Quantifying the Effects of Propagation on Classification of Cetacean Vocalizations – Final Report

PDF

Authors
  1. Hines, P.C.
  2. Binder, C.M.
Corporate Authors
Defence Research and Development Canada, Atlantic Research Centre, Halifax NS (CAN);Dalhousie Univ, Halifax NS (CAN) Dept of Oceanography;DALHOUSIE UNIV, HALIFAX NS (CAN) DEPT OF ELECTRICAL AND COMPUTER ENGINEERING
Abstract
In previous work a unique automatic classifier that uses perceptual signal features—features similar to those employed by the human auditory system—was employed to successfully classify anthropogenic transients, and vocalizations from five cetacean species. Although this is a significant achievement, successful implementation of this (or any) classifier requires that it be temporally and spatially robust. The primary goal of this research was to address the question: “Will the aural classifier work on vocalization data from these species collected under different environmental conditions?” This was done by performing a propagation experiment using cetacean vocalizations and synthetically generated calls as source signals, and testing the received signals with the classifier. The measurements were complemented by comparing the experimental results to propagation model results to examine the impact of propagation on classifier performance. Lastly metrics were identified to determine if individual features were altered in a statistically significant way as signals propagated over increasingly longer ranges. Experimental results indicated a decrease in classifier performance as propagation range increased. Through simulation, it was found that decreasing signal-to-noise ratio was the dominating influence; however, additional propagation issues—like multipath propagation—did contribute to decreasing classifier performance due to their impact on the individual perceptua

Il y a un résumé en français ici.

Keywords
passive acoustic monitoring;marine mammals;acoustic propagation;aural classifier;perceptual features;ONR external client report
Report Number
DRDC-RDDC-2017-R126 — External Client Report
Date of publication
01 Nov 2017
Number of Pages
21
DSTKIM No
CA045505
CANDIS No
805920
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: