Message-Id: <mailto:199409181017.FAA08589@library.wustl.edu> Date: Thu, 15 Sep 1994 14:20:42 -0700 From: Navid Haddadi <mailto:nhaddadi@UCS.USC.EDU> Subject: Re: 8-bit vs. 24-bit color To: Multiple recipients of list IMAGELIB <mailto:IMAGELIB@ARIZVM1.BITNET>
A color photo is basically three B&W photos. One photo for each of three color channels Red, Green, Blue (RGB). Each color plane is quantized using 8 bits (256 levels) for a total of 24 bits.A color photo can be converted back to a B&W photo by averaging the three color planes. This means that a 24-bit color image is equivalent to a 8-bit gray scale image; in other words, if a B&W photo is scanned at 24-bit color and 8-bit gray scale, result should be approx the same.
Almost anyone can resolve more that 256 shades of gray. 10 or 12 bit gray scale is not unreasonable requirement (30 or 36-bit color). However, for most images where a complex natural scene is depicted 256 shades of gray is sufficient. Images with smooth transitions among the shades of gray (e.g. X-ray) may require more bits.
BTW, you may want to scan a B&W photo using 30-bit color scanner if you wish to preserve the natural discoloration of the original. Navid mailto:nhaddadi@ucs.usc.edu