• Baccini, A., M. A. Friedl, C. E. Woodcock, and R. Warbington, 2004: Forest biomass estimation over regional scales using multisource data. Geophys. Res. Lett., 31, L10501, https://doi.org/10.1029/2004GL019782.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Blackman, S., 1986: Multiple-Target Tracking with Radar Applications. Artech House, 449 pp.

  • Breiman, L., 2001: Random forests. Mach. Learn., 45, 532, https://doi.org/10.1023/A:1010933404324.

  • Burges, C. J. C., 1998: A tutorial on support vector machines for pattern recognition. Data Min. Knowl. Discovery, 2, 121167, https://doi.org/10.1023/A:1009715923555.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Calambokidis, J., B. L. Taylor, S. D. Carter, G. H. Steiger, P. K. Dawson, and L. D. Antrim, 1987: Distribution and haul-out behavior of harbor seals in Glacier Bay, Alaska. Can. J. Zool., 65, 13911396, https://doi.org/10.1139/z87-219.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Copping, A., and et al. , 2016: Annex IV 2016 state of the science report: Environmental effects of marine renewable energy development around the world. Pacific Northwest National Laboratory Tech. Rep., 224 pp., https://tethys.pnnl.gov/sites/default/files/publications/Annex-IV-2016-State-of-the-Science-Report_MR.pdf.

  • Cotter, E., P. Murphy, and B. Polagye, 2017: Benchmarking sensor fusion capabilities of an integrated instrumentation package. Int. J. Mar. Energy, 20, 6479, https://doi.org/10.1016/j.ijome.2017.09.003.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dobeck, G. J., J. C. Hyland, and L. Smedley, 1997: Automated detection/classification of sea mines in sonar imagery. Proc. SPIE, 3079, 90110, https://doi.org/10.1117/12.280846.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Duan, K., and S. S. Keerthi, 2005: Which is the best multiclass SVM method? An empirical study. Multiple Classifier Systems, N. C. Oza et al., Eds., Springer, 278–285.

    • Crossref
    • Export Citation
  • Dudani, S. A., 1976: The distance-weighted k-nearest-neighbor rule. IEEE Trans. Syst. Man Cybern., SMC-6, 325327, https://doi.org/10.1109/TSMC.1976.5408784.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fawcett, T., 2006: An introduction to ROC analysis. Pattern Recognit. Lett., 27, 861874, https://doi.org/10.1016/j.patrec.2005.10.010.

  • Foote, K. G., D. Chu, T. R. Hammar, K. C. Baldwin, L. A. Mayer, L. C. Hufnagle, and J. M. Jech, 2005: Protocols for calibrating multibeam sonar. J. Acoust. Soc. Amer., 117, 20132027, https://doi.org/10.1121/1.1869073.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francisco, F., and J. Sundberg, 2019: Detection of visual signatures of marine mammals and fish within marine renewable energy farms using multibeam imaging sonar. J. Mar. Sci. Eng., 7, 22, https://doi.org/10.3390/JMSE7020022.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grellier, K., P. M. Thompson, and H. M. Corpe, 1996: The effect of weather conditions on harbour seal (Phoca vitulina) haulout behaviour in the Moray Firth, northeast Scotland. Can. J. Zool., 74, 18061811, https://doi.org/10.1139/z96-201.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hamilton, C. D., C. Lydersen, R. A. Ims, and K. M. Kovacs, 2014: Haul-out behaviour of the world’s northernmost population of harbour seals (Phoca vitulina) throughout the year. PLOS ONE, 9, e86055, https://doi.org/10.1371/journal.pone.0086055.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hastie, G. D., D. Gillespie, J. Gordon, J. Macaulay, B. McConnel, and C. Sparling, 2014: Tracking technologies for quantifying marine mammal interactions with tidal turbines: Pitfalls and possibilities. Marine Renewable Energy Technology and Environmental Interactions, M. A. Shields and A. I. L. Payne, Eds., Springer, 127–139.

