Developers build imaging sensors to systems
George Kotelly
Executive Editor
In developing image sensors, processors, and machine-vision systems, designers are cost-effectively harnessing off-the-shelf hardware and software-development tools. This approach has led to the design of sophisticated equipment such as CMOS-based devices that can capture and preprocess images and novel foveal image sensors. Two innovative developments in image-sensor technology involve a prosthetic retinal-implement device aimed at restoring vision to patients afflicted with retinal disease and a CCD used to remove image motion caused by atmospheric turbulence.
The retinal implant, says contributing editor John Haystead, aims at restoring vision to patients afflicted with retina pigmentosa, a degenerative disease where the photoreceptor cells of the retina fail to respond to light. To that end, researchers at North Carolina State University and Johns Hopkins University have completed a prototype implantable power and data receiver and a neural stimulator.
Other sensor architectures can remove image motion caused by atmospheric turbulence. To do so, says contributing editor Larry Curran, astronomers at the University of Hawaii and the Massachusetts Institute of Techology have designed a CCD that removes this type of image motion at lower cost than traditional tip/tilt mirrors.
Whereas researchers build new image sensors to perform specific image-processing functions, OEMs are using off-the-shelf components to build image-processing and machine vision systems for research and machine-vision applications. At the University of Wisconsin Applied Superconductivity Center, researchers are studying the magnetic fields of large-area superconductor materials using PC-based systems running off-the-shelf software. According to contributing editor Winn Hardin, this system converts optical images to mathematical representations of the material`s magnetic field.
Cameras, lighting, and processors networked over high-speed digital networks are also in use in machine-vision systems. As spotlighted by Larry Curran, such a system ensures that hypodermic needles are free of tiny metal burrs that would make their one-time use uncomfortable and checks whether the gradations on the syringes are clear.
In the Product Focus, editor-at-large Andy Wilson discusses how off-the-shelf processors, frame grabbers, image-processing boards, and real-time operating systems are used in imaging systems where real-time performance is mandatory.
