Use this zoom link to join

“Using computer vision to study animal behavior in natural environments”

Abstract: High resolution imaging paired with automated software processing is transforming how we study animals in natural habitats.  By recording a snapshot of the landscape and the individuals within it, we can quantify the presence and behavior of multiple individuals simultaneously. The result is a record of behavior in the context of animals’ natural physical as well as social landscapes. In terms of management, these techniques can enable consistent population monitoring across space and time and provide data about things like fine scale habitat use. Beyond pure management concerns, these methods also allow researchers to investigate topics such as multi-animal interactions like predator-prey dynamics, collective navigation, or decision making in social contexts. As a postdoc working with Professor Andrew Berdahl, I am using these techniques in the context of Pacific salmon migration and reproduction. In this presentation, I will discuss the ways in which automated image processing has enabled my previous work, such as monitoring the world’s largest fruit bat colony, using drones to study terrestrial animal groups at sub-second, sub-meter resolution, and quantifying honey bee nest use and construction. I will then discuss how I am using these techniques to study Pacific salmon in my current work, specifically topics like quantifying nest location and timing, finescale bear-salmon hunting dynamics, and navigation in a social context.

Biography: Ben Koger is a Washington Research Foundation Postdoctoral Fellow working with Professor Andrew Berdahl at SAFS focused on designing and using automated image based tools for studying animal behavior in natural landscapes. He did my Ph.D. with Iain Couzin at the Max Planck Institute of Animal Behavior in Konstanz, Germany. He is interested in studying the social context in which animals move and make decisions in natural environments. In addition to designing novel image processing pipelines, this work has spanned systems ranging from ungulate herds in Kenya to one of the world’s largest bat migrations in Zambia. Ben is currently counting and recording Pacific salmon as they make decisions about how to navigate home and reproduce while interacting with conspecifics and being hunted by grizzly bears.

The UW Data Science Seminar is an annual lecture series at the University of Washington that hosts scholars working across applied areas of data science, such as the sciences, engineering, humanities and arts along with methodological areas in data science, such as computer science, applied math and statistics. Our presenters come from all domain fields and include occasional external speakers from regional partners, governmental agencies and industry.

The 2022-2023 seminars will be virtual, and are free and open to the public.