    • Crossref
    • Export Citation
  • Hastie, G. D., and et al. , 2019: Automated detection and tracking of marine mammals: A novel sonar tool for monitoring effects of marine industry. Aquat. Conserv., 29, 119130, https://doi.org/10.1002/aqc.3103.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Japkowicz, N., 2000: The class imbalance problem: Significance and strategies. Proc. 2000 Int. Conf. on Artificial Intelligence, Las Vegas, Nevada, ICAI, 111–117, http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.35.1693&rep=rep1&type=pdf.

  • Jepp, P., 2017: Target tracking using sonars for marine life monitoring around tidal turbines. Proc. 12th European Wave and Tidal Energy Conf., Cork, Ireland, University of Southampton, https://tethys.pnnl.gov/publications/target-tracking-using-sonars-marine-life-monitoring-around-tidal-turbines.

  • Kasutani, E., and A. Yamada, 2001: The MPEG-7 color layout descriptor: A compact image feature description for high-speed image/video segment retrieval. Proc. 2001 Int. Conf. on Image Processing, Thessaloniki, Greece, IEEE, 674–677, https://doi.org/10.1109/ICIP.2001.959135.

    • Crossref
    • Export Citation
  • Kohavi, R., and G. H. John, 1997: Wrappers for feature subset selection. Artif. Intell., 97, 273324, https://doi.org/10.1016/S0004-3702(97)00043-X.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lieber, L., T. Nilsen, C. Zambrano, and L. Kregting, 2017: Optimising multiple multibeam sonars to assess marine life interactions with an underwater kite. Proc. 12th European Wave and Tidal Energy Conf., Cork, Ireland, University of Southampton, https://tethys.pnnl.gov/publications/optimising-multiple-multibeam-sonars-assess-marine-life-interactions-underwater-kite.

  • Melvin, G. D., 2016: Observations of in situ Atlantic bluefin tuna (Thunnus thynnus) with 500-khz multibeam sonar. ICES J. Mar. Sci., 73, 19751986, https://doi.org/10.1093/icesjms/fsw077.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Melvin, G. D., and N. A. Cochrane, 2015: Multibeam acoustic detection of fish and water column targets at high-flow sites. Estuaries Coasts, 38, 227240, https://doi.org/10.1007/s12237-014-9828-z.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Platt, J. C., 1999: Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in Large Margin Classifiers, A. J. Smola et al., Eds., MIT Press, 61–74.

  • Polagye, B., and J. Thomson, 2013: Tidal energy resource characterization: Methodology and field study in Admiralty Inlet, Puget Sound, WA (USA). Proc. Inst. Mech. Eng., 227A, 352367, https://doi.org/10.1177/0957650912470081.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rosa, I. M. D., A. T. Marques, G. Palminha, H. Costa, M. Mascarenhas, C. Fonseca, and J. Bernardino, 2016: Classification success of six machine learning algorithms in radar ornithology. Int. J. Avian Sci., 158, 2842, https://doi.org/10.1111/ibi.12333.

    • Search Google Scholar
    • Export Citation
  • Rothery, P., and D. B. Roy, 2001: Application of generalized additive models to butterfly transect count data. J. Appl. Stat., 28, 897909, https://doi.org/10.1080/02664760120074979.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schneider, D. C., and P. M. Payne, 1983: Factors affecting haul-out of harbor seals at a site in southeastern Massachusetts. J. Mammal., 64, 518520, https://doi.org/10.2307/1380370.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Song, F., Z. Guo, and D. Mei, 2010: Feature selection using principal component analysis. Int. Conf. on System Science, Engineering Design and Manufacturing Informatization, Yichang, China, IEEE, 27–30, https://doi.org/10.1109/ICSEM.2010.14.

    • Crossref
    • Export Citation
  • Thyng, K. M., C. A. Greene, R. D. Hetland, H. M. Zimmerle, and S. F. DiMarco, 2016: True colors of oceanography: Guidelines for effective and accurate colormap selection. Oceanography, 29 (3), 913, https://doi.org/10.5670/oceanog.2016.66.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Trenkel, V. M., V. Mazauric, and L. Berger, 2008: The new fisheries multibeam echosounder ME70: Description and expected contribution to fisheries research. ICES J. Mar. Sci., 65, 645655, https://doi.org/10.1093/icesjms/fsn051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Urban, P., K. Köser, and J. Greinert, 2017: Processing of multibeam water column image data for automated bubble/seep detection and repeated mapping. Limnol. Oceanogr. Methods, 15, 121, https://doi.org/10.1002/lom3.10138.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Viehman, H. A., and G. B. Zydlewski, 2014: Fish interactions with a commercial-scale tidal energy device in the natural environment. Estuaries Coasts, 38, 241252, https://doi.org/10.1007/S12237-014-9767-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wallace, B. C., K. Small, C. E. Brodley, and T. A. Trikalinos, 2011: Class imbalance, redux. 2011 IEEE 11th Int. Conf. on Data Mining, Vancouver, BC, Canada, IEEE, 754–763, https://doi.org/10.1109/ICDM.2011.33.

