Averaging spectra to improve the classification of the noise radiated by ships using neural networks

The noise radiated from ships in the ocean contains information about their machinery, being normally used for detection and identification purposes. In this work we use a neural classifier to identify the radiated noise received by a hydrophone that was far from the ship. The classification is performed in the frequency domain using a feedforward neural network, which is trained using the backpropagation algorithm. It is shown that the use of an averaged spectral information during the production phase improves significantly the efficiency of the classifier, when it is compared to a neural classifier that processes frequency domain data obtained from individual acquisition windows.