(Press-News.org) Images
A new real-time, 3D motion tracking system developed at the University of Michigan combines transparent light detectors with advanced neural network methods to create a system that could one day replace LiDAR and cameras in autonomous technologies.
While the technology is still in its infancy, future applications include automated manufacturing, biomedical imaging and autonomous driving. A paper on the system is published in Nature Communications.
50928751578_a3702cc26c_q.jpgThe imaging system exploits the advantages of transparent, nanoscale, highly sensitive graphene photodetectors developed by Zhaohui Zhong, U-M associate professor of electrical and computer engineering, and his group. They're believed to be the first of their kind.
"The in-depth combination of graphene nanodevices and machine learning algorithms can lead to fascinating opportunities in both science and technology," said Dehui Zhang, a doctoral student in electrical and computer engineering. "Our system combines computational power efficiency, fast tracking speed, compact hardware and a lower cost compared with several other solutions."
The graphene photodetectors in this work have been tweaked to absorb only about 10% of the light they're exposed to, making them nearly transparent. Because graphene is so sensitive to light, this is sufficient to generate images that can be reconstructed through computational imaging. The photodetectors are stacked behind each other, resulting in a compact system, and each layer focuses on a different focal plane, which enables 3D imaging.
But 3D imaging is just the beginning. The team also tackled real-time motion tracking, which is critical to a wide array of autonomous robotic applications. To do this, they needed a way to determine the position and orientation of an object being tracked. Typical approaches involve LiDAR systems and light-field cameras, both of which suffer from significant limitations, the researchers say. Others use metamaterials or multiple cameras. Hardware alone was not enough to produce the desired results.
They also needed deep learning algorithms. Helping to bridge those two worlds was Zhen Xu, a doctoral student in electrical and computer engineering. He built the optical setup and worked with the team to enable a neural network to decipher the positional information.
The neural network is trained to search for specific objects in the entire scene, and then focus only on the object of interest--for example, a pedestrian in traffic, or an object moving into your lane on a highway. The technology works particularly well for stable systems, such as automated manufacturing, or projecting human body structures in 3D for the medical community.
"It takes time to train your neural network," said project leader Ted Norris, professor of electrical and computer engineering. "But once it's done, it's done. So when a camera sees a certain scene, it can give an answer in milliseconds."
Doctoral student Zhengyu Huang led the algorithm design for the neural network. The type of algorithms the team developed are unlike traditional signal processing algorithms used for long-standing imaging technologies such as X-ray and MRI. And that's exciting to team co-leader Jeffrey Fessler, professor of electrical and computer engineering, who specializes in medical imaging.
"In my 30 years at Michigan, this is the first project I've been involved in where the technology is in its infancy," Fessler said. "We're a long way from something you're going to buy at Best Buy, but that's OK. That's part of what makes this exciting."
The team demonstrated success tracking a beam of light, as well as an actual ladybug with a stack of two 4x4 (16 pixel) graphene photodetector arrays. They also proved that their technique is scalable. They believe it would take as few as 4,000 pixels for some practical applications, and 400x600 pixel arrays for many more.
While the technology could be used with other materials, additional advantages to graphene are that it doesn't require artificial illumination and it's environmentally friendly. It will be a challenge to build the manufacturing infrastructure necessary for mass production, but it may be worth it, the researchers say.
"Graphene is now what silicon was in 1960," Norris said. "As we continue to develop this technology, it could motivate the kind of investment that would be needed for commercialization."
INFORMATION:
The paper is titled "Neural Network Based 3D Tracking with a Graphene Transparent Focal Stack Imaging System." The research is funded by the W.M. Keck Foundation and the National Science Foundation. Devices were fabricated in the U-M Lurie Nanofabrication Facility. Norris is also the Gérard A. Mourou Professor of Electrical Engineering and Computer Science, and a professor of applied physics. Fessler is also the William L. Root Collegiate Professor of Electrical Engineering and Computer Science, and also a professor of biomedical engineering, radiology, and applied physics.
Written by Catharine June
The tree of life just got a little bigger: A team of scientists from the U.S. and China has identified an entirely new group of microbes quietly living in hot springs, geothermal systems and hydrothermal sediments around the world. The microbes appear to be playing an important role in the global carbon cycle by helping break down decaying plants without producing the greenhouse gas methane.
"Climate scientists should take these new microbes into account in their models to more accurately understand how they will impact climate change," said END ...
A patient arrives at hospital with chest pain. Doctors suspect heart attack and rapid diagnosis is important, but the tests to confirm it can be invasive and it could easily be something else. Could a simple blood test help to non-invasively rule heart attack in or out? A new study in open access journal Frontiers in Cardiovascular Medicine certainly suggests so. The study identified telltale markers in the blood of heart attack patients that distinguished them from patients suffering chest pain with other causes. The researchers hope that the results will lead to new diagnostic tests for heart attacks.
