Symposium on image-based techniques in species detection and identification

We bring together experts at Aarhus University and beyond to raise awareness of the potential of machine learning in image analyses for detection of biological organisms in order to develop synergy between groups and disciplines.

02.11.2018 | Alexandros Iosifidis

Dato tor 08 nov
Tid 09:30 16:00
Sted Aarhus Institute of Advanced Studies, Auditorium (room 1630-201), Høegh- Guldbergs Gade 6B, 8000 Aarhus C

Image data is increasingly being used to record organisms with greater temporal and spatial resolution than would be possible by direct observation. While it has long been recognized that images represent a useful way to record information for a wide range of purposes, the application of tools to automatically derive relevant information from such images are not widely used. The rapid developments in computer science and software engineering is holding the promise of a transformation of the use of images to store and automatically extract information across a wide range of research disciplines. The aim of this symposium is to bring together experts at Aarhus University and beyond to raise awareness of the potential of machine learning in image analyses for detection of biological organisms in order to develop synergy between groups and disciplines.


PROGRAMME

9:30-10.00 Coffee and rolls

10.00-12.10

  • Welcome (Toke Thomas Høye)
  • Computer vision and machine learning for object detection
    Alexandros Iosifidis, Department of Engineering, Aarhus University
  • Advanced computational and statistical techniques for biomonitoring and aquatic ecosystem service management
    Jenni Raitoharju and Johanna Ärje, Technical University Tampere, Finland
  • Automation of sorting and identification of arthropod samples using robots
    Thomas Lyhne and Kristoffer Fink Løwenstein, AU Engineering, Aarhus University
  • Using beetle specimens of known identity to quantify sources of uncertainty in image-based identification
    Oskar LP Hansen, Department of Bioscience, Aarhus University
  • Studying plant-pollinator interactions one flower at a time
    Toke Thomas Høye, Department of Bioscience, Aarhus University
  • Arctic insect detection
    Johanna Ärje, Technical University Tampere, Finland

12.10-13.00 Lunch Aarhus University

13.00-14.00

  • Detection of seals from aerial thermal and RGB imagery
    Eric Jürgen Haase, Department of Bioscience, Aarhus University
  • Detection of birds from orthophotos - off shore transect flights
    Ib Kragh Petersen, Department of Bioscience, Aarhus University
  • Detection of demographic parameters from time lapse images – cliff nesting birds
    Flemming Merkel, Department of Bioscience, Aarhus University

14.00-14.30 Coffee

14.30-16.00

  • Honey bee monitoring
    Yoko Dupont, Department of Bioscience, Aarhus University
  • A computer vision system to monitor the infestation level of Varroa destructor in a honey bee colony
    Kim Bjerge, AU Engineering, Aarhus University
  • Application of vision based machine learning algorithms within the agricultural domain and the challenges of knowledge transfer to commercial partners
    Mads Dyrmann and Søren Skovsen, Department of Engineering, Aarhus University
  • Discussion

Participation is free of charge, but please sign up by Monday 5 November to help plan catering by sending an e-mail to Toke Thomas Høye (tth@bios.au.dk).

Seminar