Computers Math

A new type of active pixel sensor that uses a novel two-dimensional material may both enable ultra-sharp cellphone photos and create a new class of extremely energy-efficient Internet of Things (IoT) sensors, according to a team of Penn State researchers.
“When people are looking for a new phone, what are the specs that they are looking for?” said Saptarshi Das, associate professor of engineering science and mechanics and lead author of the study published Nov. 17 in Nature Materials. “Quite often, they are looking for a good camera, and what does a good camera mean to most people? Sharp photos with high resolution.”
Most people just snap a photo of a friend, a family gathering, or a sporting event, and never think about what happens “behind the scenes” inside the phone when one snaps a picture. According to Das, there is quite a bit happening to enable you to see a photo right after you take it, and this involves image processing.
“When you take an image, many of the cameras have some kind of processing that goes on in the phone, and in fact, this sometimes makes the photo look even better than what you are seeing with your eyes,” Das said. “These next generation of phone cameras integrate image capture with image processing to make this possible, and that was not possible with older generations of cameras.”
However, the great photos in the newest cameras have a catch — the processing requires a lot of energy.
“There’s an energy cost associated with taking a lot of images,” said Akhil Dodda, a graduate research assistant at Penn State at the time of the study who is now a research staff member at Western Digital, and co-first author of the study. “If you take 10,000 images, that is fine, but somebody is paying the energy costs for that. If you can bring it down by a hundredfold, then you can take 100 times more images and still spend the same amount of energy. It makes photography more sustainable so that people can take more selfies and other pictures when they are traveling. And this is exactly where innovation in materials comes into the picture.”
The innovation in materials outlined in the study revolves around how they added in-sensor processing to active pixel sensors to reduce their energy use. So, they turned to a novel 2D material, which is a class of materials only one or a few atoms thick, molybdenum disulfide. It is also a semiconductor and sensitive to light, which makes it ideal as a potential material to explore for low-energy in-sensor processing of images. More

Hummingbirds occupy a unique place in nature: They fly like insects but have the musculoskeletal system of birds. According to Bo Cheng, the Kenneth K. and Olivia J. Kuo Early Career Associate Professor in Mechanical Engineering at Penn State, hummingbirds have extreme aerial agility and flight forms, which is why many drones and other aerial vehicles are designed to mimic hummingbird movement. Using a novel modeling method, Cheng and his team of researchers gained new insights into how hummingbirds produce wing movement, which could lead to design improvements in flying robots.
Their results were published this week in the Proceedings of Royal Society B.
“We essentially reverse-engineered the inner working of the wing musculoskeletal system — how the muscles and skeleton work in hummingbirds to flap the wings,” said first author and Penn State mechanical engineering graduate student Suyash Agrawal. “The traditional methods have mostly focused on measuring activity of a bird or insect when they are in natural flight or in an artificial environment where flight-like conditions are simulated. But most insects and, among birds specifically, hummingbirds are very small. The data that we can get from those measurements are limited.”
The researchers used muscle anatomy literature, computational fluid dynamics simulation data and wing-skeletal movement information captured using micro-CT and X-ray methods to inform their model. They also used an optimization algorithm based on evolutionary strategies, known as the genetic algorithm, to calibrate the parameters of the model. According to the researchers, their approach is the first to integrate these disparate parts for biological fliers.
“We can simulate the whole reconstructed motion of the hummingbird wing and then simulate all the flows and forces generated by the flapping wing, including all the pressure acting on the wing,” Cheng said. “From that, we are able to back-calculate the required total muscular torque that is needed to flap the wing. And that torque is something we use to calibrate our model.”
With this model, the researchers uncovered previously unknown principles of hummingbird wing actuation.
The first discovery, according to Cheng, was that hummingbirds’ primary muscles, that is, their flight engines, do not simply flap their wings in a simple back and forth motion, but instead pull their wings in three directions: up and down, back and forth, and twisting — or pitching — of the wing. The researchers also found that hummingbirds tighten their shoulder joints in both the up-and-down direction and the pitch direction using multiple smaller muscles.
“It’s like when we do fitness training and a trainer says to tighten your core to be more agile,” Cheng said. “We found that hummingbirds are using similar kind of a mechanism. They tighten their wings in the pitch and up-down directions but keep the wing loose along the back-and-forth direction, so their wings appear to be flapping back and forth only while their power muscles, or their flight engines, are actually pulling the wings in all three directions. In this way, the wings have very good agility in the up and down motion as well as the twist motion.”
While Cheng emphasized that the results from the optimized model are predictions that will need validation, he said that it has implications for technological development of aerial vehicles.
“Even though the technology is not there yet to fully mimic hummingbird flight, our work provides essential principles for informed mimicry of hummingbirds hopefully for the next generation of agile aerial systems,” he said.
The other authors were Zafar Anwar, a doctoral student in the Penn State Department of Mechanical Engineering; Bret W. Tobalske of the Division of Biological Sciences at the University of Montana; Haoxiang Luo of the Department of Mechanical Engineering at Vanderbilt University; and Tyson L. Hedrick of the Department of Biology at the University of North Carolina.
The Office of Naval Research funded this work.
Story Source:
Materials provided by Penn State. Original written by Sarah Small. Note: Content may be edited for style and length. More

Virtual environments, including those for commerce, are increasingly common so as to provide an experience for the user that is as realistic as possible. However, virtual environments also provide a new opportunity for researchers to conduct experiments that would not be possible in the real world. Researchers from the University of Tsukuba have done just that by exploring how changing the position of the virtual shop assistant’s voice from its visual position would impact the shopping experience of humans in a virtual reality store.
Humans locate sound by combining visual and auditory cues. Because the visual cues are generally less variable, they can override audio cues, leading to the well-known ventriloquism effect, which occurs when a human perceives the location of a sound to be different from its actual source. It is also well known that humans have personal space, which varies according to social, personal, and environmental factors. Although both phenomena have long been studied individually, until the development of virtual reality, it has not been possible to study how the ventriloquism effect alters personal space.
“In particular, we wanted to know how it affects the rapport between the user and shop assistant,” says Professor Zempo Keiichi, lead author of the study. Rapport, or the quality of interpersonal service, strongly affects loyalty and satisfaction, and skilled salespeople use several techniques to build rapport with customers.
In their experiments, the researchers asked 16 people in the virtual shop environment to define their personal space and record their impression when approached by shop assistants. Some assistants had both a voice and an image at the same position, and others had a voice that was located at different distances between the user and assistant.
“We found that rapport was not affected when the deviation between the sound and visual positions could not be tolerated; however, when it could be tolerated, we found two distinct phenomena,” explains Professor Zempo Keiichi. The first was similar to the “uncanny valley,” which occurs when an imperfect human representation invokes feelings of uneasiness in a real human. This decreased rapport with the virtual assistant. But when the sound moved even closer to the human, the rapport increased.
The authors call this phenomenon the “mouth-in-the-door” phenomenon because it is similar to the “foot-in-the-door” phenomenon, in which a small, unconscious consent, such as not moving away when someone starts to speak, causes a person to improve their evaluation of the other person. Without these virtual experiments, this phenomenon would have likely remained undiscovered. But now that it is known, the authors believe it can be used to improve the user experience, especially in virtual shop scenarios.
Story Source:
Materials provided by University of Tsukuba. Note: Content may be edited for style and length. More

Self-driving cars, or autonomous vehicles, have long been earmarked as the next generation mode of transport. To enable the autonomous navigation of such vehicles in different environments, many different technologies relating to signal processing, image processing, artificial intelligence deep learning, edge computing, and IoT, need to be implemented.
One of the largest concerns around the popularization of autonomous vehicles is that of safety and reliability. In order to ensure a safe driving experience for the user, it is essential that an autonomous vehicle accurately, effectively, and efficiently monitors and distinguishes its surroundings as well as potential threats to passenger safety.
To this end, autonomous vehicles employ high-tech sensors, such as Light Detection and Ranging (LiDaR), radar, and RGB cameras that produce large amounts of data as RGB images and 3D measurement points, known as a “point cloud.” The quick and accurate processing and interpretation of this collected information is critical for the identification of pedestrians and other vehicles. This can be realized through the integration of advanced computing methods and Internet-of-Things (IoT) into these vehicles, which allows for fast, on-site data processing and navigation of various environments and obstacles more efficiently.
