| Aresa Biodetection A Danish biotech company has developed a genetically modified flower that could help detect land mines and it hopes to have a prototype ready for use within a few years. The genetically modified weed has been coded to change color when its roots come in contact with nitrogen-dioxide (NO2) evaporating from explosives buried in soil. | ![]() |
|
![]() |
Within three to six weeks from being sowed over land mine infested areas the small plant, a Thale Cress, will turn a warning red whenever close to a land mine. Aresa's invention, based on research at the Institute of Molecular Biology at Copenhagen University, uses a plant's normal reaction to turn red or brown when subjected to stressful conditions such as cold or drought, |
|
| but has genetically coded it to react only to nitrogen-dioxide. Aresa has succeeded in growing the genetically modified plant and hopes to launch restricted tests this year and to apply for field tests in Denmark and abroad after that. Aresa, a private company, is currently seeking strategic partners to speed up its development, both through financial and intellectual support, and has filed for intellectual property protection under the Patent Cooperation Treaty (PCT). www.aresa.dk | ||
![]() |
Sanyo
TS41 Mobile Phone Japanese
electronics giant Sanyo has developed a truly silent mobile phone. Instead
of a loudspeaker, the handset, the TS41, uses the user's own skull to
transmit the incoming call. Using
bone to conduct sound waves isn't new, but this is the first time we've
heard the technique being used for a mobile phone. Sanyo claims it's a
world's first. |
|
Sanyo's handset will be offered by Japanese network provider Tu-Ka at the end of this month. The clamshell handset weighs 98g. Inside is a 2.1in, 176 x 132 16-bit colour LCD. On the back is a 1in, 64 x 64 16-bit colour screen that display the time. It can play 40-voice polyphonic ringtones. The TS41 contains a conventional speaker. Using it, the phone can provide 140 minutes' worth of talktime. With the sub-sonic speaker, it can only manage 130 minutes. Standby time is 400 minutes. |
|
| Dust Inc. Each Smart Dust sensor (called a "mote") is about the size of a character in this sentence. It contains a device that measures such things as temperature, motion and light intensity, and it sports a miniature antenna and a radio with a range of 10 to 50 meters. Another mote inside that range picks up the first one's signal and grabs its data, adding readings of its own and passing the package along to a third mote. The data eventually move to a base station 5 inches tall, from where the info is sent to a PC or onto a wireless network. In a test for the military, eight sensors (with clocks, motion detectors and electronic compasses) were dropped from an airplane. | ![]() |
| They self-organized and successfully determined the direction, speed and size of a series of armored vehicles. Other possible uses include scattering motes across a forest floor so firefighters can spot a fire when it has burned just a few trees. You could leave your friends personal notes or restaurant reviews by dropping message chips that come alive when your friends' sensors come into range. www.dust-inc.com |
![]() |
IO2Technology
Heliodisplay
The Heliodisplay modifies the properties of
air within a localized environment. Air comes into the device, is ejected
and illuminated to produce the image. It is not holography. There is no
harmful gas or liquid. Nothing needs to be refilled. There is no
overall change in environmental properties of the room in which the device
operates. The input for the Ģimageģ is just the air in the room.
The images are easily viewed in an office environment. Like any computer monitor or TV, they appear brighter the lower the ambient light. Viewing in direct sun light is almost impossible in our current prototype. The image is planar (2D), not volumetric (3D). It appears 3D when viewed more than a few feet away, however, because there is no physical depth reference. Thus, like any computer monitor, it can project simulated 3D. Images can be seen up to 75°s off aspect for a total viewing area of 150°s. |
|
The images of the current prototypes are invisible (transparent) from behind. In future versions, the behind image will be togglable between invisible, same as front image or different from front image. Viewing requires no special glasses or background/foreground screening. In the current prototypes, the images float above the device. In future versions, the Heliodisplay will be rotatable, so that images can be projected to the side or even down. Furthermore, the device can be interactive, like a virtual touchscreen. A hand or finger can act as a mouse. No special glove or pointing device is required. www.io2technology.com |
![]() |
Toyota Prius - Toyota plans to release a car next month in Japan that parallel parks itself, an Australian newspaper reported. Toyota has rigged its hybrid Prius model with special gear to accomplish the feat. It uses a rear-mounted camera and a computer program to perform the task, consistently making a perfect reverse park without the driver touching the wheel, said Toby Hagon, an editor of The Age, who test-drove the vehicle. "This system assures you don't hit other cars," he said. Hagon, along with other journalists from around the globe, test-drove the car in Tokyo three weeks ago. Toyota declined to comment about the car or its capabilities. |
|
Nonetheless, automotive experts expect the car to make splashy headlines when Toyota officially unveils it to the public next month. It will initially be offered as a high-end feature for the $20,000 Prius, a model that uses an electric motor to assist a gasoline engine to conserve fuel. Hagon said the self-parking option is the first step toward self-driving cars. |
![]() |
| Self-driving systems have been in research
laboratories for years. Before details leaked out about Toyota's self-park
system, scientists at Roke Manor Research
in Hampshire, England, developed a network of radars and miniature cameras
that monitored a car and let it measure a parking space to let drivers
know whether they could park there. While drivers still had to turn the
steering wheel and hit the gas pedal to park, the system forced the car to
stop if an unexpected obstacle got in the way, or if the car got too close
to another vehicle. Roke showed off the system two years ago, but it isn't
in automobiles yet.
