Deep learning to identify and count wild animals

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. This will dramatically improve our ability to both study and conserve wildlife and precious ecosystems.

Motion sensor camera trap' unobtrusively take pictures of animals in their natural environment, oftentimes yielding images not otherwise observable. The information in these photographs is only useful once it has been converted into numerical data. For years, the best method for extracting such information was to involve crowdsourced teams of human volunteers to label each image manually.

A team of researchers form the US and the UK has developed a system to automatically extract such information from images by using deep neural networks. The result is a system that can automate animal identification for up to 99.3 percent of images while still performing at the same 96.6 percent accuracy rate of crowdsourced teams of human volunteers. Deep neural networks are artificial neural networks with multiple hidden layers between the input and output layers. They require vast amounts of training data to work well, and the data must be accurately labeled (e.g., each image being correctly tagged with which species of animal is present, how many there are, etc.). For this study such data was available through Snapshot Serengeti, a citizen science project. Snapshot Serengeti has deployed a large number of camera traps in Tanzania that collect millions of images of animals in their natural habitat, such as lions, leopards, cheetahs and elephants. For this study 3.2 million labeled images tagged by more than 50,000 human volunteers over several years were used as training set.

Not only does the artificial intelligence system tell you which of 48 different species of animal is present, but 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. We estimate that the deep learning technology pipeline we describe would save more than eight 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.