Artificial Intelligence Development


Dangerous ways computer worms are spreading among smartphones

Scanning 2D barcodes, finding free Wi-Fi access points, sending SMS messages, listening to music, and watching MP4 videos: these are very common activities that we do using our smartphones. Can you imagine that simply doing these things can get your smarphones infected with "worms" that can not only steal personal information from your phone, but also infect your friends's phones.

Sound scary? It will not be long before worms like this spread among smartphones. What makes the attacks feasible is an emerging technology called HTML5-based app development, and it has been rapidly gaining popularity in the mobile industry. When the adoption of this technology reaches certain threshold, attacks like this will become quite common, unless we do something to stop it. A recent Gartner report says that by 2016, fifty percent of the mobile apps will be using HTML5-based technologies.

What platforms are affected?

All major mobile systems will be affected, including Android, iOS, Blackberry, Windows Phone, etc., because they all support HTML5-based mobile apps.

A notorious problem of the HTML5-based technology is that malicious code can be easily injected into the program and get executed. That is why the Cross-Site Scripting (XSS) attack is still one of the most common attacks in the Web. XSS attacks can only target at web applications through a single channel (i.e. the Internet), but with the adoption of the same technology in mobile devices, we have found out that a similar type of attack can not only be launched against mobile apps, it can attack from many channels, including 2D barcode, Wi-Fi scanning, Bluetooth pairing, MP3 songs, MP4 videos, SMS messages, NFC tags, Contact list, etc. As long as an HTML5-based app displays information obtained from outside or from anohter app, it may be a potential victim.

Story Source:

The above story is based on materials provided by Syracuse University. Note: Materials may be edited for content and length.


See what a child will look like using automated age-progression software

It's a guessing game parents like to ponder: What will my child look like when she grows up? A computer could now answer the question in less than a minute.

University of Washington researchers have developed software that automatically generates images of a young child's face as it ages through a lifetime. The technique is the first fully automated approach for aging babies to adults that works with variable lighting, expressions and poses.

"Aging photos of very young children from a single photo is considered the most difficult of all scenarios, so we wanted to focus specifically on this very challenging case," said Ira Kemelmacher-Shlizerman, a UW assistant professor of computer science and engineering. "We took photos of children in completely unrestrained conditions and found that our method works remarkably well."

The research team has posted a paper on the new technique and will present its findings at the June IEEE Computer Vision and Pattern Recognition conference in Columbus, Ohio.

The shape and appearance of a baby's face -- and variety of expressions -- often change drastically by adulthood, making it hard to model and predict that change. This technique leverages the average of thousands of faces of the same age and gender, then calculates the visual changes between groups as they age to apply those changes to a new person's face.

More specifically, the software determines the average pixel arrangement from thousands of random Internet photos of faces in different age and gender brackets. An algorithm then finds correspondences between the averages from each bracket and calculates the average change in facial shape and appearance between ages. These changes are then applied to a new child's photo to predict how she or he will appear for any subsequent age up to 80.

The researchers tested their rendered images against those of 82 actual people photographed over a span of years. In an experiment asking random users to identify the correct aged photo for each example, they found that users picked the automatically rendered photos about as often as the real-life ones.

"Our extensive user studies demonstrated age progression results that are so convincing that people can't distinguish them from reality," said co-author Steven Seitz, a UW professor of computer science and engineering. "When shown images of an age-progressed child photo and a photo of the same person as an adult, people are unable to reliably identify which one is the real photo."

Real-life photos of children are difficult to age-progress, partly due to variable lighting, shadows, funny expressions and even milk moustaches. To compensate for these effects, the algorithm first automatically corrects for tilted faces, turned heads and inconsistent lighting, then applies the computed shape and appearance changes to the new child's face.

Perhaps the most common application of age progression work is for rendering older versions of missing children. These renderings usually are created manually by an artist who uses photos of the child as well as family members, and editing software to account for common changes to a child's face as it ages, including vertical stretching, wrinkles and a longer nose.

But this process takes time, and it's significantly harder to produce an accurate image for children younger than age 5, when facial features more closely resemble that of a baby.

The automatic age-progression software can run on a standard computer and takes about 30 seconds to generate results for one face. While this method considered gender and age, the research team that also includes UW doctoral student Supasorn Suwajanakorn hopes to incorporate other identifiers such as ethnicity, and cosmetic factors such as hair whitening and wrinkles to build a robust enough method for representing every human face.

