Fused Classification of Surface Ships Based on Hydroacoustic and Electromagnetic Signatures

Surface Ship Classification in a Littoral Environment using Fusion of Hydroacoustic and Electromagnetic Data

“…We have earlier reported a study on ship classification using delay-differential equations, autoregressive models, and Gaussian classifiers with Bayesian fusion inference, see [2]. In that work, however, we used a different approach to validate the models, so the results will not be comparable to the results presented later in this paper.…”

ShipsEar: An underwater vessel noise database

“…e l s e v i e r . c o m / l oc a t e / a p a c o u s t AUTEC (Bahamas); Erbe [9], who recorded 66 jet ski pass-bys to characterize their sound; Roth et al [5], who characterized the noise of an icebreaker under the ice of the Arctic Ocean with a sonobuoy that provided hours of recording before it exited the range of the radio link; Lennartsson et al [17], who used hydroacoustic and electromagnetic signatures to create a database of 15 vessels; Das et al [13], who trained a classifier by completing the recorded sound of 6 boats with synthetic data; Bao et al [15], who used recordings of 6 boats to train a classifier; and Yang et al [16] and Zak [2], who trained a neural network with sounds from 5 Polish Navy ships. Many of these authors expressed the desirability of having better databases for their research, but to date no database of recordings has been made available to the research community.…”

“…The task is challenging, due to the ongoing evolution in engine design, the complexity of sound propagation in the sea (especially in shallow waters) and the frequent presence of high background noise in the sensor. Researchers have applied various signal processing strategies to address these problems: Das et al [13] used spectral characteristics and cepstral coefficients, Wang et al [14] used a bark-wavelet analysis combined with Hilbert-Huang transform, Bao et al [15] exploited the nonlinear features of radiated sound through empirical mode decomposition, Zak [2] used Kohonen neural networks, Yang et al [16] proposed fractal approaches and Lennartsson et al [17] fused sound and electromagnetic signatures for classification purposes.…”

An automated approach to passive sonar classification using binary image features

“…Due to a wide range of applications, both in military and civilian purposes the research on underwater acoustic has particular importance. This includes identification and tracking of ships or submarines via the noise radiated by their machinery components (Urick, 2008;Chen et al, 2000;Soares-Filho et al, 2001;Yang et al, 2002;Lennartsson et al, 2006;Rajagopal et al, 1990) and underwater acoustic communication (Luo et al, 2012;Diamant and Lampe, 2013). This also includes identifying marine mammals (Zimmer et al, 2008) and oceanography (Howell and Wood, 2003).…”