of land cover (De Beurs et al, 2004; De Fries et al, 1998). The above studies have compared these reconstruction techniques in many aspects, the results can be distinctly different because the methods have obvious merits and demerits and can be affected by vegetation types, study area and noisy pixel percentage Among these methods, HANTS is a widely used and one of the best methods in NDVI time series reconstruction. It is vital to evaluate the performances of denoising methods, and ground observation is the most effective way of verification This kind of reference data, which need to be spatially appropriate, is hard to obtain. The rootmean-square error (Eq (1), abbreviated as RMSE) between the reconstruction and reference NDVI was calculated as a statistical indicator to evaluate the denoising performances of our method and HANTS algorithm.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
