Some video's included
Poachers hunting some of the world's most endangered animals could be stopped using night-vision drones loaded with Nasa's galaxy-tracking technology
Technology used by Nasa for studying faint stars and galaxies could be used to trap poachers in a rare collaboration between astronomers and ecologists.
The system combines flying drones, infra-red thermal imaging, and artificial intelligence to monitor animals at night when most poaching occurs.
In an early field trial in South Africa, it was used to detect elusive riverine rabbits, one of the most endangered mammals in the world.
Researchers hope in future the heat-sensitive drones will make it easier to track animals such as rhinos and spot poachers hunting them under cover of darkness.
The technology works using AI software designed to pick out distant stars and galaxies in images of space that are invisible to the naked eye.
Nasa has used similar technology loaded onto its space probes to spot celestial objects that would otherwise be missed by its scientists.
Researchers at Liverpool John Moores University worked alongside experts at Chester Zoo and Knowsley Safari Park to re-programme the software using thousands of images of animals.
Project scientist Dr Claire Burke, from Liverpool John Moores University, said: 'With thermal infrared cameras, we can easily see animals as a result of their body heat, day or night, and even when they are camouflaged in their natural environment.
'Since animals and humans in thermal footage "glow" in the same way as stars and galaxies in space, we have been able to combine the technical expertise of astronomers with the conservation knowledge of ecologists to develop a system to find the animals or poachers automatically.'
The researchers 'trained' the AI software to recognise different types of animals in a range of landscapes and vegetation.
The team has also developed software that models the effects of vegetation blocking body heat, enabling the detection of animals hidden by trees or leaves.
Further upgrades will compensate for atmospheric effects, weather and other environmental factors.
Most poaching happens at night in the dark, when it is difficult for gamekeepers to spot hunters and the animals.
The new drone technology can scan large areas of terrain and monitor regions that are hard to reach without disturbing the animals.
In a trial in South Africa last September, the scientists used the system to track riverine rabbits - a notoriously elusive species.
Describing the trial, Dr Burke said: 'The rabbits are very small, so we flew the drone quite low to the ground at a height of 20 metres [65 feet]. Although this limited the area we could cover with the drone, we managed five sightings.
'Given that there have only been about 1,000 sightings of riverine rabbits by anyone in total, it was a real success.'
In May the astro-ecologists will carry out more field trials looking for orangutans in Malaysia and spider monkeys in Mexico.
Then in June they will carry out a search for Brazilian river dolphins.
'Our aim is to make a system that is easy for conservationists and game wardens to use anywhere in the world, which will allow endangered animals to be tracked, found and monitored easily and poaching to be stopped before it happens,' said Dr Burke.
HOW DOES ARTIFICIAL INTELLIGENCE LEARN?
AI systems rely on artificial neural networks (ANNs), which try to simulate the way the brain works in order to learn.
ANNs can be trained to recognise patterns in information - including speech, text data, or visual images - and are the basis for a large number of the developments in AI over recent years.
Conventional AI uses input to 'teach' an algorithm about a particular subject by feeding it massive amounts of information.
Practical applications include Google's language translation services, Facebook's facial recognition software and Snapchat's image altering live filters.
The process of inputting this data can be extremely time consuming, and is limited to one type of knowledge.
A new breed of ANNs called Adversarial Neural Networks pits the wits of two AI bots against each other, which allows them to learn from each other.
This approach is designed to speed up the process of learning, as well as refining the output created by AI systems.
Source: http://www.dailymail.co.uk/sciencetech/article-5574577/Out-world-thermal-imaging-aimed-poachers.html
Poachers hunting some of the world's most endangered animals could be stopped using night-vision drones loaded with Nasa's galaxy-tracking technology
- System combines drones, thermal imaging, and AI to monitor animals at night
- In an early field trial in South Africa, it was used to detect elusive riverine rabbits
- The heat-sensitive drones could make it easier to track animals such as rhinos
- They could also help law enforcement agencies spot illegal poachers hunting under cover of darkness
Technology used by Nasa for studying faint stars and galaxies could be used to trap poachers in a rare collaboration between astronomers and ecologists.
The system combines flying drones, infra-red thermal imaging, and artificial intelligence to monitor animals at night when most poaching occurs.
In an early field trial in South Africa, it was used to detect elusive riverine rabbits, one of the most endangered mammals in the world.
Researchers hope in future the heat-sensitive drones will make it easier to track animals such as rhinos and spot poachers hunting them under cover of darkness.
The technology works using AI software designed to pick out distant stars and galaxies in images of space that are invisible to the naked eye.
Nasa has used similar technology loaded onto its space probes to spot celestial objects that would otherwise be missed by its scientists.
Researchers at Liverpool John Moores University worked alongside experts at Chester Zoo and Knowsley Safari Park to re-programme the software using thousands of images of animals.
Project scientist Dr Claire Burke, from Liverpool John Moores University, said: 'With thermal infrared cameras, we can easily see animals as a result of their body heat, day or night, and even when they are camouflaged in their natural environment.
'Since animals and humans in thermal footage "glow" in the same way as stars and galaxies in space, we have been able to combine the technical expertise of astronomers with the conservation knowledge of ecologists to develop a system to find the animals or poachers automatically.'
The researchers 'trained' the AI software to recognise different types of animals in a range of landscapes and vegetation.
The team has also developed software that models the effects of vegetation blocking body heat, enabling the detection of animals hidden by trees or leaves.
Further upgrades will compensate for atmospheric effects, weather and other environmental factors.
Most poaching happens at night in the dark, when it is difficult for gamekeepers to spot hunters and the animals.
The new drone technology can scan large areas of terrain and monitor regions that are hard to reach without disturbing the animals.
In a trial in South Africa last September, the scientists used the system to track riverine rabbits - a notoriously elusive species.
Describing the trial, Dr Burke said: 'The rabbits are very small, so we flew the drone quite low to the ground at a height of 20 metres [65 feet]. Although this limited the area we could cover with the drone, we managed five sightings.
'Given that there have only been about 1,000 sightings of riverine rabbits by anyone in total, it was a real success.'
In May the astro-ecologists will carry out more field trials looking for orangutans in Malaysia and spider monkeys in Mexico.
Then in June they will carry out a search for Brazilian river dolphins.
'Our aim is to make a system that is easy for conservationists and game wardens to use anywhere in the world, which will allow endangered animals to be tracked, found and monitored easily and poaching to be stopped before it happens,' said Dr Burke.
HOW DOES ARTIFICIAL INTELLIGENCE LEARN?
AI systems rely on artificial neural networks (ANNs), which try to simulate the way the brain works in order to learn.
ANNs can be trained to recognise patterns in information - including speech, text data, or visual images - and are the basis for a large number of the developments in AI over recent years.
Conventional AI uses input to 'teach' an algorithm about a particular subject by feeding it massive amounts of information.
Practical applications include Google's language translation services, Facebook's facial recognition software and Snapchat's image altering live filters.
The process of inputting this data can be extremely time consuming, and is limited to one type of knowledge.
A new breed of ANNs called Adversarial Neural Networks pits the wits of two AI bots against each other, which allows them to learn from each other.
This approach is designed to speed up the process of learning, as well as refining the output created by AI systems.
Source: http://www.dailymail.co.uk/sciencetech/article-5574577/Out-world-thermal-imaging-aimed-poachers.html
