Group protects rainforest with recycled phones, machine learning

Group protects rainforest with recycled phones, machine learning

Rainforest Connection, a San Francisco-based non-profit organisation, is using recycled phones and Google’s open source machine learning framework, TensorFlow, to protect the rainforest.

The group has created a low-cost network that listens for sounds of illegal deforestation, and analyses the data.

According to Topher White, founder and CEO of Rainforest Connection, destruction of forests accounts for a rise of nearly 20 percent in greenhouse gas emissions every year. Tropical deforestation has been accelerated by rampant logging, 90 percent of which is done illegally and undercover.

Protecting the world’s rainforests may be the fastest and cheapest way to slow climate change, believes White, and locals and indigenous tribes are best suited to protect critical regions.

“Rainforest Connection is a group of engineers and developers focused on building technology to help locals – like the Tembé tribe from central Amazon – protect their land, and in the process, protect the rest of us from the effects of climate change,” said White in a blog post.

“Chief Naldo Tembé reached out to me a couple years ago seeking to collaborate on ways that technology could help stop illegal loggers from destroying their land. Together, we embarked on an ambitious plan to address this issue using recycled cellphones and machine learning.”

White’s team has built what it believes is the world’s first scalable, real-time detection and alert system for logging and environmental conservation.

The team has hidden modified smartphones powered by solar panels – dubbed ‘Guardian’ devices’ – in trees across threatened areas. The phones monitor the sounds of the forest, and send the audio to cloud-based servers over the local cellphone network.

This is where machine learning steps in. TensorFlow is used to analyse all of the audio data in real-time, listening out for chainsaws, logging trucks, and other evidence of illegal activity. The audio is constantly being sent to the cloud from every phone, 24 hours a day.

White said that the stakes of missing illegal activity are high. “That’s why we’ve come to use TensorFlow, due to its ability to analyse every layer of our data-heavy detection process.

“The versatility of the machine learning framework empowers us to use a wide range of AI techniques with deep learning on one unified platform. This allows us to tweak our audio inputs and improve detection quality.

“Without the help of machine learning, this process would be impossible. When fighting deforestation, every improvement can mean one more saved tree,” he explained.

Read: Bee robotic: Walmart files patents on automating agriculture 

Internet of Business says

This brilliant scheme mixes low-cost, but sophisticated technology – recycled phones – and cloud-based smart analytics. The environmental benefits are twofold: not only does the scheme help protect the rainforest, but it also keeps the phones out of landfills and uses both their processing power and their network connections.

The use of audio, rather than video is smart too: less data, and zero reliance on light and visibility in dense areas of foliage, especially if illegal activities take place at night.

A brilliant, low-cost, connected scheme that gathers data, gets smarter, and has obvious benefits for human beings: a model for IoT developments of every kind.

Read more: How IoT, smart supply chains can avert global food crisis

Read more: Dell takes a fresh look at IoT with Aerofarms