For both of my hobbies -- photography and model building -- I'm often concerned with color matching between various things. A professional Colorimeter is typically an expensive piece of equipment, but with extensive knowledge of optics and electronics I don't think that constructing one would be beyond my skill levels. Does anyone know of any web sites or magazine articles which give information that would be useful in designing or building one "at home"?
TIA Norm
I've been round this loop for a client. You can get cheapish colour sensors, for instance the TAOS TCS230. The problem turns out to be determining the response of the filters in those sensors - they tend to overlap somewhat, and also there are typically IR leaks to deal with. It is easy to get numbers, but somewhat harder to relate them to what you can see ...
If I do it again I'll use a broad band sensor (TAOS TSLxx) and spend some money on precision colour filters.
Dave
If you just want to know what color something is, could you take a picture of it with a digital camera and look at the R,G,B values of individual pixels? (like, with a paint program)
Good Luck! Rich
Ages ago there was a project in Nuts and Volts that used several colored LEDs and measured the reflected light from the surface using an unfiltered sensor.
Have a look at this:
Been there -- done that. Also used a scanner. Imported the output of both into Photoshop and sampled colors that way. It works but it stinks as a workable procedure.
Norm
Yeah -- they used an old CD as a diffraction grating. I've thought of doing the same thing with some surplus prisms I have. Just add some optics to spread out the output light and a measuring instrument. It may yet happen that way.
Norm
Maybe this guy (COTS product, not a design/build)
I belive it works by sampling the reflections from a sequence of differently colored LEDs. Shouldn't be unreasonably hard to do with LEDs of defined spectra and sensors (photo-transistors?) of defined sensitivity. Calibrating it might be tough if you wanted absolute measurements (use white/gray/black targets?) but used for color matching it should be do-able.
-- Rich Webb Norfolk, VA
| | I belive it works by sampling the reflections from a sequence of | differently colored LEDs. Shouldn't be unreasonably hard to do with | LEDs of defined spectra and sensors (photo-transistors?) of defined | sensitivity. Calibrating it might be tough if you wanted absolute | measurements (use white/gray/black targets?) but used for color | matching it should be do-able. | | -- | Rich Webb Norfolk, VA
Yeah -- if all I want to do is to color match from physical objects then yes, calibration isn't really necessary though in the case of using it to match paint, I'd still need enough calibration to know what color(s) to add to what I already have to get where I want to be.
Thanks Norm
You also need more than RGB to catch 'metamerism' - colors looking different under different lighting.
Thomas
Very true. But how would a commercial colorimeter do it -- if it indeed does?
Norm
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.