Time-delay recurrent neural network for exo-atmospheric infrared target recognition

Abstract

A time-delay recurrent neural network (TDRNN) is proposed for exo-atmospheric infrared target recognition. The TDRNN employs adaptive time delays and recurrences where the adaptive time delays make the network choose the optimal values of time delays for the temporal location of important information in the input sequence and the recurrences enable the network to have dynamic recognition function. A TDRNN is trained and tested using simulated data. The target recognition experiment suggests that TDRNN have a better recognition ability.