"I'm really interested in trying to find some representation of everyone in the world by leveraging the massive amounts of captured face photos," Kemelmacher-Shlizerman said. "The aging process is one of many dimensions to consider."

This research was funded by Google and Intel Corp.

Story Source:

The above story is based on materials provided by University of Washington. Note: Materials may be edited for content and length.


Google Glass puts the focus on Parkinson's

The next generation of wearable computing is being trialled for the first time to evaluate its potential to support people with Parkinson's.

Experts at Newcastle University are investigating Google Glass as an assistive aid to help people with Parkinson's retain their independence for longer.

Glass is a wearable computer being developed by Google. Likened to the kind of technology fictionalized in the Hollywood Blockbuster Minority Report, at first glance Glass appears to be no more than a pair of designer glasses. But the system works like a hands-free smartphone, displaying information on the lens of the Glass. The technology is voice-operated and linked to the internet.

Not currently available outside the US, the five pairs of Glass at Newcastle University were donated by Google to allow researchers to test how they could be used to support people with long-term conditions.

Initial studies by the team -- who are based in the University's Digital Interaction Group in Culture Lab, part of the School of Computing Science -- have focussed on the acceptability of Glass. They have been working with a group of Parkinson's volunteers aged between 46-70 years.

Now they are working on the next stage of the project, using the technology to provide discreet prompts linked to key behaviours typical of Parkinson's, such as reminding the individual to speak up or to swallow to prevent drooling. Glass can also be used as a personal reminder for things such as medication and appointments.

The team will also be exploring how the motion sensors in Glass can be used to support people with 'freezing', a behaviour caused by motor blocking a common symptom of Parkinson's.

Led by Dr John Vines, PhD student Roisin McNaney and Dr Ivan Poliakov, this is the first UK trial of Glass. Presenting their initial findings later this month at the ACM Human Factors in Computing Systems (CHI) 2014 conference in Toronto, Canada, the team will show how emerging technologies can potentially be used to support people with progressive diseases such as Parkinson's and dementia.

"Glass opens up a new space for exploring the design and development of wearable systems," explains Dr Vines, who along with colleagues in Culture Lab is working on a number of projects investigating how technology can be used to support people in everyday life.

"It is very early days -- Glass is such new technology we are still learning how it might be used but the beauty of this research project is we are designing the apps and systems for Glass in collaboration with the users so the resulting applications should exactly meet their needs.

"What was really encouraging from this early study was how well our volunteers took to the wearable technology and the fact that they could see the potential in it."

Parkinson's disease is a progressive neurological condition affecting up to 10 million people worldwide, with onset generally in those over 50.

The condition manifests itself in motor symptoms including rigidity, tremor and 'bradykinesia' or slowness of movement.

These affect balance, gait, arm and facial movements. Motor blocks commonly affect people's legs during walking causing them to 'freeze'; speech and voice are typically affected in terms of volume and clarity and the automatic swallowing mechanism is blocked so individuals often drool.

Aside from the physical signs, there are a myriad of emotional and social factors relating to loss of independence, social confidence, embarrassment and stigma.

Roisin, a speech and language therapist whose PhD has primarily focussed on the use of external cues as behavioural prompts, says one of the big challenges is finding technology that is not only useful to people but is also discreet.

"People with Parkinson's are already coping with so much and one of the main causes of social isolation is the stigma around behaviours such as drooling and tremor which they have no control over," explains Roisin, who is using discreet prompts to remind people with Parkinson's to swallow to prevent drooling, a common side effect.

"The last thing we want is a system of cueing which is so obvious it adds to people's overall embarrassment. Wearable computing is still quite novel but as more people buy into the technology and start to wear it out and about for leisure then systems such as Glass offer us a real opportunity for the long-term treatment of progressive conditions."

Dr Vines adds: "Technology has the potential to play a central role in the development and improvement of people's lives. The challenge is understanding everyone's different needs and tailoring that technology so that it makes a real impact on society."

Claire Bale, Research Communications Manager at Parkinson's UK, said: "This new study looking into Google Glass is an exciting example of how new technologies could be used to improve the lives of people living with Parkinson's by tackling a wide variety of problems -- from freezing to remembering to take their medication on time.

