Degree of Contribution

Lead

Document Type

Article

Keywords

Computer Vision, Animal Tracking

Disciplines

Artificial Intelligence and Robotics | Behavior and Ethology | Robotics

Abstract

The Kihansi spray toad (Nectophrynoides asperginis), classified as Extinct in the Wild by the IUCN, is being bred at the Wildlife Conservation Society’s (WCS) Bronx Zoo as part of an effort to successfully reintroduce the species into the wild. Thousands of toads live at the Bronx Zoo presenting an opportunity to learn more about their behaviors for the first time, at scale. It is impractical to perform manual observations for long periods of time. This paper reports on the development of a RGB-D tracking and analytics approach that allows researchers to accurately and efficiently gather information about the toads’ behavior. A method of automated tracking based on raw movement information acquired via depth stream provided by Intel’s SR300 RGB-D camera is developed. Depth and color video streams are used to effectively track movement of small targets. Depth information identifies regions of interest from any motion detected, whereas color correlation is used to identify targets from detected motion. Challenges to be overcome include dealing with small and variable shapes of targets, their varying motion speeds, and the sheer quantity of target information. The resulting tracking system outputs sets of 3D movement tracks with timestamps. These movement tracks are then processed to display the toad target track graphs in 2D and 3D graphs, 2D heatmaps showing areas of localized activity, and multi-track analytics to detect gross activity, potential fight activity, and potential mating activity graphs. Results from automated tracking of toads under various lighting conditions over multiple days are reported and discussed.

Publication Title

Herpetological Review

Volume

51

Issue

1

Article Number

1079

Publication Date

Spring 3-2020

First Page

37

Last Page

46

Extent

9

Publisher

Society for the Study of Amphibians and Reptiles

Language

United States

Peer Reviewed

1

Version

Published

Subjects

Computer Vision, Animal Tracking

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

COinS