USING ARTIFICIAL BAT SONAR NEURAL NETWORKS FOR COMPLEX PATTERN-RECOGNITION - RECOGNIZING FACES AND THE SPEED OF A MOVING TARGET

Two sets of studies examined the viability of using bat-like sonar inp ut for artificial neural networks in complex pattern recognition tasks . In the first set of studies, a sonar neural network was required to perform two face recognition tasks. In the first task, the network was trained to recognize different faces regardless of facial expressions . Following training, the network was tested on its ability to general ize and correctly recognize faces using echoes of novel facial express ions that were not included in the training set. The neural network wa s able to recognize novel echoes of faces almost perfectly (above 96% accuracy) when it was required to recognize up to five faces. In the s econd face recognition task, a sonar neural network was trained to rec ognize the sex of 16 faces (eight males and eight females). After trai ning, the network was able to correctly recognize novel echoes of thos e faces as 'male' or as 'female' faces with accuracy levels of 88%. Ho wever, the network was not able to recognize novel faces as 'male' or 'female' faces. In the second set of studies, a sonar neural network w as required to learn to recognize the speed of a target that was movin g towards the viewer. During training, the target was presented in a v ariety of orientations, and the network's performance was evaluated wh en the target was presented in novel orientations that were not includ ed in the training set. The different orientations dramatically affect ed the amplitude and the frequency composition of the echoes. The neur al network was able to learn and recognize the speed of a moving targe t, and to generalize to new orientations of the target. However, the n etwork was not able to generalize to new speeds that were not included in the training set. The potential and limitations of using bat-like sonar as input for artifical neural networks are discussed.