Deep learning and t-SNE projection for plankton images clusterization

A. Goulart, Alexandre Morimitsu, Renan Jacomassi, N. Hirata, Rubens M. Lopes

2021 Oceans Cited 1 times

Abstract

In this paper we present a pipeline to cluster unlabelled image samples. Although not restricted to plankton image applications, we present the system within this context. Feature maps obtained from a deep learning architecture (DenseNet) are fed to the t-SNE projection in order to obtain 2D clusters. The method successfully creates clusters that can be used in interactive software, for quick manual classification of images batches.

BibTeX

@inproceedings{Goulart2021DeepLearning,
  author = {Goulart, Antonio José Homsi and Morimitsu, Alexandre and Jacomassi, Renan and Hirata, Nina and Lopes, Rubens},
  booktitle = {OCEANS 2021: San Diego – Porto},
  title = {Deep learning and t-SNE projection for plankton images clusterization},
  year = {2021},
  volume = {},
  number = {},
  pages = {1-4},
  abstract = {In this paper we present a pipeline to cluster unlabelled image samples. Although not restricted to plankton image applications, we present the system within this context. Feature maps obtained from a deep learning architecture (DenseNet) are fed to the t-SNE projection in order to obtain 2D clusters. The method successfully creates clusters that can be used in interactive software, for quick manual classification of images batches.},
  keywords = {Deep learning;Pipelines;Supervised learning;Software algorithms;Computer architecture;Feature extraction;Software;clusterization;dataset labelling;convolutional neural network;plankton imaging},
  doi = {10.23919/OCEANS44145.2021.9706043},
  issn = {0197-7385},
  month = {Sep.},
}