In a recent study published in the IEEE Transactions of Intelligent Transport Systems journal on 17 October 2022, a group of international researchers, led by Professor Gwanggil Jeon from Incheon National University, Korea have now developed a smart IoT-enabled end-to-end system for 3D object detection in real time based on deep learning and specialized for autonomous driving situations.
“For autonomous vehicles, environment perception is critical to answer a core question, ‘What is around me?’ It is essential that an autonomous vehicle can effectively and accurately understand its surrounding conditions and environments in order to perform a responsive action,” explains Prof. Jeon. “We devised a detection model based onYOLOv3, a well-known identification algorithm. The model was first used for 2D object detection and then modified for 3D objects,” he elaborates.
The team fed the collected RGB images and point cloud data as input to YOLOv3, which, in turn, output classification labels and bounding boxes with confidence scores. They then tested its performance with the Lyft dataset. The early results revealed that YOLOv3 achieved an extremely high accuracy of detection ( >96%) for both 2D and 3D objects, outperforming other state-of-the-art detection models.
The method can be applied to autonomous vehicles, autonomous parking, autonomous delivery, and future autonomous robots as well as in applications where object and obstacle detection, tracking, and visual localization is required. “At present, autonomous driving is being performed through LiDAR-based image processing, but it is predicted that a general camera will replace the role of LiDAR in the future. As such, the technology used in autonomous vehicles is changing every moment, and we are at the forefront,” highlights Prof. Jeon. “Based on the development of element technologies, autonomous vehicles with improved safety should be available in the next 5-10 years,” he concludes optimistically.
Story Source:
Materials provided by Incheon National University. Note: Content may be edited for style and length. More

Scientists have shown how three vortices can be linked in a way that prevents them from being dismantled. The structure of the links resembles a pattern used by Vikings and other ancient cultures, although this study focused on vortices in a special form of matter known as a Bose-Einstein condensate. The findings have implications for quantum computing, particle physics and other fields.
Postdoctoral researcherToni Annala uses strings and water vortices to explain the phenomenon: ‘If you make a link structure out of, say, three unbroken strings in a circle, you can’t unravel it because the string can’t go through another string. If, on the other hand, the same circular structure is made in water, the water vortices can collide and merge if they are not protected.’
‘In a Bose-Einstein condensate, the link structure is somewhere between the two,’ says Annala, who began working on this in Professor Mikko Möttönen’s research group at Aalto University before moving back to the University of British Columbia and then to the Institute for Advanced Study in Princeton. Roberto Zamora-Zamora, a postdoctoral researcher in Möttönen’s group, was also involved in the study.
The researchers mathematically demonstrated the existence of a structure of linked vortices that cannot break apart because of their fundamental properties. ‘The new element here is that we were able to mathematically construct three different flow vortices that were linked but could not pass through each other without topological consequences. If the vortices interpenetrate each other, a cord would form at the intersection, which binds the vortices together and consumes energy. This means that the structure cannot easily break down,’ says Möttönen.
From antiquity to cosmic strands
The structure is conceptually similar to the Borromean rings, a pattern of three interlinked circles which has been widely used in symbolism and as a coat of arms. A Viking symbol associated with Odin has three triangles interlocked in a similar way. If one of the circles or triangles is removed, the entire pattern dissolves because the remaining two are not directly connected. Each element thus links its two partners, stabilising the structure as a whole.
The mathematical analysis in this research shows how similarly robust structures could exist between knotted or linked vortices. Such structures might be observed in certain types of liquid crystals or condensed matter systems and could affect how those systems behave and develop.
‘To our surprise, these topologically protected links and knots had not been invented before. This is probably because the link structure requires vortices with three different types of flow, which is much more complex than the previously considered two-vortex systems,’ says Möttönen.
These findings may one day help make quantum computing more accurate. In topological quantum computing, the logical operations would be carried out by braiding different types of vortices around each other in various ways. ‘In normal liquids, knots unravel, but in quantum fields there can be knots with topological protection, as we are now discovering,’ says Möttönen.
Annala adds that ‘the same theoretical model can be used to describe structures in many different systems, such as cosmic strings in cosmology.’ The topological structures used in the study also correspond to the vacuum structures in quantum field theory. The results could therefore also have implications for particle physics.
