Computer Vision for Ecology
The cost, time, and logistics of human observation limit many studies, yet ecologists have only begun to use automated tools. Computer vision is a field of computer science that extracts information from images and combines data taken from the field with automated image analysis.
Ecologists are increasingly relying on diverse datasets from air-borne photographs to deep sea videos, but our ability to manage and analyze these datasets is limited. This session will bring together researchers to think broadly about the major opportunities, challenges and connections between ecologists using image data. I am actively gathering researchers to create a small network of projects focusing on automated biodiversity analysis. Stay tuned for new meetings at ESA 2020. |
Past Meetings
Ecological Society for America. Annual Meeting in Baltimore, MD.
Ignite ESA Session #10813: "Image Processing and Computer Vision for Ecology"
Tuesday, August 11, 2015
1:30 PM - 3:00 PM
Moderator: Ben Weinstein, Stony Brook University.
Presenters:
Applications:
Hotspotter, a fast, accurate algorithm for identifying individual animals against a labeled database.
Nature pattern match, a computer vision tool designed for extracting and matching visual features.
Leaf snap, a series of electronic field guides for flowers and trees
stalkless, Quickly (trying to) analyse leaf morphology.
idtracker: is a videotracking identifies individual during the whole video.
Other Active Researchers:
While putting together this session, i've talked to many great researchers who could not attend.
Tanya Berger-Wolf - Image Matching
Roland Kays - Camera trapping
Margaret Kosmala - Citizen Science
Aaron Olsen - Morphometrics
Mary Caswell Stoddard - Visual Mimicry
Tony Dell - Automated Tracking
Iain Couzin - Social Behavior
Jeff Kerby - Landscape Mapping
The coral reef computer vision group - UCSD
Ecological Society for America. Annual Meeting in Baltimore, MD.
Ignite ESA Session #10813: "Image Processing and Computer Vision for Ecology"
Tuesday, August 11, 2015
1:30 PM - 3:00 PM
Moderator: Ben Weinstein, Stony Brook University.
Presenters:
- Tavis Forrester, Smithsonian Conservation Biology Institute. “Image analysis for camera traps”
- Phil McDowell, Stony Brook University. “Ultra high-res elevation mapping using computer vision”
- Chandi Witharana, Stony Brook University. “High resolution remote sensing of penguin colonies”
- Ali Swanson, Zooniverse. "Engaging the public through image data"
- Katie Christie, University of Alaska. “Unmanned aerial vehicles in ecology and wildlife science"
- Paul Conn, NOAA. "Statistical models for automated datasets"
Applications:
Hotspotter, a fast, accurate algorithm for identifying individual animals against a labeled database.
Nature pattern match, a computer vision tool designed for extracting and matching visual features.
Leaf snap, a series of electronic field guides for flowers and trees
stalkless, Quickly (trying to) analyse leaf morphology.
idtracker: is a videotracking identifies individual during the whole video.
Other Active Researchers:
While putting together this session, i've talked to many great researchers who could not attend.
Tanya Berger-Wolf - Image Matching
Roland Kays - Camera trapping
Margaret Kosmala - Citizen Science
Aaron Olsen - Morphometrics
Mary Caswell Stoddard - Visual Mimicry
Tony Dell - Automated Tracking
Iain Couzin - Social Behavior
Jeff Kerby - Landscape Mapping
The coral reef computer vision group - UCSD