Wednesday, November 10, 2010


0 comentarios

Scientists at The University of Manchester have developed software for mobile phones that can track your facial features in real-time. Eventually it will be able to tell who the user is, where they are looking and even how they are feeling.

The method is believed to be unrivalled for speed and accuracy and could lead to facial recognition replacing passwords and PIN numbers to log into internet sites from a mobile phone.

"Existing mobile face trackers give only an approximate position and scale of the face," said Dr Phil Tresadern, lead researcher on the project. "Our model runs in real-time and accurately tracks a number of landmarks on and around the face such as the eyes, nose, mouth and jaw line.

"A mobile phone with a camera on the front captures a video of your face and tracks twenty-two facial features. This can make face recognition more accurate, and has great potential for novel ways of interacting with your phone."

Originally intended as part of a face- and voice-verification system for access to mobile internet applications such as email, social networking and online banking, alternative uses for the device could include fun applications that, for instance, attach virtual objects to the user's face as they move around.

"At this stage, we're particularly interested in demonstrating uses for the face-tracking part of the technology, which is the area The University of Manchester is involved in," said Dr Tresadern, who is based in Manchester's School of Cancer and Enabling Sciences. "It is very fast and I can't find anything that can rival it on a mobile phone."

Face verification is already used in laptops, webcams and the Xbox 360 Kinect but this is the first time the technology is being used with such sophistication in mobile devices such as smartphones.

The new software, built on 20 years of research at the University, has been demonstrated on a Nokia N900 for the EU-funded "Mobile Biometrics" (MoBio) project.

(Photo: U. Manchester)

The University of Manchester


0 comentarios

In a galaxy far away, an exceptionally massive black hole is traveling around a massive star in an unusually tight orbit. Also odd, the star is not as bright as it should be.

Astronomers have puzzled over this X-ray binary system, named M33 X-7, but no one could explain all of its features. Now a Northwestern University research team has.

The researchers have produced a model of the system’s evolutionary history and formation that explains all of the system’s observational characteristics: the tight orbit, the large masses of the star and black hole, the X-ray luminosity of the black hole and why its companion star is less luminous than one would expect, given its mass.

The evolutionary model is published by the journal Nature. The research will improve astronomers’ understanding of how massive stars evolve and interact with their host environment as well as shed light on the physics behind the process of black hole formation.

“We were attracted to this system because it has one of the most massive black holes to have formed from a star, and yet the rest of its characteristics, especially the mass of its companion star and its orbit, did not make any sense from an evolutionary point of view,” said Vicky Kalogera, professor of physics and astronomy in the Weinberg College of Arts and Sciences.

M33 X-7 is one of the few known X-ray binary systems containing a black hole outside our galaxy, and its star is the most massive star ever discovered in such a system.

The researchers’ evolutionary model of M33 X-7 starts with two stars in a binary system (or in orbit one around the other). One star is 100 solar masses (100 times the sun’s mass), and the other is 30 solar masses. The stars are in a close orbit, with the larger star growing faster until it nearly envelops the other. The initially smaller star then gains material from its companion, while the initially larger and more massive star collapses into a black hole at the end of its nuclear-burning lifetime. The orbit becomes even tighter.

The star, which is now 70 solar masses, is not as luminous as stars of similar mass partially because of the way it gained its mass and partially because of the inclination of the system with respect to us. On one hand, the star accreted matter so quickly from its interaction with the other star (now a black hole) that it could not adjust fast enough to its new, greater mass. Therefore, the star does not burn as bright as an undisturbed star of this greater mass would. On the other hand, the star is deformed due to the close presence of the massive black hole, and the star’s temperature and luminosity are not uniform across the surface. This effect, combined with the inclination of the system with respect to our line of sight, means we are looking at the star’s dimmest equatorial regions.

And now the massive black hole is growing even larger. The companion star is feeding matter, via a stellar wind, to the black hole. In the process X-rays are emitted, allowing astronomers to observe the black hole.

“Solitary black holes are very difficult to observe, but X-ray binary systems, such as M33 X-7, make black holes visible to us,” said Francesca Valsecchi, a doctoral student in Kalogera’s research group and lead author of the paper. “These systems provide a unique physical laboratory for the study of massive compact objects.”

