Abstract
Space-borne sensors can provide valuable spatial image data, which play an irreplaceable role in ground target observation. In this paper, we have derived a parameter called the scattering efficiency per unit area (SEA), which is used to describe the ability of different media to change the energy of an incident field in the far zone. Over the range of Hyperion wavelengths, the SEA are unequal, which can be used as the characteristic signal for target classification. In order to increase the accuracy of this model, we selected five groups of characteristic bands unaffected by noise using principal component analysis. We then combined this with a backpropagation neural network; the identification of five ground media was implemented, and the residuals were very low. These theoretical and experimental analyses verify that SEA can be used as the basis for studying the physical characteristics of scattering of ground targets.
© 2017 Optical Society of America
Full Article | PDF ArticleMore Like This
Jiang Xu, Weixian Qian, Qian Chen, and Yang Zhou
Appl. Opt. 57(5) 1254-1259 (2018)
Rosalba Saija, Maria Antonia Iatì, Paolo Denti, Ferdinando Borghese, Arianna Giusto, and Orazio I. Sindoni
Appl. Opt. 42(15) 2785-2793 (2003)
Yangyang Zhang, Jian Yang, Xiuguo Liu, Lin Du, Shuo Shi, Jia Sun, and Biwu Chen
Appl. Opt. 58(36) 9904-9913 (2019)