I know that expensive instruments can be used to determine the color of an object, but is there a poor man's way to determine color using filters? By "determine color" I mean know the relative proportions of primary colors. In other words, using filters to know that the proportions of red, blue and green are 5, 3, and 1, for example, or maybe something using the natural color system.

  • 2
    Color in the real world is not composed of proportions of primary colors. That's only how we approximate color on display devices. – twalberg Mar 25 '20 at 16:05
  • Related? photo.stackexchange.com/questions/10757/… – Rafael Mar 25 '20 at 16:30
  • This "poor man" probably would buy a cheap-o color webcam, and use captured images from it. Next steps might depend on what you actually mean by "color of an object." Of course, good luck finding a web-cam of any description right now. Almost as hard to find as toilet paper. – Solomon Slow Mar 25 '20 at 17:45
  • 1
    There's no such thing as "color" apart from perception. Light, and in fact all electromagnetic radiation, only has various wavelengths. For more please see: Why are Red, Green, and Blue the primary colors of light? – Michael C Mar 25 '20 at 20:02
  • 1
    What is your objective? An old colorimeter to calibrate your monitor can be purchased for about $25 (eg, ColorVision Spyder2). – xiota Mar 25 '20 at 22:37

One of the earliest forms of colorimetry involved a light sensor (or film) with a known sensitivity curve over the visible spectrum and, if film was involved, a densitometer to measure the resulting density.

You'd arrange constant lighting and measure the light level received (or density above bas+fog with a standardized development process) for each of three or more filters (also of known absorption characteristics, of course).

As you can probably see by now, this isn't exactly a "poor man's method", because of all the science needed behind the instrument(s). Still, it probably cost less to do it this way with a panchromatic film, in the 1950s, than to buy and use an actual colorimeter.

An alternative that was available at various times was color chips. This had limits, but the idea was that you'd hold a sample of known color against the object, and in the same light, and then move to a sample that was, say, a little more green, a little less blue, etc. until you had an exact match. This was time consuming, but if the fineness of your measurement needn't be too high, it was very likely to be less so than the film and filter method.

  • Rather than chips couldn't you approximate using a known calibrated color swatch either capured in image or in the same light. One of the printer scanner combination products I worked on would print a 120 color swatch and then scan it for calibration of the combination. for basics you could do a standard 24 color to calibrate the capture device and then measure the offset from your object? – Rowan Hawkins Mar 28 '20 at 0:37

What you need is a color densitometer, likely you can find a used one on ebay. You can buy red, green and blue separation filters from many sources. Red 25, Green 58 and Blue 47. Try close focusing a through-the-lens metering camera on a uniform color patch you wish to analyze. Have the camera display the exposure as metered. Note the shutter speed, aperture, and ISO associated with this reading. Now mount one of separation filters and take a reading. When finished, you will have data from the four readings. Noting the differences you can figure out the differences. This method crudely replicates how a densitometer works.

  • Of course, you'd need to calibrate your meter and filter set against a full spectrum white light source -- and then you'd get no better than 1 dB (1/3 stop) precision in your measurements with a camera's meter. – Zeiss Ikon Mar 27 '20 at 17:24
  • These filters are narrow cut -- The light source is not critical. – Alan Marcus Mar 27 '20 at 18:12
  • Except that if its emission is in emission lines (like some fluorescents) or very uneven over its visible spectrum (like a low wattage incandescent), your calibration is, at best, good only under that exact light. – Zeiss Ikon Mar 27 '20 at 18:13
  • Three separation filters... Four exposures? What are you using for the fourth exposure? Also, to clarify, is this method using film? – xiota Mar 28 '20 at 9:25
  • 1
    Color densitometers, like the ones at ESECO Speedmaster where I was an engineer, can take four readings, white light, red light, green light and blue light. The white light reading is call "visual", The filters in the visual position trim the response to match the human eye. – Alan Marcus Mar 28 '20 at 20:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.