Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2

Ecology and Evolution

Ryan S. Miller¹  Jesse Lewis²  Eric S. Newkirk³  Jacob S. Ivan³  Reesa Y. Conrey³  Eric A. Odell³  Mohammad S. Norouzzadeh⁴  Fabiola Iannarilli⁵  Ryan K. Brook⁶  John Erb⁷  David W. Wolfson⁸  Amy J. Davis⁹  Kurt C. VerCauteren^1,10  Jeff Clune^1,11  Michael A. Tabak^1,12  Raoul K. Boughton^1,13  James C. Beasley^1,14  Daniel P. Walsh^1,15  Erica J. Newton^1,16  Jennifer Stenglein^1,17 

[1] Center for Epidemiology and Animal Health United States Department of Agriculture Fort Collins CO USA;College of Integrative Sciences and Arts Arizona State University Mesa AZ USA;Colorado Parks and Wildlife Fort Collins CO USA;Computer Science Department University of Wyoming Laramie WY USA;Conservation Sciences Graduate Program University of Minnesota St. Paul MN USA;Department of Animal and Poultry Science University of Saskatchewan Saskatoon SK Canada;Forest Wildlife Populations and Research Group Minnesota Department of Natural Resources Grand Rapids MN USA;Minnesota Cooperative Fish and Wildlife Research Unit Department of Fisheries, Wildlife and Conservation Biology University of Minnesota St. Paul MN USA;National Wildlife Research Center United States Department of Agriculture Fort Collins CO USA;National Wildlife Research Center United States Department of Agriculture, Animal and Plant Health Inspection Service Fort Collins CO USA;OpenAI San Francisco CA USA;Quantitative Science Consulting, LLC Laramie WY USA;Range Cattle Research and Education Center, Wildlife Ecology and Conservation University of Florida Ona FL USA;Savannah River Ecology Laboratory Warnell School of Forestry and Natural Resources University of Georgia Aiken SC USA;US Geological SurveyNational Wildlife Health Center Madison WI USA;Wildlife Research and Monitoring Section Ontario Ministry of Natural Resources and Forestry Peterborough ON Canada;Wisconsin Department of Natural Resources Madison WI USA;

关键词: computer vision;    deep convolutional neural networks;    image classification;    machine learning;    motion‐activated camera;    R package;   

DOI  :  10.1002/ece3.6692

来源: DOAJ

【 摘 要 】

Abstract Motion‐activated wildlife cameras (or “camera traps”) are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter‐out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the “species model,” and one that determines if an image is empty or if it contains an animal, the “empty‐animal model.” Our species model and empty‐animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out‐of‐sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36%–91% across all out‐of‐sample datasets) and the empty‐animal model achieved an accuracy of 91%–94% on out‐of‐sample datasets from different continents. Our software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty‐animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths.

【 授权许可】

Unknown