3 years ago

Ant genera identification using an ensemble of convolutional neural networks

Ellen Marianne B. Cavalheiro, Christiano Lyra, Fernando J. Von Zuben, Marcos M. Raimundo, Luis F. P. Salles, Alan Caio R. Marques
Works requiring taxonomic knowledge face several challenges, such as arduous identification of many taxa and an insufficient number of taxonomists to identify a great deal of collected organisms. Machine learning tools, particularly convolutional neural networks (CNNs), are then welcome to automatically generate high-performance classifiers from available data. Supported by the image datasets available at the largest online database on ant biology, the AntWeb (www.antweb.org), we propose here an ensemble of CNNs to identify ant genera directly from the head, profile and dorsal perspectives of ant images. Transfer learning is also considered to improve the individual performance of the CNN classifiers. The performance achieved by the classifiers is diverse enough to promote a reduction in the overall classification error when they are combined in an ensemble, achieving an accuracy rate of over 80% on top-1 classification and an accuracy of over 90% on top-3 classification.

Publisher URL: http://journals.plos.org/plosone/article

DOI: 10.1371/journal.pone.0192011

You might also like
Discover & Discuss Important Research

Keeping up-to-date with research can feel impossible, with papers being published faster than you'll ever be able to read them. That's where Researcher comes in: we're simplifying discovery and making important discussions happen. With over 19,000 sources, including peer-reviewed journals, preprints, blogs, universities, podcasts and Live events across 10 research areas, you'll never miss what's important to you. It's like social media, but better. Oh, and we should mention - it's free.

  • Download from Google Play
  • Download from App Store
  • Download from AppInChina

Researcher displays publicly available abstracts and doesn’t host any full article content. If the content is open access, we will direct clicks from the abstracts to the publisher website and display the PDF copy on our platform. Clicks to view the full text will be directed to the publisher website, where only users with subscriptions or access through their institution are able to view the full article.