If you have ever suffered chest pain, the possibility of a heart attack may have popped into your head. While chest pain is ...
New research provides evidence that people have transmitted SARS-CoV-2, the virus that causes COVID-19, to cats during the pandemic in the UK. The study, which is published in Veterinary Record, detected the virus last year in cats that developed mild or severe respiratory disease.
Investigators used a range of laboratory techniques to show that two domestic cats from households with suspected cases of COVID-19 were infected with SARS-CoV-2.
"These findings indicate that human-to-cat transmission of SARS-CoV-2 occurred during the COVID-19 pandemic in ...
Researchers from the University of Oxford and their partners have today reported findings from a Phase IIb trial of a candidate malaria vaccine, R21/Matrix-M, which demonstrated high-level efficacy of 77% over 12-months of follow-up. In their findings (posted on SSRN/Preprints with The Lancet) they note that they are the first to meet the World Health Organization's Malaria Vaccine Technology Roadmap goal of a vaccine with at least 75% efficacy.
The authors report (in findings in press with The Lancet) from a Phase IIb randomised, controlled, double-blind trial conducted at the Clinical Research Unit of Nanoro (CRUN) / Institut de Recherche en Sciences de la Santé (IRSS), Burkina Faso. 450 participants, aged 5-17 months, were recruited from the catchment area of Nanoro, covering ...
New research led by Curtin University has revealed how radar satellites can improve the ability to detect, monitor, prepare for and withstand natural disasters in Australia including bushfires, floods and earthquakes.
The research used Synthetic Aperture Radar data obtained by the European Space Agency Sentinel-1 satellite, amongst others, to evaluate Australia-specific case studies.
Lead researcher Dr Amy Parker, an ARC Research Fellow from Curtin's School of Earth and Planetary Sciences, said the Sentinel-1 satellite mission provided the first complete global Synthetic Aperture Radar (SAR) dataset and the first opportunity to use this type of data to assess hazards in new locations, including Australia.
"What makes SAR so valuable is that it provides all-weather and night-and-day ...
Researchers from Nanjing University, National Sun Yat-sen University, and Northwestern University published a new paper in the Journal of Marketing that shows that the spatial distance between products and consumers can affect perceived value and willingness to pay.
The study, forthcoming in the Journal of Marketing, is titled "Values Created from Far and Near: Influence of Spatial Distance on Brand Evaluation" and is authored by Xing-Yu (Marcos) Chu, Chun-Tuan Chang, and Angela Y. Lee.
No one ever questions why some retail products are on display in cabinets behind the sales counter, ...
A new species of deep-sea dwelling dumbo octopus called Grimpoteuthis imperator sp. nov. has been described using a combination of MRI, micro-CT and minimally invasive gene analysis rather than traditional dissection methods. The findings are presented in the open access journal BMC Biology.
The single specimen, which was identified as a mature male, was named G. imperator because it was discovered in the northern part of the Emperor Seamounts, an undersea mountain ridge in the Northwest Pacific Ocean. In addition to the scientific name in Latin, the authors suggest possible common names for this new species such as Emperor dumbo, Dumbo impérial (French), and Kaiserdumbo (German).
Finned or dumbo octopods, although considered rare organisms, form ...
Grass crops are able to bend the rules of evolution by borrowing genes from their neighbours, giving them a competitive advantage, a new study has revealed.
Research, led by the University of Sheffield, is the first to show that grasses can incorporate DNA from other species into their genomes through a process known as lateral gene transfer.
The stolen genetic secrets give them an evolutionary advantage by allowing them to grow faster, bigger or stronger and adapt to new environments quicker. These findings could inform future work to create crops that are more resistant to the effects of climate change and help to tackle food security problems.
The Sheffield team studied grasses, which include some of the most economically and ecologically important plants, such as the most ...
A radically reformed approach to education, in which different subjects teach connected themes, like climate change or food security, is being proposed by researchers, who argue that it would better prepare children for future crises.
In a newly-published study, education researchers from the Universities of Cambridge and Edinburgh argue that there is a compelling case for a drastic shake-up of the school curriculum, so that subjects are no longer taught independently of one another. Instead, they argue that the arts and sciences should 'teach together' around real-world problems, and in a manner ...
California produces nearly 4 million tons of world-class wine each year, but with that comes thousands of tons of residue like grape skins, seeds, stems and pulp. What if scientists could harness that viticultural waste to help promote human health?
Maybe they can, according to new research from food scientists at the University of California, Davis. In a study published in the journal LWT - Food Science and Technology, the team discovered a wealth of potentially health-enhancing compounds and sugar molecules called oligosaccharides within chardonnay wine-grape pomace.
Oligosaccharides are found in many plant and animal tissues, including human breast milk. Recent advances have revealed oligosaccharides' vast potential ...
