How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!
By Lee J. Nelson
In recognition of our twentieth anniversary, we decided to take a look at five disciplines: medical imaging, research and development, remote sensing, imaging in academia and machine vision. We solicited opinions and observations from an august group of imaging veterans:
AI: What was the state of imaging technology in your discipline in 1986?
Mr. Biegel: Back then, medical imaging systems used proprietary data models, custom hardware and consisted of distinct dedicated applications—most of which drove diagnostic image display via “film on a light box.” Radiology work flows were built around analog output and storage. Sharing image data with colleagues was problematic. It required custom programming and often involved “cracking” the modality vendor’s singular data format. Any hope of seamlessly communicating metadata (demographics, measurements, imaging parameters) via electronic means was a pipe dream.
Mr. Gibson: Imaging research and development largely was limited to national labs and major corporations in 1986. The space requirements were significant. In addition to the relatively high cost, the availability of optics to perform complex imaging tasks was very limited.
Dr. Heric: In 1986, remote sensing systems were regularly restricted by the 30-meter spatial resolution of spaceborne collection systems. The SPOT sensor—with its 10-meter panchromatic resolution—was slowly coming on the scene and we were amazed at the improvement! In our wildest dreams, we never imagined that the high resolutions of today’s systems would be allowed. The policy path from 30-meter to sub-meter has been a long one, indeed.