Implementation of an Artificial Intelligence Approach to GPR Systems for Landmine Detection

Artificial Neural Network (ANN) approaches are applied to detect and determine the object class using a special set of UltraWideBand (UWB) pulse Ground Penetrating Radar (GPR) sounding results. It used results GPR with antenna system, consisting one radiator four receiving antennas located around transmitting antenna. The presence and, accordingly, signals received from spatially separated positions provide collection after reflection an at different angles due this, location in coordinate connected We considered sums differences by two six possible combinations: (1 2, 1 3, 2 4, etc.). These combinations were then stacked sequentially into long signal. Synthetic constructed such way contain many more notable specific information about which belongs as well searched compared obtained system just radiating therefore increases accuracy determining object’s coordinates its classification. radiation, propagation, scattering numerically simulated finite difference time domain (FDTD) method. Results experiment on mine detection examine ANN too. objects having distances was training testing dataset for ANN. aims recognize classify detected landmine or other location. influence Gaussian noise added immunity investigated. recognition ensemble presented. consists fully recurrent neural networks, gated units, long-short term memory network. all ANNs processed meta network better quality underground