In: Proceedings of the Second International Workshop on Earth Observation and Remote Sensing Applications (EORSA 2012), Shanghai, China, 8-11 June 2012. - Shanghai, China : IEEE Xplore - ISBN 9781467319478 - p. 31 - 35.
Shanghai, China : IEEE Xplore - ISBN 9781467319478 Second International Workshop on Earth Observation and Remote Sensing Applications (EORSA 2012), 2012-06-08/2012-06-11
In recent decades, researchers have developed methods and models to reconstruct time series of irregularly spaced observations from satellite remote sensing, among which the widely used Harmonic Analysis of Time Series (HANTS) method. Many studies based on time series reconstructed with HANTS documented the excellent performance of this method. While some limitations of HANTS have been noticed in these applications, there is no dedicated study on a systematic evaluation on the performance of the HANTS method. In this study, we evaluated the impact of gaps on the time series reconstruction of NDVI by HANTS. For global representativeness, a simulated NDVI time series dataset was constructed for four generic patterns and was applied as a reference dataset. Then random gaps were introduced into the reference series and both the reference and gapped series were reconstructed by harmonic analysis. The deviations between the two reconstructed results were used to evaluate statistically the accuracy of harmonic analysis under different gap conditions. The size of maximum gap (MGS), the number of loss (NL) and the number of gaps (NG) were selected to parameterize the gap distribution. The results showed that MGS, NL and NG were significant factors in the process of reconstruction and the two terminals and the peak of the series are crucial positions. MGS and NL should not be too large in the time series for all seasonal or non-seasonal case; otherwise the reconstructed series is not reliable. These conclusions can be taken as a reference to indicate the reliability of HANTS for particular cases towards the definition of a quality indicator of any time series.
There are no comments yet. You can post the first one!
Post a comment
Please log in to use this service. Login as Wageningen University & Research user or guest user in upper right hand corner of this page.