Deep Active Learning for In Situ Plankton Classification

  • Erik BochinskiEmail author
  • Ghassen Bacha
  • Volker Eiselein
  • Tim J. W. Walles
  • Jens C. Nejstgaard
  • Thomas Sikora
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11188)


Ecological studies of some of the most numerous organisms on the planet, zooplankton, have been limited by manual analysis for more than 100 years. With the development of high-throughput video systems, we argue that this critical bottle-neck can now be solved if paired with deep neural networks (DNN). To leverage their performance, large amounts of training samples are required that until now have been dependent on manually created labels. To minimize the effort of expensive human experts, we employ recent active learning approaches to select only the most informative samples for labelling. Thus training a CNN using a nearly unlimited amount of images while limiting the human labelling effort becomes possible by means of active learning. We show in several experiments that in practice, only a few thousand labels are required to train a CNN and achieve an accuracy-level comparable to manual routine analysis of zooplankton samples. Once trained, this CNN can be used to analyse any amount of image data, presenting the zooplankton community the opportunity to address key research questions on transformative scales, many orders of magnitude, in both time and space, basically only limited by video through-put and compute capacity.


Classification Zooplankton Active learning Automatic identification and sizing Cost-effective active learning In situ 


  1. 1.
    Benfield, M.C., et al.: RAPID: research on automated plankton identification. Oceanography 20, 172–187 (2007)CrossRefGoogle Scholar
  2. 2.
    Connor, R.: The United Nations World Water Development Report 2015: Water for a Sustainable World. UNESCO Publishing, Paris (2015)Google Scholar
  3. 3.
    Cowen, R.K., Guigand, C.M.: In situ ichthyoplankton imaging system (ISIIS): system design and preliminary results. Limnol. Oceanogr. Methods 6, 126–132 (2008)CrossRefGoogle Scholar
  4. 4.
    Dai, J., Wang, R., Zheng, H., Ji, G., Qiao, X.: ZooplanktoNet: deep convolutional network for zooplankton classification. In: OCEANS (2016)Google Scholar
  5. 5.
    Douglas, K., Juniper, S., Jenkyns, R., Hoeberechts, M., Macoun, P., Hillier, J.: Developing spatial management tools for offshore marine protected areas. In: OCEANS-Anchorage (2017)Google Scholar
  6. 6.
    Hofmann, E.E., Klinck, J.M.: Future marine zooplankton research - a perspective. Marine Ecol. Progress Ser. 222, 297–308 (2001)CrossRefGoogle Scholar
  7. 7.
    Gorriz, M., Carlier, A., Faure, E., Giro-i Nieto, X.: Cost-effective active learning for melanoma segmentation. arXiv preprint arXiv:1711.09168 (2017)
  8. 8.
    Gorsky, G., et al.: Digital zooplankton image analysis using the zooscan integrated system. J. Plankton Res. 32, 285–303 (2010)CrossRefGoogle Scholar
  9. 9.
    Hirata, N.S., Fernandez, M.A., Lopes, R.M.: Plankton image classification based on multiple segmentations. In: ICPR CVAUI Workshop (2016)Google Scholar
  10. 10.
    Kingma, D., Ba, J.: Adam: a method for stochastic optimization. In: International Conference for Learning Representations (2015)Google Scholar
  11. 11.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: NIPS (2012)Google Scholar
  12. 12.
    Lee, H., Park, M., Kim, J.: Plankton classification on imbalanced large scale database via convolutional neural networks with transfer learning. In: IEEE International Conference on Image Processing (2016)Google Scholar
  13. 13.
    Lin, L., Wang, K., Meng, D., Zuo, W., Zhang, L.: Active self-paced learning for cost-effective and progressive face identification. IEEE Trans. Pattern Anal. Mach. Intell. 40, 7–19 (2018)CrossRefGoogle Scholar
  14. 14.
    MacLeod, N., Benfield, M., Culverhouse, P.: Time to automate identification. Nature 467, 154 (2010)CrossRefGoogle Scholar
  15. 15.
    Orenstein, E.C., Beijbom, O., Peacock, E.E., Sosik, H.M.: WHOI-plankton-a large scale fine grained visual recognition benchmark dataset for plankton classification. arXiv preprint arXiv:1510.00745 (2015)
  16. 16.
    Otsu, N.: A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern. 9, 62–66 (1979)CrossRefGoogle Scholar
  17. 17.
    Scheffer, T., Decomain, C., Wrobel, S.: Active hidden Markov models for information extraction. In: International Symposium on Intelligent Data Analysis (2001)CrossRefGoogle Scholar
  18. 18.
    Settles, B.: Active learning literature survey. Computer Sciences Technical report 1648 (2010)Google Scholar
  19. 19.
    Shannon, C.E.: A mathematical theory of communication. ACM SIGMOBILE Mob. Comput. Commun. Rev. 5, 3–55 (2001)CrossRefGoogle Scholar
  20. 20.
    Stanitsas, P., Cherian, A., Truskinovsky, A., Morellas, V., Papanikolopoulos, N.: Active convolutional neural networks for cancerous tissue recognition. In: IEEE International Conference on Image Processing (2017)Google Scholar
  21. 21.
    Vörösmarty, C.J., et al.: Global threats to human water security and river biodiversity. Nature 467, 555 (2010)CrossRefGoogle Scholar
  22. 22.
    Wang, C., Yu, Z., Zheng, H., Wang, N., Zheng, B.: CGAN-plankton: towards large-scale imbalanced class generation and fine-grained classification. In: IEEE International Conference on Image Processing (2017)Google Scholar
  23. 23.
    Wang, K., Zhang, D., Li, Y., Zhang, R., Lin, L.: Cost-effective active learning for deep image classification. IEEE TCSVT 27, 2591–2600 (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Erik Bochinski
    • 1
    Email author
  • Ghassen Bacha
    • 1
  • Volker Eiselein
    • 1
  • Tim J. W. Walles
    • 2
  • Jens C. Nejstgaard
    • 2
  • Thomas Sikora
    • 1
  1. 1.Communication System GroupTechnische Universität BerlinBerlinGermany
  2. 2.Leibnitz-Institute of Freshwater Ecology and Inland FisheriesStechlinGermany

Personalised recommendations