Biodiversity Information Science and Standards : Conference Abstract
|
Corresponding author: Sohaib Younis (muhammad-sohaib.younis@senckenberg.de)
Received: 05 Apr 2019 | Published: 10 Jul 2019
© 2019 Sohaib Younis, Marco Schmidt, Bernhard Seeger, Thomas Hickler, Claus Weiland
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation: Younis S, Schmidt M, Seeger B, Hickler T, Weiland C (2019) A Workflow for Data Extraction from Digitized Herbarium Specimens. Biodiversity Information Science and Standards 3: e35190. https://doi.org/10.3897/biss.3.35190
|
|
Based on own work on species and trait recognition and complementary studies from other working groups, we present a workflow for data extraction from digitized herbarium specimens using convolutional neural networks. Digitized herbarium sheets contain:
In order to treat these objects appropriately, segmentation techniques (
In order to deal with new incoming digitized collections, unseen data or categories, we propose implementation of a new Deep Learning approach, so-called Lifelong Learning: Past knowledge of the network is dynamically saved in latent space using autoencoder and generatively replayed while the network is trained on new tasks which enables it to solve complex image processing tasks without forgetting former knowledge while incrementally learning new classes and knowledge.
Trait Recognition, Convolutional Neural Networks, Lifelong Learning, Herbarium specimens, Trait Semantics, Digitized Natural History Collections, Image Processing, Image Captioning, Object Detection, Object Annotation
Sohaib Younis