Beyond transfer learning: Leveraging ancillary images in automated classification of plankton

被引:0
|
作者
Ellen, Jeffrey S. [1 ,2 ]
Ohman, Mark D. [2 ]
机构
[1] NIWC Pacific, Basic & Appl Res Div, San Diego, CA 92152 USA
[2] Univ Calif San Diego, Scripps Inst Oceanog, Calif Current Ecosyst LTER Site, La Jolla, CA 92037 USA
来源
LIMNOLOGY AND OCEANOGRAPHY-METHODS | 2024年 / 22卷 / 12期
基金
美国国家科学基金会;
关键词
D O I
10.1002/lom3.10648
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
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.
引用
收藏
页码:943 / 952
页数:10
相关论文
共 50 条
  • [31] Optimization of automated tissue classification in histopathological images: use of a deep transfer learning approach on a pancreatic cancer cohort
    Haeberle, L.
    Kronberg, R.
    Pfaus, M.
    Xu, H.
    Krings, K.
    Schlensog, M.
    Rau, T.
    Pandyra, A.
    Lang, K.
    Esposito, I.
    Lang, P.
    VIRCHOWS ARCHIV, 2022, 481 (SUPPL 1) : S79 - S79
  • [32] Deep learning for plankton and coral classification
    Lumini, Alessandra
    Nanni, Loris
    Maguolo, Gianluca
    APPLIED COMPUTING AND INFORMATICS, 2023, 19 (3-4) : 265 - 283
  • [33] SoftLungX: leveraging transfer learning with convolutional neural networks for accurate respiratory disease classification in chest X-ray images
    Geroski, Tijana
    Pavic, Ognjen
    Dasic, Lazar
    Milovanovic, Dragan
    Petrovic, Marina
    Filipovic, Nenad
    JOURNAL OF BIG DATA, 2024, 11 (01)
  • [34] Machine Learning on Biomedical Images: Interactive Learning, Transfer Learning, Class Imbalance, and Beyond
    Khan, Naimul Mefraz
    Abraham, Nabila
    Hon, Marcia
    Guan, Ling
    2019 2ND IEEE CONFERENCE ON MULTIMEDIA INFORMATION PROCESSING AND RETRIEVAL (MIPR 2019), 2019, : 85 - 90
  • [35] An automated deep learning pipeline based on advanced optimisations for leveraging spectral classification modelling
    Passos, Dario
    Mishra, Puneet
    CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2021, 215
  • [36] Leveraging Gradient-Based Optimizer and Deep Learning for Automated Soil Classification Model
    Alsolai, Hadeel
    Rizwanullah, Mohammed
    Maashi, Mashael
    Othman, Mahmoud
    Alneil, Amani A.
    Abdelmageed, Amgad Atta
    CMC-COMPUTERS MATERIALS & CONTINUA, 2023, 76 (01): : 975 - 992
  • [37] Automated Plankton Classification With a Dynamic Optimization and Adaptation Cycle
    Conradt, Jan
    Boerner, Gregor
    Lopez-Urrutia, Angel
    Moellmann, Christian
    Moyano, Marta
    FRONTIERS IN MARINE SCIENCE, 2022, 9
  • [38] In-domain versus out-of-domain transfer learning in plankton image classification
    Maracani, Andrea
    Pastore, Vito Paolo
    Natale, Lorenzo
    Rosasco, Lorenzo
    Odone, Francesca
    SCIENTIFIC REPORTS, 2023, 13 (01)
  • [39] In-domain versus out-of-domain transfer learning in plankton image classification
    Andrea Maracani
    Vito Paolo Pastore
    Lorenzo Natale
    Lorenzo Rosasco
    Francesca Odone
    Scientific Reports, 13
  • [40] NollySenti: Leveraging Transfer Learning and Machine Translation for Nigerian Movie Sentiment Classification
    Shode, Iyanuoluwa
    Adelani, David Ifeoluwa
    Peng, Jing
    Feldman, Anna
    61ST CONFERENCE OF THE THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL 2023, VOL 2, 2023, : 986 - +