A Developed Algorithm for Automating the Multiple Bands Multiple Endmember Selection of Hyperion data Applied on Central of Cairo, Egypt

摘要: This study attempts to provide an answer regarding the utility of Hyperion imagery in mapping urban settings developed countries. The authors present a novel method for extracting quantitative land cover information at sub-pixel level from hyperspectral or imagery. proposed is based on multiple endmember spectral mixture (MESMA) by Roberts et al. (1998b), but extends it handle high-dimensional pixels characterizing images. utilizes multiband analysis (Multiband MESMA) model that allows both bands and endmembers vary per-pixel basis across image. goal select optimal subset maximizes separability among candidate set given pixel, accordingly minimize confusion modeled increase accuracy physical representativeness derived fractions pixel. develop tool automate this test its case using image Central Cairo, Egypt. EO-1 sensor only source data currently available unlike cities Europe North America, where sources such generally exist. scene represents very heterogeneous landscape has ecological footprint complex range interrelated socioeconomic, environmental dynamics. results show data, with rich information, can help address some limitations automated are reported previous studies. For this, proper selected used within endmember, multiple-band SMA process determine best Root Mean Square Error (RMSE) abundance percentages. better extricated (Hyperion). Keywords: Spectral Mixture Analysis, Hyperspectral Data, Egypt