Bryan Rickett, who heads the radar group at Roke, says it will be another two years before cars come equipped with the capability to park themselves. He wasn't sure whether self-driving cars would hit the market anytime soon, although Hagon estimated they would in another "10 to 15 years." "We already have automatic cruise control," Rickett said. "Parking and collision-avoidance-type functions are coming along as well." Toyota's Prius may become the first car to exhibit those features. In Hagon's test drive, the car's built-in computer electronically measured a parking spot, marked a turn-in point with a virtual flagpole and then steered the car automatically into the space -- rear end first. The car monitored white lines, gutters and nearby objects, avoiding scratches and performing a "perfect" park every time, Hagon said. But Hagon could see a major pitfall for the Prius: its price tag. "I wouldn't buy the car," Hagon said. "You can get a better car for less money." Auto insurance companies probably won't like the system either. Just as airplanes carry pilots even though most of the plane's features are automated, a car will probably always require a driver for liability reasons, Rickett said. "I doubt insurance companies will let drivers say they aren't responsible for their vehicles," he said. wired..08/25/2003 |
| Roke Manor Research Video Motion Anomaly Detection Roke Manor Research, the electronics research and development business owned by Siemens, has developed and patented VMAD (Video Motion Anomaly Detection), a unique image processing technique designed to improve the efficiency of close circuit television (CCTV) by enabling them to 'think'. The intelligent scene analysis software employs a self-learning alerting mechanism to automatically detect abnormal behaviour. According to statistics, there are 25 million CCTV cameras in operation worldwide - 2.5 million in the UK alone - with the average citizen caught on CCTV cameras 300 times a day. |
|
|
The side effect of this popularity has been the costly human monitoring operations needed to process the video information. VMAD overcomes the inefficiencies of relying on an operator to identify suspicious behaviour from many different video feeds by 'learning' to analyse the live CCTV images for out of the ordinary events. The VMAD software can be configured to alert an operator's attention to an abnormal event that may be cause for concern, and highlight it on screen. An alternate configuration could be set to an indexing point on a CCTV recording, making it easier to analyse an event after it has happened. "The heightened pressure for ports and airports to increase security measures by investing in new technology to protect people and assets has resulted in security staff being overloaded with information from CCTV, radio and other systems", says Simon Reay, Technical Sales Manager at Roke Manor Research. "VMAD improves staff efficiency by enabling them to manage more video screens or releasing them to provide extra resource on the ground. The cost savings realised could open up the potential for monitoring operations that were until now economically infeasible due to manning overhead." Video cameras and motion detectors have been commonplace for a number of years. However, it has been reported that these surveillance systems have been of limited use due to high false alarm rates and an inability to extract key information from scenes of high activity. Unlike other emerging image intelligence software solutions, VMAD does not need programming to look for a multitude of specific objects or events. Instead, it learns for itself the normal scene behaviour. Therefore, it is simple to set up and maintain and can manage a large number of possible anomalous or unforeseen events. VMAD can simultaneously process hundreds of moving features, regardless of direction and speed of travel, and is robust to lighting changes. Consequently VMAD can detect an abnormal event even in the presence of other normal movement, improving on traditional video motion detection (VMD) solutions. For ports, airports and terminals, VMAD can be applied to a wide variety of applications including detecting bags that have been abandoned or fallen from a carousel, speeding vehicles, abnormal passenger behaviour such as people running through a terminal building, jumping over barriers, approaching stationary aircraft or entering restricted areas. The benefits of VMAD are most apparent in situations where accurate event detection is required for distributed installations with many cameras and a high number of possible abnormal events. In addition to transport hubs, VMAD's flexibility means it is well suited to a number of different sectors, including security agencies, power stations, shopping malls or car parks and corporations such as banks.
|
|