Share Email Print
cover

Proceedings Paper

Application of neural networks for retrieving atmospheric gases concentration profile for lidar sounding data
Author(s): Mikhail Yu. Kataev; A. Ya. Sykhanov
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

In the report a method of ozone profile concentration retrieving from the lidar data sounding on the basis of neural networks (NN) is description. Application of neural networks in inverse tasks is connected with solving some important stages. In the first, it is necessary to carry out training of NN on the basis of the big data set (measurement - decision). In the second, basing on the results of the first stage to generate optimum NN (number of layers, transfer functions). Results of simulation inverse task of ozone profile concentration retrieving from the lidar data sounding have shown reliability of work in NN, speed of the inverse tasks solving and accuracy of retrieving ozone profile comparable to traditional methods.

Paper Details

Date Published: 23 February 2004
PDF: 4 pages
Proc. SPIE 5397, Tenth Joint International Symposium on Atmospheric and Ocean Optics/Atmospheric Physics. Part II: Laser Sensing and Atmospheric Physics, (23 February 2004); doi: 10.1117/12.548590
Show Author Affiliations
Mikhail Yu. Kataev, Institute of Atmospheric Optics (Russia)
Tomsk State Univ. of Control Systems and Radioelectronics (Russia)
A. Ya. Sykhanov, Tomsk State Univ. of Control Systems and Radioelectronics (Russia)


Published in SPIE Proceedings Vol. 5397:
Tenth Joint International Symposium on Atmospheric and Ocean Optics/Atmospheric Physics. Part II: Laser Sensing and Atmospheric Physics
Gennadii G. Matvienko; Georgii M. Krekov, Editor(s)

© SPIE. Terms of Use
Back to Top