Evaluation of single-date and multi-seasonal spatial and spectral information of Sentinel-2 imagery to assess growing stock volume of a Mediterranean forest.
International Journal of Applied Earth Observation and Geoinformation(2019)
摘要
•Single-date and multi-seasonal Sentinel-2 models for GSV prediction were developed based on the bagging LASSO algorithm.•GSV models were developed using the original bands and various spectral and spatial features extracted from the images.•The synthetic spectral bands models presented a subtle improvement over the original Sentinel-2 bands models.•Overall, the most accurate GSV models were developed using GLCM texture measures.•Images acquired during dry season, leaf-on conditions appear to be more appropriate for GSV prediction.
更多查看译文
关键词
LASSO,First-order texture,GLCM,LISA,Spectral bands,Spectral indices
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络