Next, the researchers plan to theoretically demonstrate the existence of a knot in a Bose-Einstein condensate that would be topologically protected against dissolving in an experimentally feasible scenario. ‘The existence of topologically protected knots is one of the fundamental questions of nature. After a mathematical proof, we can move on to simulations and experimental research,’ says Möttönen.
Story Source:
Materials provided by Aalto University. Note: Content may be edited for style and length. More

Spintronic devices and their operation are governed by the microstructures of magnetic domains. These magnetic domain structures undergo complex, drastic changes when an external magnetic field is applied to the system. The resulting fine structures are not reproducible, and it is challenging to quantify the complexity of magnetic domain structures. Our understanding of the magnetization reversal phenomenon is, thus, limited to crude visual inspections and qualitative methods, representing a severe bottleneck in material design. It has been difficult to even predict the stability and shape of the magnetic domain structures in Permalloy, which is a well-known material studied over a century.
Addressing this issue, a team of researchers headed by Professor Masato Kotsugi from Tokyo University of Science, Japan, recently developed an AI-based method for analyzing material functions in a more quantitative manner. In their work published in Science and Technology of Advanced Materials: Methods, the team used topological data analysis and developed a super-hierarchical and explanatory analysis method for magnetic reversal processes. In simple words, super-hierarchical means, according to research team, the connection between micro and macro properties, which are usually treated as isolated but, in the big scheme, contribute jointly to the physical explanation.
The team quantified the complexity of the magnetic domain structures using persistent homology, a mathematical tool used in computational topology that measures topological features of data persisting across multiple scales. The team further visualized the magnetization reversal process in two-dimensional space using principal component analysis, a data analysis procedure that summarizes large datasets by smaller “summary indices,” facilitating better visualization and analysis. As Prof. Kotsugi explains, “The topological data analysis can be used for explaining the complex magnetization reversal process and evaluating the stability of the magnetic domain structure quantitatively.” The team discovered that slight changes in the structure invisible to the human eye that indicated a hidden feature dominating the metastable/stable reversal processes can be detected by this analysis. They also successfully determined the cause of the branching of the macroscopic reversal process in the original microscopic magnetic domain structure.
The novelty of this research lies in its ability to connect magnetic domain microstructures and macroscopic magnetic functions freely across hierarchies by applying the latest mathematical advances in topology and machine learning. This enables the detection of subtle microscopic changes and subsequent prediction of stable/metastable states in advance that was hitherto impossible. “This super-hierarchical and explanatory analysis would improve the reliability of spintronics devices and our understanding of stochastic/deterministic magnetization reversal phenomena,” says Prof. Kotsugi.
Interestingly, the new algorithm, with its superior explanatory capability, can also be applied to study chaotic phenomenon as the butterfly effect. On the technological front, it could potentially improve the reliability of next generation magnetic memory writing, aid the development of new hardware for the next generation of devices.
Story Source:
Materials provided by Tokyo University of Science. Note: Content may be edited for style and length. More

A new automated approach to detect doxing — a form of cyberbullying in which certain private or personally identifiable information is publicly shared without an individual’s consent or knowledge — may help social media platforms better protect their users, according to researchers from Penn State’s College of Information Sciences and Technology.
The research on doxing could lead to more immediate flagging and removal of sensitive personal information that has been shared without the owner’s authorization. To date, the research team has only studied Twitter, where their novel proposed approach uses machine learning to differentiate which tweet containing personally identifiable information is maliciously shared rather than self-disclosed.
They have identified an approach that was able to automatically detect doxing on Twitter with over 96% accuracy, which could help the platform — and eventually other social media platforms — more quickly and easily identify true cases of doxing.
“The focus is to identify cases where people collect sensitive personal information about others and publicly disclose it as a way of scaring, defaming, threatening or silencing them,” said Younes Karimi, doctoral candidate and lead author on the paper. “This is dangerous because once this information is posted, it can quickly be shared with many people and even go beyond Twitter. The person to whom the information belongs needs to be protected.”