"But to really make the most of the potential of new technologies it's essential that researchers work in partnership with the real experts in the condition -- people living with Parkinson's.

"Only people with the condition can tell us if these new approaches will genuinely improve their lives in meaningful and realistic ways."

Case studies:

Partners Lynn Tearse, 46, and Ken Booth, 56, from County Durham, were some of the first volunteers to try out Glass as part of the Newcastle University trial.

Ken, who was first diagnosed with Parkinson's in 1991, underwent Deep Brain Stimulation last year in a bid to relieve some of the side effects of the condition.

"The drugs just weren't working for me anymore," explains the former salesman.

"I'd been offered the operation five years ago but I was too scared and at the time I was managing with the medication. But by last year the tremor had got so bad I couldn't carry on.

"The difference is incredible. It hasn't stopped the episodes completely and I still have to take the medication but it's helping to control the symptoms so I can live my life."

Trialling Glass for a week, Ken says both he and Lynn are complete converts.

"They're just fantastic. The potential for someone with Parkinson's is endless. For me the biggest benefit was confidence. When you freeze your legs stop working but your body carries on moving forward and it's easy to fall.

"Because Glass is connected to the internet you can link it to computers and mobile phones. So if you're alone you just have to look through the Glass and carers, friends or relatives will be able to see exactly where you are and come and get you. Or you just tell it to call someone and it rings them."

Lynn, a retired teacher who was diagnosed with Parkinson's in 2008, adds: "People would probably say you can do all these things on a smartphone but actually, with Parkinson's, negotiating a touch screen is really difficult.

"It's not just the tremor. During a 'down time' when the medication is starting to wear off and you're waiting for the next lot to kick in it can be like trying to do everything wearing a pair of boxing gloves. Your movements are very slow and your body won't do what you want it to."

Lynn says Glass could also be hugely helpful to unlock the brain when is 'freezes'.

"No-one really understands why it happens," explains Lynn, "but it happens when the flat surface in front of you breaks up or the space in front of you narrows such as a doorway. Revolving doors are particularly bad.

"Your legs gradually freeze up and the difficulty is getting started again. The brain seems to need a point beyond the blockage to fix on and people use different things -- Ken will kick the end of his walking stick out in front of him but many people use laser pens to create a virtual line beyond the barrier. This is where Glass could really make a difference."

Using it as a medication reminder is another of the applications the Newcastle University team is looking at.

"The drugs don't cure Parkinson's, they control it so it's really important to take the medication on time," explains Ken.

"I was taking two or three different drugs every two hours, different combinations at different times of the day; some with water, some with food, the instructions are endless. Having a reminder that is literally in your face wherever you are and whatever you are doing would really help."

Lynn adds: "Parkinson's can be very isolating. Ken and I work together -- we went away last month and I learnt to ski -- but the Parkinson's symptoms and the drug side effects can be frightening and often embarrassing and not always well understood.

"Any technology which promotes confidence and helps people take better control of their condition and their life should be welcomed."


Flaw in 'secure' cloud storage could put privacy at risk

Johns Hopkins computer scientists have found a flaw in the way that secure cloud storage companies protect their customers' data. The scientists say this weakness jeopardizes the privacy protection these digital warehouses claim to offer. Whenever customers share their confidential files with a trusted friend or colleague, the researchers say, the storage provider could exploit the security flaw to secretly view this private data.

The lead author of the new article is Duane C. Wilson, a doctoral student in the Department of Computer Science in the university's Whiting School of Engineering. The senior author is his faculty adviser, Giuseppe Ateniese, an associate professor in the department. Both are affiliated with the Johns Hopkins University Information Security Institute.

Their research focused on the secure cloud storage providers that are increasingly being used by businesses and others to house or back up sensitive information about intellectual property, finances, employees and customers. These storage providers claim to offer "zero-knowledge environments," meaning that their employees cannot see or access the clients' data. These storage businesses typically assert that this confidentiality is guaranteed because the information is encrypted before it is uploaded for cloud storage.

But the Johns Hopkins team found that complete privacy could not be guaranteed by these vendors. "Our research shows that as long as the data is not shared with others, its confidentiality will be preserved, as the providers claim," Wilson said. "However, whenever data is shared with another recipient through the cloud storage service, the providers are able to access their customers' files and other data."

