News Release

Ecology and AI

Harnessing artificial intelligence to automatically identify, count, and describe animals in the wild

Peer-Reviewed Publication

Harvard University

It's poised to transform fields from earthquake prediction to cancer detection to self-driving cars, and now scientists are unleashing the power of deep learning on a new field - ecology.

A team of researchers from Harvard, Auburn University, the University of Wyoming, the University of Oxford and the University of Minnesota demonstrated that the artificial intelligence technique can be used to identify animal images captured by motion-sensing cameras.

Using more than three million photographs from the citizen science project Snapshot Serengeti, researchers trained the system to automatically identify, count and describe animals in their natural habitats. Results showed the system was able to automate the process for up to 99.3 percent of images as accurately as human volunteers. The study is described in a June 5 paper published in the Proceedings of the National Academy of Sciences.

Snapshot Serengeti has deployed a large number of "camera traps," or motion-sensitive cameras in Tanzania that collect millions of images of animals in their natural habitat, such as lions, leopards, cheetahs, and elephants.

While the images can offer insight into a host of questions, from how carnivore species co-exist to predator-prey relationships, they are only useful once they have been converted into data that can be processed.

For years, the best method for extracting such information was to ask crowdsourced teams of human volunteers to label each image manually - a laborious and time-consuming process.

"Not only does the artificial intelligence system tell you which of 48 different species of animal is present, it also tells you how many there are and what they are doing. It will tell you if they are eating, sleeping, if babies are present, etc," said Margaret Kosmala, one of the leaders of Snapshot Serengeti and a co-author of the study. "We estimate that the deep learning technology pipeline we describe would save more than 8 years of human labeling effort for each additional 3 million images. That is a lot of valuable volunteer time that can be redeployed to help other projects."

"While there are a number of projects that rely on images captured by camera traps to understand the natural world, few are able to recruit the large numbers of volunteers needed to extract useful data," said Snapshot Serengeti founder Ali Swanson. "The result is that potentially important knowledge remains locked away, out of the reach of scientists.

"Although projects are increasingly turning to citizen science for image classification, we're starting to see it take longer and longer to label each batch of images as the demand for volunteers grows," Swanson added. "We believe deep learning will be key in alleviating the bottleneck for camera trap projects: the effort of converting images into usable data."

A form of computational intelligence loosely inspired by how animal brains see and understand the world, deep learning relies on training neural networks using vast amounts of data. For that process to work, though, the training data must be properly labeled.

"When I told (senior author) Jeff Clune we had 3.2 million labeled images, he stopped in his tracks," said Craig Packer, who heads the Snapshot Serengeti project. "Our citizen scientists have done phenomenal work, but we needed to speed up the process to handle ever greater amounts of data. The deep learning algorithm is amazing and far surpassed my expectations. This is a game changer for wildlife ecology."

Going forward, first-author Mohammad Sadegh Norouzzadeh believes deep learning alogrithms will continue to improve and hopes to see similar systems applied to other ecological data sets.

"Here, we wanted to demonstrate the value of the technology to the wildlife ecology community, but we expect that as more people research how to improve deep learning for this application and publish their datasets, the sky's the limit," he said. "It is exciting to think of all the different ways this technology can help with our important scientific and conservation missions."

"This technology lets us accurately, unobtrusively, and inexpensively collect wildlife data, which could help catalyze the transformation of many fields of ecology, wildlife biology, zoology, conservation biology, and animal behavior into 'big data' sciences," said Jeff Clune, the Harris Associate Professor at the University of Wyoming and a Senior Research Manager at Uber's Artificial Intelligence Labs, and the senior author on the paper. "This will dramatically improve our ability to both study and conserve wildlife and precious ecosystems."


The paper was written by Clune, his PhD student Mohammad Sadegh Norouzzadeh, his former PhD student Anh Nguyen (now at Auburn University), Margaret Kosmala (Harvard University), Ali Swanson (University of Oxford), and Meredith Palmer and Craig Packer (both from the University of Minnesota).

This research was supported with funding from the National Science Foundation.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.