An artificial intelligence system that allows self-driving cars to ‘see’ around corners in real time could help prevent accidents, according to its developers.
Researchers from Stanford University in the USA have created a system that bounces a laser beam off a wall to create an ‘image’ of objects hidden from view.
The ‘image’ captured won’t make sense to a human, but using artificial intelligence technology the system can create a visual reconstruction of the hidden view.
The research was funded by US government agency DARPA (Defence Advanced Research Projects Agency), and is one of a number of similar technology programmes being developed.
It could also be used by soldiers to see around walls, rescue workers searching for people and even in space travel to examine the interior caves of an asteroid.
The systems might one day let self-driving cars ‘look’ around parked cars or busy intersections to not only see cars but also read license plates
As well as the Stanford researchers, the team included experts from Princeton University, Southern Methodist University and Rice University.
The researchers used a commercially available camera sensor and a powerful, but standard, laser in the new system – similar to the one found in a laser pointer.
The laser beam bounces off a visible wall onto the hidden object and then back onto the wall, creating an interference pattern known as a speckle.
‘Reconstructing the hidden object from the speckle pattern requires solving a challenging computational problem’, said Metzler.
He said short exposure times are necessary for real-time imaging but produce too much noise for existing algorithms to work.
A camera uses light scattered off of a rough wall, known as a virtual detector, to reconstruct an image of the hidden object. When using a continuous-wave laser, the camera records speckle
To solve this problem, the researchers turned to deep learning, a form of machine learning that is better for interpreting large and varied data.
‘Compared to other approaches for non-line-of-sight imaging, our deep learning algorithm is far more robust to noise and thus can operate with much shorter exposure times,’ said co-author Prasanna Rangarajan.
‘By accurately characterising the noise, we were able to synthesise data to train the algorithm to solve the reconstruction problem.’
Effectively the artificial intelligence system filters out the noise to create an ‘image’ of what is hiding behind the wall or object.
He said they were able to do this using deep learning without having to capture costly training data, as would be needed by traditional machine learning.
‘Our imaging system provides uniquely high resolutions and imaging speeds,’ said research team leader Christopher A. Metzler from Stanford University.
‘These attributes enable applications that wouldn’t otherwise be possible, such as reading the license plate of a hidden car as it is driving’.
It has been designed to image small objects at high resolutions, but can be combined with other systems to produce low-resolution images of larger items.
‘Non-line-of-sight imaging has important applications in medical imaging, navigation, robotics and defence,’ said co-author Felix Heide.
‘Our work takes a step toward enabling its use in a variety of such applications.’
They tested their new technique by recreating images of 0.4 inch tall letters and numbers hidden behind a corner.
The research was funded by DARPA, the Defence Advanced Research Projects Agency and is one of a number of similar technology programmes being developed
An imaging system was setup about one metre from the wall hiding the letters and they used an exposure length of a quarter of a second.
This approach produced reconstructions of the real letters that were hidden behind the wall with a resolution a quarter of the original image height.
The study is part of DARPA’s Revolutionary Enhancement of Visibility by Exploiting Active Light-fields (REVEAL) program, which is developing a variety of different techniques to image hidden objects around corners.
DARPA says on its website: ‘The REVEAL program aims to develop a comprehensive theoretical framework to enable maximum information extraction.
‘Taking it from complex scenes by using all photon pathways and leveraging light’s multiple degrees of freedom.’
The researchers are now working to make the system practical for more applications by extending the field of view so that it can reconstruct larger objects.
The research has been published in the journal Optica.
WHAT IS DEEP LEARNING?
Deep learning is a form of machine learning concerned with algorithms which have a wide range of applications.
It is a field which was inspired by the human brain and focuses on building artificial neural networks.
It was formed originally based on brain simulations and to allow learning algorithms to become better and easier to use.
Processing vast amounts of complex data then becomes much easier and allows researchers to trust algorithms to draw accurate conclusions based on the parameters the researchers have set.
Task-specific algorithms which exist are better for specific tasks and goals but deep-learning allows for a wider scope of data collection.