AI Listening Devices Help Sumatra’s Rangers Topple Illegal Loggers

The listening devices help rangers keep tabs on Sumatra's 800 mammal and bird species - as well as illegal loggers.

The conservationists and rangers of Sumatra's diverse rainforests have gained an AI ally in their fight against illegal logging.

Autor*in Mark Newton, 08.07.19

Translation Mark Newton:

The Indonesian island of Sumatra is home to a unique ecosystem which supports some of the world’s most critically endangered animals. It is the only place on Earth where orangutans, tigers, elephants and rhinoceroses live together – although each of these species is teetering on the edge of extinction. 

There are only around 600 Sumatran tigers left in the wild, while the number of rhinos is likely less than 100. While rampant poaching has had an impact, perhaps even more destructive is the deforestation that results from the cultivation of oil palms (grown to produce palm oil) and illegal logging.

To help detect these issues in a wild and often difficult to navigate location, rangers with the Warsi Indonesian Conservation Community (KKI Warsi) are now being backed up the latest Google technology.

A tech nonprofit, named Rainforest Connection, is using artificial intelligence algorithms to plant spies across the rainforest. Their creation – dubbed The Guardian – uses a Google technology known as TensorFlow which allows for intelligent listening devices to record sounds in the rainforest, pick those most pertinent and then forward them in realtime to rangers’ smartphones. The basic technology behind the idea has been around for a few years, and previously Rainforest Connection has provided listening devices to to the Tembe tribe in the Amazon. However, it has now combined with TensorFlow, Google’s free and open-source software library designed to facilitate machine learning applications.

Rainforest Connection has been developing a searchable ‘Audio Ark’ of rainforest sounds, such as animal chirps and calls, as well as the sounds of human activity, from a community of contributors. The TensorFlow technology can compare the sounds it hears in the wild to this library and single out certain sounds for forwarding to rangers. 

The devices themselves are relatively simple, consisting of a solar panel, battery, used cellphone and voice recorder, however it is enough for sounds from the jungle to be recorded, filtered and then sent to the cloud. Within Sumatra, twelve such Guardians were installed around the edges of four villages, namely Sirukam, Pakan Rabaa Timur, Pakan Rabaa and Pasir Palang Timur.

Firstly, this can help with wildlife monitoring, allowing rangers to keep tabs on certain species without the need for invasive, expensive and time consuming monitoring expeditions. For example, Rainforest Connection first trialled the technology by tracking two species of endangered parrot which were being introduced in Ecuador. By recording and forwarding their bird calls, rangers were able to more efficiently organise and support the reintroduction efforts.

For Sumatra, detecting the sounds of logging and human activity is equally as important as tracking the animals. Around 40 to 61 percent of all Indonesian lumber results from illegal forestry, costing the government around 2 billion USD a year from corruption, uncollected taxes and general poor resource management. This deforestation also has dire consequences for Sumatra’s native species, which rely on the cover of the rainforest to survive. In the last 22 years, deforestation has removed 12 million hectares of rainforest from Sumatra – a loss of almost 50 percent.

KKI Warsi program manager Rainal Daus explained that The Guardian helped rangers to respond more efficiently and in strength to real-time incidents:

“Using this tool, our forest rangers will be able to monitor the forest’s condition real-time. This will make it easier for them to take immediate action if there is illegal logging activity taking place at the forest.”

This isn’t the first time Google technology or listening devices have been used in conservation efforts. Previously, we have heard of an eavesdropping AI that is being used to track elephants in Tanzania, while Google’s facial recognition software has been tweaked to spot animals in their habitats, as well as potential poachers.

The French Firm on a Mission to Add All Rhinos to the Internet of Things

More and more devices are becoming part of the so-called Internet of Things. Now, a leading connectivity firm has added rhinoceroses to that list.

Your Smartphone’s Face Recognition Software Is Now Being Used to Identify Dolphins

Algorithms based on Google's face recognition software are now able to identify individual dolphins based on only their dorsal fins - boosting conservation efforts.

The Eavesdropping AI That’s Helping Rangers Track Africa’s Elusive Forest Elephants

AI-backed listening devices aim to track elephants - and their poachers - in the dense African rainforests.

How Satellite Collars for Elephants Are Helping Tanzanian Rangers Get a Technological Edge Over Poachers

A new satellite collaring initiative aims to bolster the dwindling elephant numbers in Tanzania’s national game reserves.

Google’s Facial Recognition AI Helps Identify Endangered Species in the Wild – and Their Poachers

Facial recognition tech is the latest tool in the fight against poaching, with London Zoo embarking on a new trial which could drastically improve conservation efforts.A new project, developed in collaboration with Google, will use machine learning and image recognition software to identify and catalogue animals snapped by remote motion-sensitive camera traps.

PAWS: Artificial Intelligence Helps Patrollers Hunt for Poachers

Computer scientists from the University of Southern California (USC) developed an artificial intelligence tool that helps prevent poaching by predicting where it is likely to happen and suggesting patrolling routes.