Classification of Sperm Whale Clicks (Physeter Macrocephalus) with Gaussian-Kernel-Based Networks

Schaar, Mike Connor Roger Malcolm Van der; Delory, Eric; André, M.

doi:10.3390/a2031232

Cited by 7 publications

(2 citation statements)

References 19 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The polynomial kernel function is a global kernel function with strong generalization ability but weak learning ability [36], whereas the Gaussian radial basis kernel function is a local kernel function with strong learning ability but weak generalization ability. It is difficult to obtain good results in regression forecasting [37] by using only a single kernel function.…”

Section: Construction Of Mixed Kernel Functionmentioning

confidence: 99%

A Forecast Model of the Number of Containers for Containership Voyage

2018

View full text Add to dashboard Cite

Container ships must pass through multiple ports of call during a voyage. Therefore, forecasting container volume information at the port of origin followed by sending such information to subsequent ports is crucial for container terminal management and container stowage personnel. Numerous factors influence container allocation to container ships for a voyage, and the degree of influence varies, engendering a complex nonlinearity. Therefore, this paper proposes a model based on gray relational analysis (GRA) and mixed kernel support vector machine (SVM) for predicting container allocation to a container ship for a voyage. First, in this model, the weights of influencing factors are determined through GRA. Then, the weighted factors serve as the input of the SVM model, and SVM model parameters are optimized through a genetic algorithm. Numerical simulations revealed that the proposed model could effectively predict the number of containers for container ship voyage and that it exhibited strong generalization ability and high accuracy. Accordingly, this model provides a new method for predicting container volume for a voyage.

show abstract

Section: Construction Of Mixed Kernel Functionmentioning

confidence: 99%

A Forecast Model of the Number of Containers for Containership Voyage

2018

View full text Add to dashboard Cite

show abstract

“…(Unpublished results). The classification and localisation modules have been successfully validated with several types of impulsive sounds [3,4,5,6,7]. A module for the classification of short tonal sounds is currently under development.…”

Section: Introductionmentioning

confidence: 99%

Automated real-time acoustic detection of Fin Whale calls at the deep sea floor observatory off Kushiro-Tokachi, Japan

André

Iwase

Akamatsu

et al. 2011

2011 IEEE Symposium on Underwater Technology and Workshop on Scientific Use of Submarine Cables and Related Technologies

Self Cite

View full text Add to dashboard Cite

A system for the real-time and fully automated detection, classification and localisation of acoustic sources in the marine environment has been developed at the Laboratory of Applied Bioacoustics (LAB) of the Technical University of Catalonia (UPC). The system features modules for the detection, the classification and localisation of short tonal sounds and impulsive sounds and the quantification of the level and the spectral characteristics of the background noise. The system has been successfully validated for the detection of many sound-types from fixed ocean cable observatories (ESONET, ANTARES, NEPTUNE). In particular, the detection module for short tonal sounds has been effectively applied to detect baleen whales and delphinid calls and whistles. The system is operational on live acoustic data streams from fixed ocean observatories (www.listentothedeep.org), demonstrating its ability to handle continuous data streams with a sampling frequency as high as 250 kHz. Here, we applied this module for the detection of short tonal sounds has been adjusted to detect calls of baleen whales in acoustic data from the sea floor observatory off Kushiro in the Kuril Trench, Japan (www.jamstec.go.jp/scdc/top_e.html). This data was recorded at a sampling frequency of 100 samples per second. The detector was first validated on data series from several marine observatories that were either known to contain Fin Whale calls or to be from regions where Fin Whales do not occur. The 384 hours of acoustic data used off Kushiro ranged from 25 th April to 10 th May 2010 without gaps. Many sections of these recording contained in the frequency band intense impulsive noise, or interferences, from unknown anthropogenic sources. Nevertheless, the system successfully detected a train of Fin Whale calls that occurred on April, the 29 th 2010 between 02h15-03h15. The call showed a frequency down sweep in the range 30-15 Hz with duration of 1-2 seconds. The signal to noise ratio of these calls, as perceived by an experienced operator, was clearly lower than in the validation data sets. This demonstrates (1) that Fin Whale are present off Kushiro, (2) that the detector can handle calls with a very low signal to noise ratio and (3) that the real-time automated detection of baleen whale calls can be performed live at JAMSTEC network of observatories.

show abstract

Environmental Acoustic Recording System (EARS) in the Gulf of Mexico

Ioup

Sidorovskaia

et al. 2016

Modern Acoustics and Signal Processing

View full text Add to dashboard Cite

Classification of Sperm Whale Clicks (Physeter Macrocephalus) with Gaussian-Kernel-Based Networks

Cited by 7 publications

References 19 publications

A Forecast Model of the Number of Containers for Containership Voyage

A Forecast Model of the Number of Containers for Containership Voyage

Automated real-time acoustic detection of Fin Whale calls at the deep sea floor observatory off Kushiro-Tokachi, Japan

Environmental Acoustic Recording System (EARS) in the Gulf of Mexico

Contact Info

Product

Resources

About