作者: Myunghee Jung , Eunmi Chang
DOI: 10.1080/01431161.2015.1007252
关键词:
摘要: This study presents a normalized difference vegetation index NDVI-based land-cover change detection method based on harmonic analysis. Multi-temporal NDVI data show seasonal variation characteristics in the time domain. A model represents characterization of temporal variability set over local region corresponding to pixel through its components. In this research, annual is performed by tracking dynamics analysing simple but effective noise reduction process also proposed provide necessary high-quality stream for multi-temporal analysis statistics observed oscillations. The algorithm was tested and evaluated with Moderate Resolution Imaging Spectroradiometer MODIS series MYD13Q1, 16 day L3 global 250 m SIN grid v005 VI set. results indicate that provides computationally inexpensive automatic monitor conditions long-term large regions. described here particularly useful monitoring changes well-established deciduous forests developed canopies.