PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild

Research output: Contribution to conferencePoster

Abstract

Harbour porpoises (Phocoena phocoena) are difficult to observe at sea, even with good weather conditions, due to their small size and cryptic behaviour. However, they are highly vocal, producing narrow-band high-frequency (NBHF) echolocation clicks, which makes them well suited for passive acoustic monitoring (PAM). Such PAM systems must be coupled with a classification algorithm to identify the likely porpoise signals among other transient signals. We present a new harbour porpoise click classifier (PorCC) for full-waveform signals, with an improved performance over current classifiers. PorCC was developed in MATLAB and uses the coefficients of two logistic regression models in a decision-making pathway to assign each signal to one of three categories: high-quality click (HQ), low-quality click (LQ), or high-frequency noise (N). The first model uses click duration and QRMS (RMS-bandwidth / centroid frequency) to separate HQ from N. The second model uses click duration, QRMS, ratio between peak and centroid frequency, peak cross-correlation coefficient (against a model click), centroid frequency, and -3dB bandwidth to separate LQ from N. PorCC achieved hit rates > 90% for HQ clicks while keeping false alarm levels < 1%. The performance of PorCC was compared to that of PAMGuard’s Porpoise Classifier module (with standard settings) and the receiver operating characteristics curve was generated for both classifiers. The precision for HQ (HQ clips classified as HQ / total clips classified as HQ) was 31.2% for PAMGuard and 96.1% for PorCC, and the detectability index (d’) was 2.2 for PAMGuard versus 4.1 for PorCC. Results of this study show PorCC is a rapid, highly accurate method to classify NBHF clicks, which could be applied for real time monitoring, as well as to study harbour porpoises, and potentially other NBHF species, throughout their distribution range from data collected using towed hydrophones or static recorders.

Other

OtherThe 15th Danish Marine Mammal Symposium
CountryDenmark
CityOdense
Period23/01/1925/01/19

Fingerprint

Ports and harbors
Classifiers
Monitoring
Acoustics
Bandwidth
Hydrophones
MATLAB
Logistics
Decision making

Keywords

  • harbour porpoise
  • click classifier
  • PorCC

Cite this

Cosentino, M., Guarato, F., Tougaard, J., Nairn, D., Jackson, J. C., & Windmill, J. F. C. (2019). PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild. Poster session presented at The 15th Danish Marine Mammal Symposium, Odense, Denmark.
Cosentino, Melania ; Guarato, Francesco ; Tougaard, Jakob ; Nairn, David ; Jackson, Joseph C. ; Windmill, James F. C. / PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild. Poster session presented at The 15th Danish Marine Mammal Symposium, Odense, Denmark.
@conference{d570f564b8ef450a9c8102c58a7aff0e,
title = "PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild",
abstract = "Harbour porpoises (Phocoena phocoena) are difficult to observe at sea, even with good weather conditions, due to their small size and cryptic behaviour. However, they are highly vocal, producing narrow-band high-frequency (NBHF) echolocation clicks, which makes them well suited for passive acoustic monitoring (PAM). Such PAM systems must be coupled with a classification algorithm to identify the likely porpoise signals among other transient signals. We present a new harbour porpoise click classifier (PorCC) for full-waveform signals, with an improved performance over current classifiers. PorCC was developed in MATLAB and uses the coefficients of two logistic regression models in a decision-making pathway to assign each signal to one of three categories: high-quality click (HQ), low-quality click (LQ), or high-frequency noise (N). The first model uses click duration and QRMS (RMS-bandwidth / centroid frequency) to separate HQ from N. The second model uses click duration, QRMS, ratio between peak and centroid frequency, peak cross-correlation coefficient (against a model click), centroid frequency, and -3dB bandwidth to separate LQ from N. PorCC achieved hit rates > 90{\%} for HQ clicks while keeping false alarm levels < 1{\%}. The performance of PorCC was compared to that of PAMGuard’s Porpoise Classifier module (with standard settings) and the receiver operating characteristics curve was generated for both classifiers. The precision for HQ (HQ clips classified as HQ / total clips classified as HQ) was 31.2{\%} for PAMGuard and 96.1{\%} for PorCC, and the detectability index (d’) was 2.2 for PAMGuard versus 4.1 for PorCC. Results of this study show PorCC is a rapid, highly accurate method to classify NBHF clicks, which could be applied for real time monitoring, as well as to study harbour porpoises, and potentially other NBHF species, throughout their distribution range from data collected using towed hydrophones or static recorders.",
keywords = "harbour porpoise, click classifier, PorCC",
author = "Melania Cosentino and Francesco Guarato and Jakob Tougaard and David Nairn and Jackson, {Joseph C.} and Windmill, {James F. C.}",
year = "2019",
month = "1",
day = "25",
language = "English",
note = "The 15th Danish Marine Mammal Symposium ; Conference date: 23-01-2019 Through 25-01-2019",

}

Cosentino, M, Guarato, F, Tougaard, J, Nairn, D, Jackson, JC & Windmill, JFC 2019, 'PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild' The 15th Danish Marine Mammal Symposium, Odense, Denmark, 23/01/19 - 25/01/19, .

PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild. / Cosentino, Melania; Guarato, Francesco; Tougaard, Jakob; Nairn, David; Jackson, Joseph C.; Windmill, James F. C.

2019. Poster session presented at The 15th Danish Marine Mammal Symposium, Odense, Denmark.

Research output: Contribution to conferencePoster

TY - CONF

T1 - PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild

AU - Cosentino, Melania

AU - Guarato, Francesco

AU - Tougaard, Jakob

AU - Nairn, David

AU - Jackson, Joseph C.

AU - Windmill, James F. C.

PY - 2019/1/25

Y1 - 2019/1/25

N2 - Harbour porpoises (Phocoena phocoena) are difficult to observe at sea, even with good weather conditions, due to their small size and cryptic behaviour. However, they are highly vocal, producing narrow-band high-frequency (NBHF) echolocation clicks, which makes them well suited for passive acoustic monitoring (PAM). Such PAM systems must be coupled with a classification algorithm to identify the likely porpoise signals among other transient signals. We present a new harbour porpoise click classifier (PorCC) for full-waveform signals, with an improved performance over current classifiers. PorCC was developed in MATLAB and uses the coefficients of two logistic regression models in a decision-making pathway to assign each signal to one of three categories: high-quality click (HQ), low-quality click (LQ), or high-frequency noise (N). The first model uses click duration and QRMS (RMS-bandwidth / centroid frequency) to separate HQ from N. The second model uses click duration, QRMS, ratio between peak and centroid frequency, peak cross-correlation coefficient (against a model click), centroid frequency, and -3dB bandwidth to separate LQ from N. PorCC achieved hit rates > 90% for HQ clicks while keeping false alarm levels < 1%. The performance of PorCC was compared to that of PAMGuard’s Porpoise Classifier module (with standard settings) and the receiver operating characteristics curve was generated for both classifiers. The precision for HQ (HQ clips classified as HQ / total clips classified as HQ) was 31.2% for PAMGuard and 96.1% for PorCC, and the detectability index (d’) was 2.2 for PAMGuard versus 4.1 for PorCC. Results of this study show PorCC is a rapid, highly accurate method to classify NBHF clicks, which could be applied for real time monitoring, as well as to study harbour porpoises, and potentially other NBHF species, throughout their distribution range from data collected using towed hydrophones or static recorders.

AB - Harbour porpoises (Phocoena phocoena) are difficult to observe at sea, even with good weather conditions, due to their small size and cryptic behaviour. However, they are highly vocal, producing narrow-band high-frequency (NBHF) echolocation clicks, which makes them well suited for passive acoustic monitoring (PAM). Such PAM systems must be coupled with a classification algorithm to identify the likely porpoise signals among other transient signals. We present a new harbour porpoise click classifier (PorCC) for full-waveform signals, with an improved performance over current classifiers. PorCC was developed in MATLAB and uses the coefficients of two logistic regression models in a decision-making pathway to assign each signal to one of three categories: high-quality click (HQ), low-quality click (LQ), or high-frequency noise (N). The first model uses click duration and QRMS (RMS-bandwidth / centroid frequency) to separate HQ from N. The second model uses click duration, QRMS, ratio between peak and centroid frequency, peak cross-correlation coefficient (against a model click), centroid frequency, and -3dB bandwidth to separate LQ from N. PorCC achieved hit rates > 90% for HQ clicks while keeping false alarm levels < 1%. The performance of PorCC was compared to that of PAMGuard’s Porpoise Classifier module (with standard settings) and the receiver operating characteristics curve was generated for both classifiers. The precision for HQ (HQ clips classified as HQ / total clips classified as HQ) was 31.2% for PAMGuard and 96.1% for PorCC, and the detectability index (d’) was 2.2 for PAMGuard versus 4.1 for PorCC. Results of this study show PorCC is a rapid, highly accurate method to classify NBHF clicks, which could be applied for real time monitoring, as well as to study harbour porpoises, and potentially other NBHF species, throughout their distribution range from data collected using towed hydrophones or static recorders.

KW - harbour porpoise

KW - click classifier

KW - PorCC

UR - https://www.dhm2019.dk/

M3 - Poster

ER -

Cosentino M, Guarato F, Tougaard J, Nairn D, Jackson JC, Windmill JFC. PorCC: a new high-accuracy click classifier to study harbour porpoises in the wild. 2019. Poster session presented at The 15th Danish Marine Mammal Symposium, Odense, Denmark.