The cost of time - implications of hyperspectral data volume and feature selection routines for conservation science

摘要: The recent greater availability of airborne hyperspectral imagery in the tropics has allowed for analysis increasingly complex analytical questions ecology such as remote tree species identification. In comparison to identification spectral expression gender dioecious been generally overlooked despite its effects on plant ecophysiological functioning and prevalence tropics. A problem often implied but not frequently addressed these analyses is complexity posed by data volume collected sensors. We examine effect this specifically feature selection routines classification implication resultant limitations use at regional operational scales. conclude based an examination time cost high performance computing systems, that efficient alternative large scale academic or NGO research a cluster PlayStation™ 3s.