Beyond transfer learning: Leveraging ancillary images in automated classification of plankton
Open Access
- 25 September 2024
- journal article
- research article
- Published by Wiley in Limnology and Oceanography: Methods
- Vol. 22 (12) , 943-952
Abstract
We assess whether a supervised machine learning algorithm, specifically a convolutional neural network (CNN), achieves higher accuracy on planktonic image classification when including non‐plankton and ancillary plankton during the training procedure. We focus on the case of optimizing the CNN for a single planktonic image source, while considering ancillary images to be plankton images from other instruments. We conducted two sets of experiments with three different types of plankton images (from a Zooglider, Underwater Vision Profiler 5, and Zooscan), and our results held across all three image types. First, we considered whether single‐stage transfer learning using non‐plankton images was beneficial. For this assessment, we used ImageNet images and the 2015 ImageNet contest‐winning model, ResNet‐152. We found increased accuracy using a ResNet‐152 model pretrained on ImageNet, provided the entire network was retrained rather than retraining only the fully connected layers. Next, we combined all three plankton image types into a single dataset with 3.3 million images (despite their differences in contrast, resolution, and pixel pitch) and conducted a multistage transfer learning assessment. We executed a transfer learning stage from ImageNet to the merged ancillary plankton dataset, then a second transfer learning stage from that merged plankton model to a single instrument dataset. We found that multistage transfer learning resulted in additional accuracy gains. These results should have generality for other image classification tasks.Keywords
Funding Information
- Division of Ocean Sciences (OCE‐2224726, OCE‐2243190)
This publication has 19 references indexed in Scilit:
- Deep learning for plankton and coral classificationApplied Computing and Informatics, 2020
- Improving plankton image classification using context metadataLimnology and Oceanography: Methods, 2019
- Automated species-level identification of planktic foraminifera using convolutional neural networks, with comparison to human performanceMarine Micropaleontology, 2019
- Zooglider: An autonomous vehicle for optical and acoustic sensing of zooplanktonLimnology and Oceanography: Methods, 2018
- Validation methods for plankton image classification systemsLimnology and Oceanography: Methods, 2016
- Deep Residual Learning for Image Recognition2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Digital zooplankton image analysis using the ZooScan integrated systemJournal of Plankton Research, 2010
- ImageNet: A large-scale hierarchical image database2009