Effective learning algorithm for a neural-like opto-electronic tomographical system
Oleg T. Kamenev, Yu. N. Kulchin, A. V. Panov, Yu. S. Petrov · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2005
In this paper we report the results of the working out the fiber-optic measuring system. The fiber-optic measuring lines consisting from sensors are described. The reference piezoelectric modulator is used for the control of the workin gpoint of the measuring line. The tomographical problem arising in data processing is solved by the neural network. An efficient algorithm for neural network training is presented.