University of Cambridge > Talks.cam > AI4ER Seminar Series > Using Deep Learning to Count Albatrosses from Space

  Browse A-Z

Log in

Googledetails Microsoftdetails

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹī¸ Information

Using Deep Learning to Count Albatrosses from Space

Download to your calendar using vCal

  • UserEllie Bowler, University of East Anglia, British Antarctic Survey
  • ClockTuesday 25 February 2020, 12:00-13:15
  • HouseBullard Lab, Seminar Room.

If you have a question about this talk, please contact Jonathan Rosser .

Chair: Emily Shuckburgh Abstract: In this project we aim to automate the detection of Wandering Albatrosses in super high resolution satellite imagery (DigitalGlobe’s WorldView-3), using state of the art deep learning approaches. We train a convolutional neural network to classify and detect potential albatrosses, achieving accuracy values of approximately 80% on the test set. By analysing the agreement between manually generated labels, we show that these results are in fact in line with human performance. We hope that the methods will streamline the analysis of WorldView-3 imagery, allowing more frequent monitoring of a species which is of high conservation concern.

This talk is part of the AI4ER Seminar Series series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Harker Seminar Room, Department of Earth Sciences Harrods Rm, Queens Building, Emmanuel College, Cambridge CB2 3AP Centre for Scientific Computing

Other talks

Talk title TBC TBC CUSCRS and Renal Medicine Title tbc Washing machines and their rare components: lessons from kidney patients Ensemble inequivalence for systems with long-range interactions
 

Š 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity