Kodak boosts digital camera sensitivity

TalkBack 4 of 10:

Next »

« Previous

• Thread View
• Flat View

Sorry, but that's even worse

I can generate white from just and equal amount of two saturated colors, namely pure cyan light at 485-488 (argon ion laser) and pure yellow-orange light (filtered sodium lamp or dye laser). Both of these will not be completely detected using just red and blue filtered pixels. Your approach will infer the the residual value in the luminance channel is from green - but there is no green component to this optical field (i.e. no optical radiation in the 500 to 530 nm region. That leads to ambiguity and frustrates accurate color measurements.

I used to be involved in design laser projection systems so the color generation issues where very important considerations. this is the inverse problem of color measurement, which we also needed to in order to test our displays (these were HD computer graphics projections on 12' tall 18' to 20 ' wide displays in military command centers in 1988). The CIE color chart* shows how pure saturated color (a photon has a distinct frequency and wavelength) combine to make the color map that the "standard observer" can perceive. Drawing a line between the color regions represented by the color centers (either the detector filter response regions or the phosphor/LED/laser emission regions) gives the full range of the set of color values that can be measured or displayed. Replacing any component of RGB with a region near the center of the map at white for a "luminance" channel only decrease the area on the color map that can created or represented. Blue was actually the best choice because it's represents about the least loss of area.

Kodak actually did their homework when choosing the filter pattern and pixel color sets. I made it through college at the University of Rochester (with a degree in Optics from the College of Engineering) because of a generous Kodak scholarship and I even worked for them for a very brief time. I did sit in on technical presentations (at meetings for the Optical Society of America) regarding their studies about psycho-physics, resolution perception and color perception and how they choose the requirements for their film and camera offerings long before digital photography. Kodak has some bright people; give 'em a little credit.

*1936 CIE Color Chart -for an example see:
http://www.hdtvexpert.com/pages_b/calibration.html

Posted by: jacarter3   Posted on: 06/15/07 You are currently: a Guest | Members login | Terms of Use

Alert moderator to an offensive message

Subscribe to this discussion via Email or RSS

SponsoredWhite Papers, Webcasts, and Downloads

• Three Steps You Need to Know to Stop Data Loss Varonis Sensitive data exposed to misuse or loss... it is the stuff of nightmares ... Download Now
• Reducing Server Total Cost of Ownership with VMware Virtualization Software VMware VMware virtualization enables customers to reduce their server TCO and ... Download Now
• The True Costs of Virtual Server Solutions VMware In an economic environment that is repeatedly heralding the message "do ... Download Now