In their work, the researchers collected and curated a dataset of nearly 180,000 tweets that were likely to contain doxed information. Using machine learning techniques, they categorized the data as containing personal information tied to either an individual’s identity — their social security number — or an individual’s location — their IP address — and manually labeled more than 3,100 of the tweets that were found to contain either piece of information. They then further classified the data to differentiate malicious disclosures from self-disclosures. Next, the researchers examined the tweets for common potential motivations behind disclosures, determined whether the intent was likely defensive or malicious, and indicated whether it could be characterized as doxing.
“Not all doxing instances are necessarily malicious,” explained Karimi. “For example, a parent of a missing child might benignly share their private information with the desperate hope of finding them.”
Next, the researchers used nine different approaches based on existing natural language processing methods and models to automatically detect instances of doxing and malicious disclosures of two types of most sensitive private information, social security number and IP address, in their collected dataset. They compared the results and identified the approach with the highest accuracy rate, presenting their findings in November at the 25th ACM Conference on Computer-Supported Cooperative Work and Social Computing.
According to Karimi, this work is especially critical in a time when leading social media platforms — including Twitter — are conducting mass layoffs, minimizing the number of workers responsible for reviewing content that may violate the platforms’ terms of service. One platform’s policy, for example, states that unless a case of doxing has clearly abusive intent, the owner of the publicly shared information or their authorized representative must contact the platform before enforcement action is taken. Under this policy, private information could remain publicly available for long periods of time if the owner of the information is not aware that it has been shared.
“While there exist some prior studies on detection of private information in general and some automated approaches for detecting cyberbullying are applied by social media platforms, they do not differentiate self-disclosures from malicious disclosures of second- and third-parties in tweets,” he said. “Fewer people are now in charge of taking action for these manual user reports, so adding automation can help them to narrow down the most important and sensitive reports and prioritize them.”
Karimi collaborated with Anna Squicciarini, Frymoyer Chair in Information Sciences and Technology, and Shomir Wilson, assistant professor of information sciences and technology, on the paper.
Story Source:
Materials provided by Penn State. Original written by Jess Hallman. Note: Content may be edited for style and length. More

During the holidays, kids often spend more time on screens, leaving parents to wonder: Is it causing harm? Possibly.
For preteens, the odds of developing OCD over a two-year period increased by 13% for every hour they played video games and by 11% for every hour they watched videos, according to a new national study led by UC San Francisco researchers that publishes Dec. 12 in the Journal of Adolescent Health.
“Children who spend excessive time playing video games report feeling the need to play more and more and being unable to stop despite trying,” said Jason Nagata, MD, lead author of the study and assistant professor of pediatrics at UCSF. “Intrusive thoughts about video game content could develop into obsessions or compulsions.”
Watching videos, too, can allow for compulsive viewing of similar content — and algorithms and advertisements can exacerbate that behavior, he added.
OCD is a mental health condition involving recurrent and unwanted thoughts as well as repetitive behaviors that a person feels driven to perform. These intrusive thoughts and behaviors can become severely disabling for the sufferers and those close to them.
“Screen addictions are associated with compulsivity and loss of behavioral control, which are core symptoms of OCD,” Nagata said.
Create a Family Media Plan
Researchers asked 9,204 preteens ages 9-10 years how much time they spent on different types of platforms; the average was 3.9 hours per day. Two years later, the researchers asked their caregivers about OCD symptoms and diagnoses. Use of screens for educational purposes was excluded.
At the two-year mark, 4.4% of preteens had developed new-onset OCD. Video games and streaming videos were each connected to higher risk of developing OCD. Texting, video chat and social media didn’t link individually with OCD, but that may be because the preteens in the sample didn’t use them much, researchers said. Results may differ for older teens, they added.
In July, Nagata and his colleagues discovered excessive screen time was linked to disruptive behavior disorders in 9-11 year olds, though social media was the biggest contributor in that case. In 2021, they found adolescent screen time had doubled during the pandemic.
“Although screen time can have important benefits such as education and increased socialization, parents should be aware of the potential risks, especially to mental health,” said Nagata. “Families can develop a media use plan which could include screen-free times including before bedtime.”
Story Source:
Materials provided by University of California – San Francisco. Original written by Jess Berthold. Note: Content may be edited for style and length. More