Google has made its AI model, SpeciesNet, open-source, enabling the identification of animal species through the analysis of camera trap photos.

Worldwide, researchers utilize camera traps, which are digital cameras connected to infrared sensors, to study wildlife populations. However, while these traps provide valuable insights, they generate an enormous amount of data that can take weeks or even days to analyze.

To address this issue, Google introduced Wildlife Insights, an initiative under the company’s Google Earth Outreach philanthropy program, about six years ago. Wildlife Insights offers a platform where researchers can share, identify, and analyze wildlife images online, facilitating collaboration and speeding up the analysis of camera trap data.

Many analysis tools within Wildlife Insights are powered by SpeciesNet, which Google claims was trained on over 65 million publicly available images, as well as images from organizations such as the Smithsonian Conservation Biology Institute, the Wildlife Conservation Society, the North Carolina Museum of Natural Sciences, and the Zoological Society of London.

According to Google, SpeciesNet has the ability to classify images into more than 2,000 labels, including animal species, taxa such as “mammalian” or “Felidae,” and non-animal objects like “vehicle.”

The release of the SpeciesNet AI model is expected to enable tool developers, academics, and biodiversity-related startups to scale the monitoring of biodiversity in natural areas, as stated by Google in a blog post published on Monday.

SpeciesNet is available on GitHub under an Apache 2.0 license, allowing it to be used commercially with minimal restrictions.

It’s worth noting that Google’s SpeciesNet is not the only open-source tool for automating the analysis of camera trap images. The AI for Good Lab at Microsoft maintains PyTorch Wildlife, an AI framework that offers pre-trained models fine-tuned for animal detection and classification.