GRI Publications and Scholarship

Aerial Wildlife Image Repository

Abstract

The availability of an ever-improving repository of datasets allows machine learning algorithms to have a robust training set of images, which in turn allows for accurate detection and classification of wildlife. This repository (AWIR---Aerial Wildlife Image Repository) would be a step in creating a collaborative rich dataset both in terms of taxa of animals and in terms of the sensors used to observe (visible, infrared, Lidar etc.). Initially, priority would be given to wildlife species hazardous to aircrafts, and common wildlife damage-associated species. AWIR dataset is accompanied by a classification benchmarking website showcasing examples of state-of-the-art algorithms recognizing the wildlife in the images.

DOI

https://doi.org/10.54718/WVGF3020

Publication Date

2023

Research Center

Geosystems Research Institute

AWIR-Cow.zip (70707 kB)
Cow

AWIR-Deer.zip (59738 kB)
Deer

AWIR-Horse.zip (55623 kB)
Horse

This document is currently not available here.

Share

COinS