The problem, Wilson said, is that privacy during file-sharing is normally preserved by the use of a trusted third party, a technological "middle-man" who verifies the identify of the users who wish to share files. When this authentication process is finished, this third party issues "keys" that can unscramble and later re-encode the data to restore its confidentiality.

"In the secure cloud storage providers we examined," Wilson said, "the storage businesses were each operating as their own 'trusted third party,' meaning they could easily issue fake identity credentials to people using the service. The storage businesses could use a phony 'key' to decrypt and view the private information, then re-encrypt it before sending it on to its intended recipient."

Wilson added, "As a result, whenever data is shared with another user or group of users, the storage service could perform a man-in-the-middle attack by pretending to be another user or group member. This would all happen without alerting the customers, who incorrectly believe that the cloud storage provider cannot see or access their data."

These storage services generally do not share the details of how their technology works, so Wilson and Ateniese substantiated the security flaw by using a combination of reverse engineering and network traffic analysis to study the type of communication that occurs between a secure cloud storage provider and its customers.

The researchers pointed out that their study focused only on three storage providers that claimed their customers' data would remain completely confidential. Other file-sharing services, such as Dropbox and Google Drive, make no pledge of privacy. Instead, they say that after a user's data is uploaded, it is encrypted with keys that are owned by the file-sharing service.

To solve the security flaw, the researchers recommend that the arrangements between customers and secure storage providers be revised so that an independent third party serves as the file-sharing "middle-man," instead of the storage company itself.

"Although we have no evidence that any secure cloud storage provider is accessing their customers' private information, we wanted to get the word out that this could easily occur," said Ateniese, who supervised the research. "It's like discovering that your neighbors left their door unlocked. Maybe no one has stolen anything from the house yet, but don't you think they'd like to know that it would be simple for thieves to get inside?"

Story Source:

The above story is based on materials provided by Johns Hopkins University. Note: Materials may be edited for content and length.


Researchers propose network-based evaluation tool to assess relief operations feasibility

The United Nations Office for Disaster Risk Reduction reported that disasters have affected around 2.9 billion people worldwide from 2000-2012- killing more than a million, and damaging around 1.7 trillion US dollars in estimates. Moreover, natural disasters and their damages have been documented to occur with increasing intensity. Given the staggering numbers, effective disaster preparedness and relief response plans is compelling, especially considering the fact that natural disasters are usually unpredictable and damage cannot be avoided.

Implementing a speedy and effectual outreach post-disaster is a nontrivial challenge "due to potential infrastructural changes such as destruction of road systems that make some highways impassable, and damage to the facilities and/or warehouses that serve as storage for relief supplies." A Singapore-based team of scientists from the Institute of High Performance Computing, A*STAR and The Logistics Institute-Asia Pacific has presented a model that looks into the logistics of disaster relief using open data and tools and measures developed in the field of network science. The work was recently published in the International Journal of Modern Physics C.

Based on OpenStreetMap- a collaborative project that provides open geodata to the world, the team reported a procedure that automatically converts a road map system into a road network of nodes and edges. It then utilizes contemporary tools in complex networks to assess several dynamics on the system, particularly, the flow of goods and other relief efforts, and quantify the reachability of critical loci within a geographic area where a disaster has struck. The proposed model is highly-flexible- allowing for inclusion of damage information, such as information coming from the Humanitarian OpenStreetMap team, in the analyses. The procedure developed also enables evaluation of the various effects of a range of possible hypothetical infrastructure destruction scenarios even before a disaster strikes a region-this was shown to be crucial in formulating contingency plans for the logistics of disaster response and relief operations.

To illustrate the utility of the methodology developed, the team considered the roadmap of the city of Tacloban in the central Philippines that was hit by Typhoon Haiyan, which claimed 6000 lives and displaced around 4.1 million more. Among others, the work quantifies the extent at which the inherent structure of the road network plays a role in facilitating, or hindering, landbound relief efforts, especially in the critical hours and days immediately following a disaster event. It also discusses the inaccuracy of assuming that road networks follow a structure similar to the more commonly studied scale-free, random, and/or grid (regular) network configurations.

This research was supported by Singapore A*STAR Complex Systems Programme research grant (# 1224504056 for EFL, CM) A*STAR SERC research grant (#1121790043 for JFV, XF, RG, RdS).

Story Source:

The above story is based on materials provided by World Scientific. Note: Materials may be edited for content and length.


Page 10 of 28

Our Partners