EMD-Based Energy Spectrum Entropy Distribution Signal Detection Methods for Marine Mammal Vocalizations.

To develop a passive acoustic monitoring system for diversity detection and thereby adapt to the challenges of a complex marine environment, this study harnesses the advantages of empirical mode decomposition in analyzing nonstationary signals and introduces energy characteristics analysis and entropy of information theory to detect marine mammal vocalizations. The proposed detection algorithm has five main steps: sampling, energy characteristics analysis, marginal frequency distribution, feature extraction, and detection, which involve four signal feature extraction and analysis algorithms: energy ratio distribution (ERD), energy spectrum distribution (ESD), energy spectrum entropy distribution (ESED), and concentrated energy spectrum entropy distribution (CESED). In an experiment on 500 sampled signals (blue whale vocalizations), in the competent intrinsic mode function (IMF2) signal feature extraction function distribution of ERD, ESD, ESED, and CESED, the areas under the curves (AUCs) of the receiver operating characteristic (ROC) curves were 0.4621, 0.6162, 0.3894, and 0.8979, respectively; the Accuracy scores were 49.90%, 60.40%, 47.50%, and 80.84%, respectively; the Precision scores were 31.19%, 44.89%, 29.44%, and 68.20%, respectively; the Recall scores were 42.83%, 57.71%, 36.00%, and 84.57%, respectively; and the F1 scores were 37.41%, 50.50%, 32.39%, and 75.51%, respectively, based on the threshold of the optimal estimated results. It is clear that the CESED detector outperforms the other three detectors in signal detection and achieves efficient sound detection of marine mammals.

Extraction of Energy Characteristics of Blue Whale Vocalizations Based on Empirical Mode Decomposition.

This study extracts the energy characteristic distributions of the intrinsic mode functions (IMFs) and residue functions (RF) for a blue whale sound signal, with empirical mode decomposition (EMD) as the basic theoretical framework. A high-resolution marginal frequency characteristics extraction method, based on EMD with energy density intensity (EDI) parameters for blue B call vocalizations, was proposed. The extraction algorithm included six steps: EMD, energy analysis, marginal frequency (MF) analysis with EDI parameters, feature extraction (FE), classification, and Hilbert spectrum (HS) analysis. The blue whale sound sources were obtained from the website of the Scripps Whale Acoustics Lab of the University of California, San Diego, USA. The source is a type of B call with a time duration of 46.65 s, from which 59 analysis samples with a time duration of 180 ms were taken. The average energy distribution ratios of the IMF1, IMF2, IMF3, IMF4, and RF are 49.06%, 20.58%, 13.51%, 10.94% and 3.84%, respectively. New classification criteria and EDI parameters were proposed to extract the blue whale B call vocalization (BWBCV) characteristics. The analysis results show that the main frequency bands of the signal are distributed at 41-43 Hz in the MF of IMF1 for Class I BWBCV and 11-13 Hz in the MF of IMF2 for Class II BWBCV, respectively.