Svenska Systematikföreningen: Automatic identification of organism
Convener: Miroslav Valan (Miroslav.Valan@nrm.se).
The dream of having computers identify biological species simply by analyzing specimen images has been around for decades, but actual progress in this direction has been relatively slow. This is bound to change, as the field of computer vision has developed dramatically in recent years thanks to the introduction of deep learning algorithms based on convolutional neural networks (ConvNets), and the increasing availability of powerful graphical processing units (GPUs). This session will focus on the potential of these cutting-edge techniques in moving the dream of automated species identification closer to reality. The aim is to bring together biodiversity researchers, experts on state-of-the-art automated taxon identification systems, and computer vision researchers working on similar projects in a stimulating discussion of the potential of the new techniques and the barriers to further progress.
Friday, August 18th. Afternoon session.
13.30 – 14.20
The Automated Assessment and Identification of Organisms from Mor-phological Data.
1The Natural History Museum, Cromwell Road, London, UK; 2Department
of Earth Science, University College, London, UK; 2Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, China.
14.20 – 14.40
Allison Y. Hsiang1, Pincelli M. Hull, Kaylea Nelson, Leanne E. Elder, Luke C. Strotz, Gregory D. Meyer:
Best practices for developing imaging datasets for object recognition using supervised machine learning.
1Naturhistoriska Riksmuseet, Stockholm.
14.40 – 15.00
James Warren, and Brenton Lang:
Plantkey - An apparatus and method for the instantaneous
identification of plant species from leaf samples taken whilst in
15.00 – 15.30
15.30 – 15.50
Can Convolutional Neural Networks be used in taxonomy?
1Department of Bioinformatics and Genetics, Swedish Museum of Natural History; 2Savantic AB.
15.50 – 16.10
Ethics of data sharing in the age of digital imaging?
16.10 – 17.00