How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!
Some of the earliest entrants into the market were often key to the development of the product. Aware joined the OpenGL standards committee project at the 2001 Siggraph trade show and was a member, along with 11 other companies of the standards architectural review board (ARB) that approved the standard, which was released in September. The ARB included Apple, ATI, Dell Computer, Evans & Sutherland, Hewlett-Packard, IBM, Intel, Matrox, NVIDIA, SGI, and Sun.
Pegasus Imaging Corp. (Tampa, FL) began working on JPEG algorithms in 1991 and continues to support that protocol. Jack Berlin, president of Pegasus, says that the company began working on JPEG 2000 in 1999. "As far as technology goes," Berlin says, "JPEG 2000 continues to be our biggest investment as far as research and development dollars."
JPEG 2000 is an ISO standard for image compression based on wavelet technology (for more information on wavelets, see sidebar "Why Wavelets?" on page 26), and is considered by some to be a better compression technology than JPEG. A working group of the Digital Imaging and Communication in Medicine (DICOM) standard has approved JPEG 2000 use as a compression option on medical images. Both the lossless compression algorithm, an algorithm to archive and extract data without any data loss, and lossy compression algorithm, a compression technique that may lead to data loss, have been accepted for use. The International Standards Organization (ISO) has adopted JPEG2000.
"JPEG 2000 does compress better than JPEG," says Dave Benini, director of product marketing for Aware (Bedford, MA). "You can get better quality for a given file size across a spectrum of compression ratios."
Aware has added the compression standard into its JPEG 2000 Codec for Medical Applications. The software features lossless and lossy compression options. In the lossless mode, compression ratios are 2:1 to 3:1 and the archived image can be compressed and saved in lossless or near lossless form. Multiple smaller blocks of data can be extracted from these files, decoded and viewed by clients via a standard protocol network -- for example, a particular area of the image, a sub-resolution version of the image, or a lower quality version of the image can be extracted from a single JPEG 2000 image file.