F. Del Frate et G. Schiavon, Nonlinear principal component analysis for the radiometric inversion of atmospheric profiles by using neural networks, IEEE GEOSCI, 37(5), 1999, pp. 2335-2342
Citations number
20
Categorie Soggetti
Eletrical & Eletronics Engineeing
Journal title
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
A new neural network algorithm for the inversion of radiometric data to ret
rieve atmospheric profiles of temperature and vapor has been developed. The
potentiality of the neural networks has been exploited not only for invers
ion purposes but also for data feature extraction and dimensionality reduct
ion, In its complete form, the algorithm uses a neural network architecture
consisting of three stages: 1) the input stage reduces the dimension of th
e input vector; 2) the middle stage performs the mapping from the reduced i
nput vector to the reduced output vector; 3) the third stage brings the out
put of the middle stage to the desired actual dimension. The effectiveness
of the algorithm has been evaluated comparing its performance to that obtai
nable with more traditional linear techniques.