home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ukma!usenet.ins.cwru.edu!agate!doc.ic.ac.uk!uknet!mucs!lilleyc
- From: lilleyc@cs.man.ac.uk (Chris Lilley)
- Newsgroups: comp.graphics
- Subject: Re: Color Sample to RGB Conversion
- Keywords: color conversion
- Message-ID: <6759@m1.cs.man.ac.uk>
- Date: 17 Nov 92 02:00:30 GMT
- References: <hewitt.721337021@usafa.af.mil>
- Sender: news@cs.man.ac.uk
- Organization: Dept Computer Science, University of Manchester, U.K.
- Lines: 94
-
- In article <hewitt.721337021@usafa.af.mil> hewitt@usafa.af.mil (W. Joe
- Hewitt) writes:
-
- > I am looking for a way to generate a color on a workstation that closely
- >matches a given sample from a color plate.
-
- The simple answer is, mix it by eye. This is also a pretty accurate
- answer. However you may want to do this automatically, perhaps for a
- large number of colours. So you need a way of spacifying the colour
- numerically.
-
- > I have tried scanning the image (24 bit scanner) with little success.
- >The scanning software only gives percentages of the R,G, & B components.
- >Using a little math I converted each percentage to a value in the range
- >0..255 (24 bit color system) for each RGB component.
-
- OK so you scanned it in as an RGB image. That will not help solve your
- problem, however.
-
- >The problem is that
- >the new color doesn't look very much like the original sample.
-
- No, it won't. Red, green, and blue are very loose specifications for
- colours. They differ between different monitors, and between different
- scanners.
-
- >I know
- >that the monitor can be adjusted to do minor compenstaion, but it still
- >doesn't get close enough to the original sample.
-
- Nope. And the adjustment would need to be non linear in any case -
- some colours get closer, others get worse.
-
- > Is there a better way to accurately reproduce a color sample on a 24 bit
- >workstation? Is there something I have missed in the conversion?
-
- As you have realised, RGB is not the way to go. You basically have two
- options.
-
- One is inexpensive and slow - mix the colours by eye using a
- paint package that lets you mix colours. This may be easier if you use
- HLS or HSV rather than RGB, but do whatever you find easier. This
- method is OK if you only have a few colours to match, and can give
- very accurate results. Be aware, however, that the colours you mix up
- will only hold good under the lighting conditions you mixed them under
- - flourescent striplight, electric lightbulb, sunny day, overcast day
- - and will look wrong under different illumination conditions.
- Sometimes just a bit wrong, sometimes markedly wrong.
-
- The other method is automated but might cost a bit - in time odr
- money.
-
- You need to use a standard numerical specification for your colours.
- These is one by the International Standards Organisation that is
- widely used, called CIE XYZ. The difference from RGB is that the CIE
- XYZ specification of a colour is an objective quantity that can be
- measured and reproduced on a wide variety of equipment. You then need
- to display this on your workstation by converting to RGB. This
- conversion will be different for each monitor. Details of how to do
- this are in various books or if you are stuck mail me for details. You
- will need some data from the monitor manufacturer to do the conversion
- - the colour of that monitors red, green, blue and white using
- standard CIE colour descriptions, and also the gamma (transfer
- function) of the monitor which compensates for its nonlinearity.
-
- Given the same data for the scanner, you could continue to use your
- existing scanner set up. The conversion would then be:
-
- your scanner RGB --> CIE XYZ --> your monitor RGB
-
- Relevant data should be available from the scanner manufacturer if you
- ask persistently ;-)
-
- The other way to get CIE XYZ values is to borrow (or even buy - but
- they cost about $6k) an instrument to measure them. These instruments
- typically look like a squat gun with some sort of readout like a LCD
- (or via a serial port). You put the sample against the muzzle, press
- the trigger and the X, Y and Z values are displayed.
-
- > Please E-mail any responses directly to me since my news feed is several
- >days behind.
-
- Well so is mine! Anyway perhaps others would like to know the answer
- too.
-
-
- --
- Chris Lilley
- ---------------------------------------------------------------------------
- Technical Author, ITTI Computer Graphics and Visualisation Training Project
- Computer Graphics Unit, Manchester Computing Centre, Manchester, UK
- Internet: lilley@cgu.mcc.ac.uk Janet: lilley@uk.ac.mcc.cgu
- Voice: +44 (0)61 275 6095 Fax: +44 (0)61 275 6040
- ---------------------------------------------------------------------------
-