作者: Roger M. Hoffer , Richard S. Latty
DOI:
关键词: Context (language use) 、 Geography 、 Remote sensing 、 Divergence (statistics) 、 Multispectral Scanner 、 Statistical distance 、 Cover (algebra) 、 Channel (digital image) 、 Thematic Mapper 、 Thematic map
摘要: This study involved the evaluation of characteristics multispectral scanner data relative to forest cover type mapping, using NASA's NS-001 simulate proposed Thematic Mapper (TM). The objectives were determine: (1) optimum number wavebands utilize in computer classifications TM data; (2) which channel combinations provide highest expected classification accuracy; and (3) merit each context classes examined. Transformed divergence was used as a measure statistical distance between spectral class densities associated with twelve classes. maximum overall mean pair-wise transformed basis for evaluating all possible waveband available use computer-assisted classifications.