How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!
The beginning of a new year is often a time to look ahead, and Advanced Imaging is no different. A trio of academics, who work in various imaging disciplines, were asked to look ahead. They were not asked for pie-in-the-sky predictions, or way-down-the-road prognostications, but rather they were asked about issues that affect their work -- and could affect the imaging industry -- in the time period just ahead.
WHAT'S DOABLE NEXT?
Advanced Imaging: What is your area of special interest and research and what are the key areas for further technology development and extensions of capability within it in the next three to five years?
Stefan Robila: My interests fall within general imaging and optics education, and the development of activities that would increase the understanding of these topics in the community. Technology advances in sensor and lens technology have rendered in-house applications relatively inexpensive, and have increased the general public interest and understanding on imaging; this trend will continue and will extend in the direction of image databasing and organization. My research deals with the development of efficient processing techniques for hyperspectral data. Here, the key development is the reduction in cost of the hyperspectral sensors for both the visible and infrared. An extension of the applicability of hyperspectral technology to other areas, including quality control and medical and chemical imaging, are expected to only increase its attractivity. Finally, the technology will expand from regular tunable filters to hyperspectral tunable sensors where the band sizes could be dynamically set.
Roger Easton: My work in the last several years has focused on applying digital imaging technologies to enhance or recover handwritten text from historical manuscripts, including the Archimedes palimpsest, a 10th-century manuscript that contains the oldest known copies of seven treatises. This effort has used scientific and professional-quality digital cameras that have become available in the last few years. It seems likely that advances in acquisition technology will continue so that we expect to acquire digital imagery with better spatial and spectral resolution over the visible spectrum. I expect that the technology to collect images easily, and inexpensively, in the near ultraviolet region and farther into the near infrared region of the spectrum to become more available, and at reasonable cost. I expect that imaging technologies and capabilities to advance in other spectrum regimes, such as x-ray fluorescence and confocal microscopy. These also are likely to help in this particular application.
John Kerekes: My work has been in the area of earth remote sensing -- investigating the optimal design of sensing systems and their analysis methods for a variety of environmental and military applications. This area was initiated with the passive optical and microwave sensors of the 1960s and 1970s and grew into active optical (lidar) and microwave (synthetic aperture radar and interferometry). Increasing spatial and spectral resolution have been continuing themes with passive optical hyperspectral imaging (HSI), which are hundreds of narrow, contiguous spectral channels, emerging in the 1990s and now commercially available. Many of the HSI systems have had to rely on two or more detector array technologies to cover the full solar spectrum, 0.4 to 2.5 microns, and this has led to band-to-band spatial registration problems. The development of extended InGaAs or similar detector arrays to cover 0.4 to 1.8 microns and beyond will allow the collection of fully co-registered images and lead to more operational uses. Other areas of growth are expected in the use of uncooled bolometers for longwave infrared (7 to 14 microns) operation, and the use of multispectral lidars for active sensing of spectral phenomenology.