Automated plankton classification from holographic imagery with deep convolutional neural networks

Buyu Guo, Lisa Nyman, Aditya R. Nayak, David Milmore, Malcolm McFarland, Michael S. Twardowski, James M. Sullivan, Jia Yu, Jiarong Hong

Research output: Contribution to journalArticlepeer-review

Abstract

In situ digital inline holography is a technique which can be used to acquire high-resolution imagery of plankton and examine their spatial and temporal distributions within the water column in a nonintrusive manner. However, for effective expert identification of an organism from digital holographic imagery, it is necessary to apply a computationally expensive numerical reconstruction algorithm. This lengthy process inhibits real-time monitoring of plankton distributions. Deep learning methods, such as convolutional neural networks, applied to interference patterns of different organisms from minimally processed holograms can eliminate the need for reconstruction and accomplish real-time computation. In this article, we integrate deep learning methods with digital inline holography to create a rapid and accurate plankton classification network for 10 classes of organisms that are commonly seen in our data sets. We describe the procedure from preprocessing to classification. Our network achieves 93.8% accuracy when applied to a manually classified testing data set. Upon further application of a probability filter to eliminate false classification, the average precision and recall are 96.8% and 95.0%, respectively. Furthermore, the network was applied to 7500 in situ holograms collected at East Sound in Washington during a vertical profile to characterize depth distribution of the local diatoms. The results are in agreement with simultaneously recorded independent chlorophyll concentration depth profiles. This lightweight network exemplifies its capability for real-time, high-accuracy plankton classification and it has the potential to be deployed on imaging instruments for long-term in situ plankton monitoring.

Original languageEnglish (US)
Pages (from-to)21-36
Number of pages16
JournalLimnology and Oceanography: Methods
Volume19
Issue number1
DOIs
StatePublished - Jan 2021

Bibliographical note

Funding Information:
The field data collection efforts in East Sound were funded under grant N00014‐15‐1‐2628 from the Office of Naval Research Coastal Geophysics and Optics program (MST and JMS). Development of the HOLOCAM was supported by a National Oceanographic Partnership Program (NOPP) grant administered by ONR (JMS and MST, grant N0001410C0041). The authors acknowledge Brad Penta and the Naval Research Laboratory (NRL) for providing logistical support and ship time for acquiring the Delaware shelf data set, as part of the NRL‐funded Intermediate Trophic Levels Project (grant 61153N). The data processing efforts at FAU were partially funded by NSF grants OCE‐1634053 and OCE‐1657332 and internal support. Portions of the analysis were supported by the Harbor Branch Oceanographic Institute Foundation. BG is supported by the Chinese Scholarship Council. Embedded computer testing is supported by National Natural Science Foundation of China (61705210 and 41527901). The authors thank Ranjoy Barua for processing the holographic imagery and for helping with the creation and management of the labeled plankton database, and Kevin Mallery for his helpful advice on network architecture. The authors are also grateful for the efforts of Sean Goughan and Paige Trenchard toward helping create the labeled image database.

Funding Information:
The field data collection efforts in East Sound were funded under grant N00014-15-1-2628 from the Office of Naval Research Coastal Geophysics and Optics program (MST and JMS). Development of the HOLOCAM was supported by a National Oceanographic Partnership Program (NOPP) grant administered by ONR (JMS and MST, grant N0001410C0041). The authors acknowledge Brad Penta and the Naval Research Laboratory (NRL) for providing logistical support and ship time for acquiring the Delaware shelf data set, as part of the NRL-funded Intermediate Trophic Levels Project (grant 61153N). The data processing efforts at FAU were partially funded by NSF grants OCE-1634053 and OCE-1657332 and internal support. Portions of the analysis were supported by the Harbor Branch Oceanographic Institute Foundation. BG is supported by the Chinese Scholarship Council. Embedded computer testing is supported by National Natural Science Foundation of China (61705210 and 41527901). The authors thank Ranjoy Barua for processing the holographic imagery and for helping with the creation and management of the labeled plankton database, and Kevin Mallery for his helpful advice on network architecture. The authors are also grateful for the efforts of Sean Goughan and Paige Trenchard toward helping create the labeled image database.

Publisher Copyright:
© 2020 Association for the Sciences of Limnology and Oceanography

Fingerprint

Dive into the research topics of 'Automated plankton classification from holographic imagery with deep convolutional neural networks'. Together they form a unique fingerprint.

Cite this