Machine learning to classify animal species in camera trap images: Applications in ecology
- US Dept. of Agriculture (USDA), Fort Collins, CO (United States). Center for Epidemiology and Animal Health; University of Wyoming, Laramie, WY (United States); DOE/OSTI
- University of Wyoming, Laramie, WY (United States)
- US Dept. of Agriculture (USDA), Fort Collins, CO (United States). Center for Epidemiology and Animal Health
- US Dept. of Agriculture (USDA), Fort Collins, CO (United States). National Wildlife Research Center
- US Dept. of Agriculture (USDA), Fort Collins, CO (United States). Center for Epidemiology and Animal Health\
- Arizona State University, Mesa, AZ (United States)
- Tejon Ranch Conservancy, Lebec, CA (United States)
- University of Georgia, Aiken, SC (United States). Savannah River Ecology Laboratory
- University of Florida, Ona FL (United States)
- Colorado Parks and Wildlife, Fort Collins, Co (United States)
- University of Saskatchewan, Saskatoon, SK (Canada)
- University of Montana, Missoula, MT (United States)
Motion-activated cameras (“camera traps”) are increasingly used in ecological and management studies for remotely observing wildlife and are amongst the most powerful tools for wildlife research. However, studies involving camera traps result in millions of images that need to be analysed, typically by visually observing each image, in order to extract data that can be used in ecological analyses. We trained machine learning models using convolutional neural networks with the ResNet-18 architecture and 3,367,383 images to automatically classify wildlife species from camera trap images obtained from five states across the United States. We tested our model on an independent subset of images not seen during training from the United States and on an out-of-sample (or “out-of-distribution” in the machine learning literature) dataset of ungulate images from Canada. We also tested the ability of our model to distinguish empty images from those with animals in another out-of-sample dataset from Tanzania, containing a faunal community that was novel to the model. The trained model classified approximately 2,000 images per minute on a laptop computer with 16 gigabytes of RAM. The trained model achieved 98% accuracy at identifying species in the United States, the highest accuracy of such a model to date. Out-of-sample validation from Canada achieved 82% accuracy and correctly identified 94% of images containing an animal in the dataset from Tanzania. We provide an r package (Machine Learning for Wildlife Image Classification) that allows the users to (a) use the trained model presented here and (b) train their own model using classified images of wildlife from their studies. The use of machine learning to rapidly and accurately classify wildlife in camera trap images can facilitate non-invasive sampling designs in ecological studies by reducing the burden of manually analysing images. Our r package makes these methods accessible to ecologists.
- Research Organization:
- University of Georgia, Athens, GA (United States)
- Sponsoring Organization:
- US Department of Agriculture (USDA); USDOE; USDOE Office of Environmental Management (EM)
- Grant/Contract Number:
- EM0004391
- OSTI ID:
- 1614652
- Journal Information:
- Methods in Ecology and Evolution (Online), Journal Name: Methods in Ecology and Evolution (Online) Journal Issue: 4 Vol. 10; ISSN 2041-210X
- Publisher:
- British Ecological SocietyCopyright Statement
- Country of Publication:
- United States
- Language:
- English
A comparison of cost and quality of three methods for estimating density for wild pig (Sus scrofa)
|
journal | February 2020 |
| Data from: Machine learning to classify animal species in camera trap images: applications in ecology | dataset | January 2019 |
| Data from: Machine learning to classify animal species in camera trap images: applications in ecology | dataset | January 2019 |
Similar Records
Seismic savanna: machine learning for classifying wildlife and behaviours using ground‐based vibration field recordings
Two-Stage Wildlife Event Classification for Edge Deployment