Comparative analysis of machine learning models in classifying hydroacoustic noises of sea vessels

Abstract

The automated identification and classification of marine objects by hydroacoustic noise is an important task in water areas monitoring and World Ocean exploration. The important stages in the development of an automated object recognition system are the choice of a classifier and choice of features. The article provides a comparative analysis of the accuracy of determining the ship class based on its hydroacoustic noise by three different models of machine learning and an artificial neural network.