Orangutans are some of the most intelligent primates in nature, with capabilities including tools, colony living, and even rudimentary construction skills. However, the species is critically endangered due to humans—specifically, illegal activities such as poaching, black market trading, and the ongoing destruction of their natural habitat.
According to WWF, orangutan numbers in Borneo have declined by more than 50% over the past 60 years or so, and one of the leading causes of this is the destruction (at least 55%) of their habitat over the last two decades. In Indonesia, WWF has pushed conservation efforts in thhe Sebangau National Park, with experts going into the field everyday to monitor and track populations, as well as to collect data on the primates.
Generally speaking, it takes up to 3 days to process each batch of data, with thousands of photos to handle and analayse. This, consequently, is a process that can be affected by human error, especially with the high volume of data being procesed on a regular basis.
How can machine learning technology help this?
According to WWF-Indonesia, the conservation efforts can be “challenging and resource consuming” due to the difficulty of identifying each species. For the study of Orangutans within their natural habitat, it can be difficult to identify individual Orangutans. This can slow down and affect monitoring for specific colonies of primates.
To better deal with that, WWF-Indonesia has announced a collaboration with Amazon Web Services (AWS) to save the critically-endangered orangutan species in Indonesia. AWS’ machine learning technologies help experts in the field to better understand the size, as well as the well-being of orangutan populations within the habitat. Additionally, these technologies will also help WWF to expand their monitoring and tracking efforts across more territories, while conserving critical resources and funding.
Basically, technologies such as Amazon SageMaker, Amazon API Gateway, Amazon CloudWatch, and AWS Lambda help WWF monitor and track populations through the help of motion-activated cameras and mobile phones. This data is then uploaded to Amazon Simple Storage Service (Amazon S3) for further processing. However, the intelligent capabilities of machine learning tech means that the analysis time—previously a 3-day process—is now reduced to under 10 minutes. Accuracy and specificity of data is also better, while human error is no longer a concern.
“By adopting machine learning, WWF-Indonesia has reduced its reliance on a limited pool of conservationist experts and improved the accuracy and breadth of its data about orangutan populations.”
Face recognition technology is also now used to identify individual primates within populations, and this is something that could possibly be integrated into mobile apps, according to WWF.
And as the partnership continues to mature, WWF intends to look into more machine learning technologies from AWS to further improve the “speed and accuracy” of their indentification and tracking efforts. Accorfing to Aria Nagasastra, Finance and Technology Director of WWF-Indonesia:
“Using AWS services like Amazon SageMaker and Amazon S3, we are starting to make a tool accessible for the field surveyors, even with limited expertise and capacity, to identify wildlife in its natural habitat with a high degree of accuracy. With careful use of technology, this innovation will help the biologists and conservationists to effectively and cost-efficiently monitor the wildlife behavior through time and thus we can allocate our resources to scale up the monitoring efforts and invest more in conservation actions.”
[ IMAGE SOURCE: WWF-Indonesia ]