08 Jul 2020
 | 08 Jul 2020
Status: this preprint was under review for the journal OS but the revision was not accepted.

Towards operational phytoplankton recognition with automated high-throughput imaging and compact convolutional neural networks

Tuomas Eerola, Kaisa Kraft, Osku Grönberg, Lasse Lensu, Sanna Suikkanen, Jukka Seppälä, Timo Tamminen, Heikki Kälviäinen, and Heikki Haario

Abstract. Plankton communities form the basis of aquatic ecosystems and elucidating their role in increasingly important environmental issues is a constantly present research question. The concealed plankton community dynamics reflect changes in environmental forcing, growth traits of competing species, and multiple food web interactions. Recent technological advances have led to the possibility of collecting real-time big data opening new horizons for testing core hypotheses in planktonic systems, derived from macroscopic realms, in community ecology, biodiversity research, and ecosystem functioning. Analyzing the big data calls for computer vision and machine learning methods capable of producing interoperable data across platforms and systems. In this paper we apply convolutional neural networks (CNN) to classify a brackish-water phytoplankton community in the Baltic Sea. For solving the classification task, we utilize compact CNN architectures requiring less computational capacity and creating an opportunity to quickly train the network. This makes it possible to (1) test various modifications to the classification method, and (2) repeat each experiment multiple times with different training and test set combinations to obtain reliable results. We further analyze the effect of large class imbalance to the CNN performance, and test relevant data augmentation techniques to improve the performance. Finally, we address the practical implications of the classification performance to aquatic research by analyzing the confused classes and their effect on the reliability of the automatic plankton recognition system, to guide further development of plankton recognition research. Our results show that it is possible to obtain good classification accuracy with relatively shallow architectures and a small amount of training data when using effective data augmentation methods even with a very unbalanced dataset.

Tuomas Eerola et al.

Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement
Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Tuomas Eerola et al.

Tuomas Eerola et al.


Total article views: 989 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
609 325 55 989 61 45
  • HTML: 609
  • PDF: 325
  • XML: 55
  • Total: 989
  • BibTeX: 61
  • EndNote: 45
Views and downloads (calculated since 08 Jul 2020)
Cumulative views and downloads (calculated since 08 Jul 2020)

Viewed (geographical distribution)

Total article views: 890 (including HTML, PDF, and XML) Thereof 889 with geography defined and 1 with unknown origin.
Country # Views %
  • 1


Latest update: 03 Dec 2023
Short summary
The role of plankton communities in important environmental issues is an active research question. Large amounts of plankton images collected using modern devices call for automated analysis methods. We consider classification of phytoplanktons using compact convolutional neural networks allowing fast model training. We analyse the confused classes and their practical implications to aquatic research. We show that good accuracy can be obtained with a limited amount of unbalanced training data.