Object classification using artificial neural network processing of data obtained by impulse GPR with 1 Tx + 4Rx antenna system

The system for detection of landmines and similar object is investigated. The processing of data obtained by impulse GPR with 1 Tx + 4Rx antenna system is proposed on the basis of artificial neural network approach. The unique antenna system collects important characteristics of subsurface object from reflected impulse electromagnetic wave. The filtration of received signals from interferences caused by experimental setup, its wave-like movement, and ground surface permits to prepare the data of the needed level of quality for further processing by artificial neural network. The network successfully classify a type of object and distance to them. The stability of the object recognition is checked for different level of additional noise.