Latest question:
How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!


By Larry Adams
Editor
Bandwidth is increasing. Storage is improving. Raw data can now be streamed into storage at 200 megabytes per second and higher. Data from dozens of MRI or CT slices, each less than a millimeter apart, can be crafted into a 3-D image.
Surfaces of parts as well as land masses can be mapped. Correlations and causal effects can be rooted out with pattern and object recognition software.
For some applications, it is a matter of getting the data to storage and allowing other software and other technologies to manipulate the data. For some applications, it is a matter of using the technology to analyze the images in real time and improve processes -- whether in the plant, the operating room or the battlefield.
Storing the data for later analysis is a common, and important, use of imaging data. It does however, bring with it indexing issues. It is not uncommon, experts say, to find images stored on a variety of mediums, in a variety of modalities.
"Finding images once they have been stored has always been a problem for some people," says Ian Lee, vice president of engineering and support for AccuSoft Corp. (Northborough, MA). "The trick is to find them and use them."