Collection: Anomalous spectrograms obtained from Ocean Networks Canada's hydrophones deployed along the North Pacific and Canadian Arctic 2014-2021, used as an additional test for an autoencoder model developed to find anomalous hydrophone data

Citation

Ocean Networks Canada Society. 2022. Collection: Anomalous spectrograms obtained from Ocean Networks Canada's hydrophones deployed along the North Pacific and Canadian Arctic 2014-2021, used as an additional test for an autoencoder model developed to find anomalous hydrophone data. Ocean Networks Canada Society. https://doi.org/10.26152/w90h-mj32

Title

Collection: Collection: Anomalous spectrograms obtained from Ocean Networks Canada's hydrophones deployed along the North Pacific and Canadian Arctic 2014-2021, used as an additional test for an autoencoder model developed to find anomalous hydrophone data

DOI

10.26152/w90h-mj32

This information is maintained also at https://doi.datacite.org/dois/10.26152%2Fw90h-mj32.

Abstract

This collection contains a dataset consisting of 161 anomalous spectrograms found within Ocean Networks Canada’s hydrophone data. Each were selected through careful curation by a hydrophone specialist, and contain spurious signals such as frequency drops, electrical noise and other non-acoustic phenomena that rarely occur, yet suggest issues that need further investigation. These spectrograms come from the hydrophones located in the North Pacific (Barkley Canyon, Saanich Inlet, Folger Passage, Clayoquot Slope and Strait of Georgia) and in the Canadian Arctic at Cambridge Bay between 2014 to 2021. This data is used as an additional test for the autoencoder model developed to find anomalous hydrophone spectrograms that is described in the publication “Anomaly detection in complex data: a practical application when outliers are few” by Engida, Z.,  Foloni-Neto, H., Slonimer, A., Bedard, J., Alam, F. S., Snauffer, A. (2022).

Creator

Ocean Networks Canada Society

Date Created

2022-08-30

Funding References

Not listed

Publisher

Ocean Networks Canada Society

Publication Year

2022

Resource Type

Collection

Rights

Creative Commons Attribution 4.0 International

For more details, please refer to our data policy page: https://www.oceannetworks.ca/data/data-policy/

Formats

PNG

Geolocations - Bounding Box

North: 69.11268462

East: -126.8724249

South: 48.345174

West: -105.06407

Contributors

Component Datasets

Version History

Related Publications

Z. Engida, H. F. Neto, A. Slonimer, J. Bedard, F. S. Alam and A. M. Snauffer, "Anomaly detection in complex data: a practical application when outliers are few," OCEANS 2022, Hampton Roads, Hampton Roads, VA, USA, 2022, pp. 1-7, doi: 10.1109/OCEANS47191.2022.9977351.