your Windows® embedded community
Other members are said to include Analog Devices, Apical, Avnet Electronics Marketing, CEVA, CogniVue, Freescale, National Instruments, Nvidia, Tokyo Electron Device, MathWorks, Ximea, XMOS, and Texas Instruments.
For now, the group is focused on its new open-access website (see link farther below). The site is intended to serve as a source of practical information for embedded vision system design. In the future, the site will host newsletters, educational webinars, industry reports, and the publication of technology standards.
The embedded vision field spans processors, image sensors, lenses, lighting, networking, and development tools that apply intelligence to cameras and other imaging devices, says the EVA. These components are said to be used in applications ranging from surveillance to safety to robotics in industries including industrial, medical, automotive, security, and consumer home automation and entertainment.
Kinect shows the way
In his founding statement, Jeff Bier, president of BDTI, singles out the Microsoft Kinect as an inspiration. The Kinect game controller add-on for the Xbox 360 has quickly become the fastest-selling consumer electronics device in history, shipping more than 10 million units in five months, according to Blier.
"Adding computer vision to embedded systems creates phenomenal new products, markets, and opportunities," writes Bier. "From automobiles that prevent accidents to security cameras that prevent crimes, embedded vision will proliferate across a multitude of markets."
Intelligent video surveillance is a particularly thriving segment of embedded vision. EVA member Texas Instruments (TI) aims many of its DaVinci ARM/DSP hybrid system on chips (SoCs) at intelligent surveillance. Its recent TMS320DMVA1 SoC, which is backward-compatible with DaVinci SoCs, combines an ARM9 core, a new Vision analytics co-processor, and a codec co-processor. The SoC is offered in a DMVA1 IP camera reference design (pictured at left), says TI.
The DMVA1's imaging technology is said to provide video stabilization, face detection, noise filtering, auto white balance, auto focus, auto exposure, and edge enhancement. Developers can also integrate TI's TMS320DM6435 DaVinci DSP with the design to "achieve advanced analytics capabilities," according to TI.
Machine vision for industrial applications is another hot area. EVA member Ximea recently announced a series of Windows -compatible "smart cameras" for machine vision that include Intel Atom processors and color image sensors ranging from WVGA (752 x 480 pixels) to five megapixels. The Curerra-R devices (pictured at right) offer features including 1GB to 4GB of flash storage, microSD slots, VGA and Ethernet ports, and isolated digital I/O.
Further information
More information on the Embedded Vision Alliance may be found at the EVA website.