    • Crossref
    • Export Citation
  • Wilding, T. A., and et al. , 2017: Turning off the DRIP (‘data-rich, information-poor’)—Rationalising monitoring with a focus on marine renewable energy developments and the benthos. Renewable Sustainable Energy Rev., 74, 848859, https://doi.org/10.1016/j.rser.2017.03.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Williamson, B., S. Fraser, P. Blondel, P. S. Bell, J. J. Waggit, and B. E. Scott, 2017: Multisensor acoustic tracking of fish and seabird behavior around tidal turbine structures in Scotland. IEEE J. Ocean Eng., 42, 948965, https://doi.org/10.1109/joe.2016.2637179.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, M., and Z. Zhou, 2005: A k-nearest neighbor based algorithm for multi-label classification. Int. Conf. on Granular Computing, Beijing, China, IEEE, 718–721, https://doi.org/10.1109/GRC.2005.1547385.

    • Crossref
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 184 184 17
Full Text Views 27 27 4
PDF Downloads 32 32 3

Automatic Classification of Biological Targets in a Tidal Channel Using a Multibeam Sonar

View More View Less
  • 1 University of Washington, Seattle, Washington
© Get Permissions
Restricted access

Abstract

Multibeam sonars are widely used for environmental monitoring of fauna at marine renewable energy sites. However, they can rapidly accrue vast volumes of data, which poses a challenge for data processing. Here, using data from a deployment in a tidal channel with peak currents of 1–2 m s−1, we demonstrate the data-reduction benefits of real-time automatic classification of targets detected and tracked in multibeam sonar data. First, we evaluate classification capabilities for three machine learning algorithms: random forests, support vector machines, and k-nearest neighbors. For each algorithm, a hill-climbing search optimizes a set of hand-engineered attributes that describe tracked targets. The random forest algorithm is found to be most effective—in postprocessing, discriminating between biological and nonbiological targets with a recall rate of 0.97 and a precision of 0.60. In addition, 89% of biological targets are correctly classified as either seals, diving birds, fish schools, or small targets. Model dependence on the volume of training data is evaluated. Second, a real-time implementation of the model is shown to distinguish between biological targets and nonbiological targets with nearly the same performance as in postprocessing. From this, we make general recommendations for implementing real-time classification of biological targets in multibeam sonar data and the transferability of trained models.

Corresponding author: Emma Cotter, ecotter@whoi.edu

Abstract

Multibeam sonars are widely used for environmental monitoring of fauna at marine renewable energy sites. However, they can rapidly accrue vast volumes of data, which poses a challenge for data processing. Here, using data from a deployment in a tidal channel with peak currents of 1–2 m s−1, we demonstrate the data-reduction benefits of real-time automatic classification of targets detected and tracked in multibeam sonar data. First, we evaluate classification capabilities for three machine learning algorithms: random forests, support vector machines, and k-nearest neighbors. For each algorithm, a hill-climbing search optimizes a set of hand-engineered attributes that describe tracked targets. The random forest algorithm is found to be most effective—in postprocessing, discriminating between biological and nonbiological targets with a recall rate of 0.97 and a precision of 0.60. In addition, 89% of biological targets are correctly classified as either seals, diving birds, fish schools, or small targets. Model dependence on the volume of training data is evaluated. Second, a real-time implementation of the model is shown to distinguish between biological targets and nonbiological targets with nearly the same performance as in postprocessing. From this, we make general recommendations for implementing real-time classification of biological targets in multibeam sonar data and the transferability of trained models.

Corresponding author: Emma Cotter, ecotter@whoi.edu
Save