Intrinsic structure study of whale vocalizations

Yin Xian; Xiaobai Sun; Wenjing Liao; Yuan Zhang; Douglas Nowacek; Loren Nolte

doi:10.1109/oceans.2016.7761101

Intrinsic structure study of whale vocalizations

Yin Xian, Xiaobai Sun, Wenjing Liao, Yuan Zhang, Douglas Nowacek, Loren Nolte

School of Continuing Education

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

1 Citation (Scopus)

Abstract

Whale vocalizations can be modeled as polynomial-phase signals, which are widely used in radar and sonar applications. Such signals lie on a nonlinear manifold parameterized by polynomial phase coefficients. In this paper, we apply manifold learning methods, in particular ISOMAP and Laplacian Eigenmap, to examine the underlying geometric structure of whale vocalizations. We can improve the classification accuracy by using the intrinsic structure of whale vocalizations. Our experiments on the DCLDE conference and MobySound data show that manifold learning methods such as ISOMAP and Laplacian eigenmap outperform linear dimension reduction methods such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS).

Original language	English
Title of host publication	OCEANS 2016 MTS/IEEE Monterey
Publisher	IEEE
Number of pages	5
ISBN (Electronic)	9781509015375
ISBN (Print)	9781509015276
DOIs	https://doi.org/10.1109/oceans.2016.7761101
Publication status	Published - Sept 2016
Event	OCEANS 2016 MTS/IEEE Monterey - Monterey, United States Duration: 19 Sept 2016 → 23 Sept 2016 https://ieeexplore.ieee.org/xpl/conhome/7750936/proceeding

Publication series

Name	OCEANS

Conference

Conference	OCEANS 2016 MTS/IEEE Monterey
Country/Territory	United States
City	Monterey
Period	19/09/16 → 23/09/16
Internet address	https://ieeexplore.ieee.org/xpl/conhome/7750936/proceeding

User-Defined Keywords

Whale vocalizations
polynomial-phase signals
manifold learning
whale classification

Access to Document

10.1109/oceans.2016.7761101

Cite this

@inproceedings{1ca2d93d3f524f1c88ff2851205502fb,

title = "Intrinsic structure study of whale vocalizations",

abstract = "Whale vocalizations can be modeled as polynomial-phase signals, which are widely used in radar and sonar applications. Such signals lie on a nonlinear manifold parameterized by polynomial phase coefficients. In this paper, we apply manifold learning methods, in particular ISOMAP and Laplacian Eigenmap, to examine the underlying geometric structure of whale vocalizations. We can improve the classification accuracy by using the intrinsic structure of whale vocalizations. Our experiments on the DCLDE conference and MobySound data show that manifold learning methods such as ISOMAP and Laplacian eigenmap outperform linear dimension reduction methods such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS).",

keywords = "Whale vocalizations, polynomial-phase signals, manifold learning, whale classification",

author = "Yin Xian and Xiaobai Sun and Wenjing Liao and Yuan Zhang and Douglas Nowacek and Loren Nolte",

year = "2016",

month = sep,

doi = "10.1109/oceans.2016.7761101",

language = "English",

isbn = "9781509015276",

series = "OCEANS",

publisher = "IEEE",

booktitle = "OCEANS 2016 MTS/IEEE Monterey",

address = "United States",

note = "OCEANS 2016 MTS/IEEE Monterey ; Conference date: 19-09-2016 Through 23-09-2016",

url = "https://ieeexplore.ieee.org/xpl/conhome/7750936/proceeding",

}

TY - GEN

T1 - Intrinsic structure study of whale vocalizations

AU - Xian, Yin

AU - Sun, Xiaobai

AU - Liao, Wenjing

AU - Zhang, Yuan

AU - Nowacek, Douglas

AU - Nolte, Loren

PY - 2016/9

Y1 - 2016/9

N2 - Whale vocalizations can be modeled as polynomial-phase signals, which are widely used in radar and sonar applications. Such signals lie on a nonlinear manifold parameterized by polynomial phase coefficients. In this paper, we apply manifold learning methods, in particular ISOMAP and Laplacian Eigenmap, to examine the underlying geometric structure of whale vocalizations. We can improve the classification accuracy by using the intrinsic structure of whale vocalizations. Our experiments on the DCLDE conference and MobySound data show that manifold learning methods such as ISOMAP and Laplacian eigenmap outperform linear dimension reduction methods such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS).

AB - Whale vocalizations can be modeled as polynomial-phase signals, which are widely used in radar and sonar applications. Such signals lie on a nonlinear manifold parameterized by polynomial phase coefficients. In this paper, we apply manifold learning methods, in particular ISOMAP and Laplacian Eigenmap, to examine the underlying geometric structure of whale vocalizations. We can improve the classification accuracy by using the intrinsic structure of whale vocalizations. Our experiments on the DCLDE conference and MobySound data show that manifold learning methods such as ISOMAP and Laplacian eigenmap outperform linear dimension reduction methods such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS).

KW - Whale vocalizations

KW - polynomial-phase signals

KW - manifold learning

KW - whale classification

U2 - 10.1109/oceans.2016.7761101

DO - 10.1109/oceans.2016.7761101

M3 - Conference proceeding

SN - 9781509015276

T3 - OCEANS

BT - OCEANS 2016 MTS/IEEE Monterey

PB - IEEE

T2 - OCEANS 2016 MTS/IEEE Monterey

Y2 - 19 September 2016 through 23 September 2016

ER -

Intrinsic structure study of whale vocalizations

Abstract

Publication series

Conference

User-Defined Keywords

Access to Document

Fingerprint

Cite this