The image sensor is a key piece of any embedded vision system design, of course. As such, many of you are probably already familiar with the term "Bayer Pattern," which I discussed in some detail in a technical article published last fall. A Bayer Pattern sensor transforms the inherently monochrome light capture process into a full-color measurement outcome by means of an additive (red, green and blue) color filter array. One filter color associates with each pixel, and the aggregate filter array contains twice as many total green filters as either red or blue, since the human visual system is disproportionately attuned to detail in green-spectrum light.
Partial-spectrum information from nearby pixels is interpolated via post-processing algorithms in order to create an approximation of each pixel's full-spectrum information. And, although niche alternative approaches such as Foveon's X3 may disagree, the Bayer Pattern approach seems to work well, judging from its longevity and ubiquity. The Bayer filter is named after its inventor, Eastman Kodak imaging scientist and "father of digital imaging" Bryce Bayer, who patented it (along with the requisite interpolation algorithms which enabled the use of it) in 1976. Bayer retired from Kodak in the 1990s and passed away on November 13, in Bath, Maine, at the age of 83. Our gratitude goes out to Bryce Bayer, as well our condolences to his family and friends.
