Hitachi Vantara, the IT arm of Hitachi, has developed a new artificial intelligence (AI) technology with non-profit Rainforest Connection to predict when illegal logging will take place in the rainforest.
The AI solution analyses eco-acoustic data gathered by sensors attached to trees. Called Guardians, these acoustic sensors are made from old smartphones and ‘listen’ to the sounds of the forest to pick out the sound of chainsaws.
Machine learning can identify the sound of illegal logging from up to a kilometre away, picking up sounds that may be undetectable to the human ear.
When a chainsaw is detected, the data is uploaded to the cloud and the audio file is sent to a ranger to review and investigate. While this method has already protected an estimated 3000km², the process of shutting down the illegal logging operation can take up to 14 days, and in this time period trees can be lost.
Therefore, the ability to predict illegal logging before it begins is invaluable.
The non-profit has been using AI and sensors to detect illegal logging since 2014. But now, Hitachi Vantara has built an algorithm using its Lumada data analytics technology to predict when illegal logging might start by listening out for sound anomalies.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below formBy GlobalData
For example, the algorithm can pick up the sound of illegal loggers walking through the forest or preparing a site before beginning their operation.
Hitachi developed the algorithm using decades of audio files to create a baseline rainforest sounds, making it easier to spot acoustic anomalies.
According to Hitachi Vantara, by detecting such sounds before the logging operation begins, response times can be improved by up to 35%, with the algorithm able to make a “very accurate prediction” that illegal logging will take place up to five days in advance.
Since the start of the partnership in 2019, the solution is now live in Sumatra, Indonesia, where over 70% of all deforestation that occurs is illegal. The solution is scheduled to roll out to the whole network of Rainforest Connection guardians, located in over 11 countries across five continents, in 2021.
Rainforest Connection hopes to triple its impact by the end of 2021, protecting 6000km² of rainforest.
According to Brazil’s National Institute for Space Research, deforestation in the Brazilian Amazon Rainforest increased by over 50% in the first three months of 2020 compared to the same period in 2019. Up to 90% of deforestation can be attributed to illegal logging, according to Rainforest Connection.
Hitachi Vantara said that the technology could have applications in a number of industries, such as predicting machinery malfunctions or safety violations in factories, predicting crop pest movement in agriculture, creating more accurate weather forecasts and in search and rescue missions.
It also believes that the audio data collected by Guardians could be valuable for researchers monitoring biodiversity or endangered species.
“The projects with Hitachi Vantara are game-changing for Rainforest Connection,” said Topher White, CEO, Rainforest Connection. “We’ll be able to scale up our operations and provide rangers with greater certainty around when logging events are likely to happen.”