Valsecchi, Kalogera and colleagues performed detailed binary system evolution calculations to explore possible evolutionary tracks. They used information known about the physics of binary stellar interactions and black hole formation processes.

In their initial work, they ran more than 200,000 sequences on a high-performance computing cluster, which took a couple of months. The researchers then examined a number of these sequences in further detail and were able to identify the final model, consistent with all observational characteristics of M33 X-7.

M33 X-7 is an X-ray binary system discovered in 2007 in the Messier 33 galaxy, known as M33. (An X-ray binary system is a class of binary stars luminous in X-rays.) The Messier 33 galaxy, slightly farther away from us than the Andromeda galaxy, is among the farthest permanent objects that can be viewed with the naked eye.

(Photo: Matthew McCrory, Francesca Valsecchi and Vicky Kalogera, Northwestern University)

Northwestern University


0 comentarios

The human hand is an amazing machine that can pick up, move and place objects easily, but for a robot, this “gripping” is a vexing challenge. Opting for simple elegance, a team of researchers from Cornell University, the University of Chicago and iRobot have created a versatile gripper that uses the jamming of particulate material inside an elastic bag to hold on to objects, bypassing traditional designs based around the human hand and fingers.

They call it a universal gripper, as it conforms to arbitrary objects it’s grabbing, rather than being designed specifically for particular objects, says Cornell researcher Hod Lipson, associate professor of mechanical engineering and computer science. The research is a collaboration between the groups of Lipson, Heinrich Jaeger at the University of Chicago and Chris Jones at iRobot Corporation.

In essence, the gripper uses the same phenomenon that makes a vacuum–packed bag of ground coffee so firm; in fact, ground coffee worked very well in the device. But the researchers found a new use for this everyday phenomenon: They placed the elastic bag against a surface and then removed the air from the bag, solidifying the ground coffee inside and forming a tight grip.

Lipson noted that the universality of the gripper makes future applications seemingly limitless — including use by the military to dismantle explosive devices or to move potentially dangerous objects, industrial applications of robotic arms in factories, using the gripper on the feet of a robot that could walk on walls, or making multi–purpose prosthetic limbs.

Here’s how it works: Particulate materials are large aggregates of individually solid particles. A special feature of this class of materials, which includes many familiar commodities like sand, grain or ground coffee, is that they can undergo a so–called jamming transition, so they behave more like a solid than a fluid. This happens when particles lose their ability to move past each other.

Evacuating the air from a bag of espresso grounds, for example, will move the particles closer until the jamming transition is crossed, at which point the material becomes rigid; releasing the vacuum by unsealing the package unjams the grounds and they can flow.

In the gripper, a latex balloon filled with particulate material is attached to a robotic arm. The balloon presses down in the unjammed state, molds around the desired object and then a vacuum sucks the air out of the balloon, solidifying its grip. When the vacuum is released, the balloon becomes soft again, and the gripper lets go.

“The concept of a jamming transition was developed to provide a unified framework for understanding and predicting behavior in a wide range of disordered, amorphous materials that all can be driven into a ‘glassy’ state, where they respond like a solid yet structurally resemble a liquid. This includes many liquids, colloids, emulsions or foams, and also particulate matter consisting of macroscopic grains,” Jaeger said. “What is particularly neat with the gripper is that here we have a case where a new concept in basic science provided a fresh perspective in a very different area, robotics, and then opened the door to applications none of us had originally thought about.”

The team split the work on the project. Eric Brown, a postdoctoral researcher, and Nick Rodenberg, a physics undergraduate, worked with Jaeger on characterizing the basic mechanisms that enable the gripping action. Prototypes of the gripper were built and tested by Lipson and graduate student John Amend as well as at iRobot.

As for the right particulate material, anything that can jam will do in principle, and in early prototypes rice, couscous and even ground–up tires were tried. For the final prototypes, the team settled on coffee because it’s light but also jams well. However, there is much room for optimization. What sets the jamming–based gripper apart is that it has performed well with almost any object, including a raw egg or a coin — both notoriously difficult for traditional robotic grippers.

(Photo: Lloyd DeGrane)

University of Chicago




Selected Science News. Copyright 2008 All Rights Reserved Revolution Two Church theme by Brian Gardner Converted into Blogger Template by Bloganol dot com