Skip to main content

Animal population censusing at scale with citizen science and photographic identification

Author(s): Parham, J; Crall, J; Stewart, C; Berger-Wolf, T; Rubenstein, Daniel I.

Download
To refer to this page use: http://arks.princeton.edu/ark:/88435/pr1s791
Abstract: Population censusing is critical to monitoring the health of an animal population. A census results in a population size estimate, which is a fundamental metric for deciding the demographic and conservation status of a species. Current methods for producing a population census are expensive, demanding, and may be invasive, leading to the use of overly-small sample sizes. In response, we propose to use volunteer citizen scientists to collect large numbers of photographs taken over large geographic areas, and to use computer vision algorithms to semi-automatically identify and count individual animals. Our data collection and processing are distributed, non-invasive, and require no specialized hardware and no scientific training. Our method also engages the community directly in conservation. We analyze the results of two population censusing events, the Great Zebra and Giraffe Count (2015) and the Great Grevy's Rally (2016), where combined we processed over 50,000 photographs taken with more than 200 different cameras and over 300 on-the-ground volunteers. © Copyright 2017, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
Publication Date: 1-Jan-2017
Citation: Parham, J, Crall, J, Stewart, C, Berger-Wolf, T, Rubenstein, D. (2017). Animal population censusing at scale with citizen science and photographic identification. AAAI Spring Symposium - Technical Report, SS-17-01 - SS-17-08 (37 - 44).
Pages: 37 - 44
Type of Material: Journal Article
Journal/Proceeding Title: AAAI Spring Symposium - Technical Report
Version: Final published version. Article is made available in OAR by the publisher's permission or policy.



Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.