Best way to find colour values of physical objects - Pantone colour bridge guide?

Forum|Forum|5 years ago
September 13, 2020
返信数 3.
5510 ビュー

Can anyone recommend which is the best Pantone colour guide (or similar) for finding the colour values of real life physical objects? I was going to just get the colour bridge, but didn't know whether the coated or uncoated version would be more suitable - I assume the 'same' colours in each would either look different on the guide due to the coating, or have different colour values assigned so they print correctly? I'm not going to be printing anything though, so that's not a consideration here. Should I use the coated or uncoated version?

The important details:

What I am trying to acheive is accurate colour representation of the object (in this case threads from a cotton thread sample book) in a sample or colour palette I will create in Photoshop (CS3). I do understand about the variation in how monitors display colour, and I know that other people's monitors may display the colours differently depending on calibration, and that's fine, as long as I can be sure that the actual colour values I'm using are as representative of the real life colour as possible. I've been creating swatches by eye in Photoshop up to now, which has mostly been fine, but some colours have been almost impossible to get right - the difference between reflective and transmissive colours doesn't help, and my monitor isn't properly colour calibrated (yet), so even what looks like a perfect match to me may not be. (If it matters, I'm using an Asus IPS HD LED-backlit LCD Monitor and NVIDIA GeForce GTX 1060 6GB graphics card.)

I don't want to use an app or a photo of the item to get colour values, because that has too many of its own problems with regard to white balance, lighting, etc. - I'm good at colour matching by eye, so my preference is to match it physically to a good guide rather than relying on a camera picking up the colours correctly, then use the colour values given in the guide to get that colour in Photoshop. I would probably be using RGB values (unless anyone recommends otherwise?) as there is no printing involved. I thought about using Lab, is there any benefit to this? Any tips on the best working space / colour settings / profiles to use in CS3 (on Windows 10) would be gratefully received! Basically I'm trying to get as close as possible to having the colour swatch on screen (on a perfectly calibrated monitor anyway) look the same colour as the physical item held in your hand (at least with the same lighting conditions it was matched under). Not simple, I know... Oh, and I'll be posting the end result online, so I know that web compression, web colours etc. might be an issue (?), but I at least want the original image I create to be as accurate as possible, so any colour discrepancies are due to viewers equipment etc., not me! I should also note that if I just had a few colours to match, I'd just eyeball it, but I have literally hundreds to do, many of which are very similar to each other, so I'd love to have a better, more streamlined method for this.

Other points:

I'd be matching under natural daylight conditions - I'm considering investing in some D65 lighting with high CRI (95+), but I'm assuming for the most part, lighting conditions aren't too important for this, given both object and guide will be viewed under the same lighting.

I'm also very tempted to build an Arduino based RGB colour picker - anyone done this? It would be interesting to know if it picks the same values from scanning a Pantone guide as the ones listed...

Sorry for the massive wall of text! I figured the details would be important for getting the most helpful and relevant answers... Any recommendations or tips at all would be very much appreciated here, I'm happy to consider alternatives, but please bear in mind I can only really afford one of the colour bridge guides, so certainly can't stretch to anything fancy like the £7.5k cotton swatch library! 😞 I do plan on properly colour calibrating my monitor at some point, but would still want what I've described here, so that wouldn't be the solution for me by itself.

Thanks in advance to anyone who can help!

Norman Sanders

Legend

“….I'm good at colour matching by eye, so my preference is to match it physically to a good guide.”

A response to your post may be distilled to four items: you, your viewing environment, your client, your client’s viewing environment.

You:

Have you ever had the precision of your color acuity evaluated? (Not a simple color blind test using a set of pseudoisochromatic test plates.)

Are you aware that the precision of physical matching is affected by many things, among them fatigue, anger and even the sequence of colors viewed? (It is generally recommended that you build up to a saturated color by viewing less saturated samples first. It is also best not to agonize over a decision for too long because your eyes suffer retinal fatigue and become less sensitive if you stare at the same object for more than 10 seconds at a time. If you are struggling to make a decision, repeat the process after resting your eyes.)

Your viewing environment:

Do you have a viewing booth available that meets industry standards? Those standards include but are not limited to color temperature, color rendering index (the visual effect of the light source on 8 specific pastel colors), and light intensity at the viewing surface…and even the surrounding area which demands a specific Munsell value. When the presentation of the match involves your desktop or laptop computer, what are those standards and controls?

Are you aware the de facto standard reference in the US is the set of Pantone swatches, and that its most useful designation is its Lab value because Lab is device independent?

Your client:

Have you considered that your client’s color acuity is out of your control? I have read that one man in seven has some form of color anomaly, and one woman in 400 has the problem. I cannot vouch for those numbers but even though I have spent a good deal of my vocational life in the letterpress and lithography fields serving the of some of the most demanding art directors on the planet, I never argue with my wife about color matches.

Your client's viewing enviornment:

Have you considered that you have no control of each client’s viewing environment? Simply put, colors may match in one lighting environment and not in another. Check out Wikipedia: Metamerism. In addition, you have no control of your clients’ computer settings regarding color. Believe me, they vary. The expression for this in New York is a simple: Oy.

Bottom line: You are attempting perfection in an area where the variables make perfection impossible. (In the spirit of full disclosure, this is a fragment from a three-hour lecture I presented to my college class each year that was devoted to “Color Theory and the Printed Page.”)

davescm

Community Expert

Hi

A few comments

1. "my monitor isn't properly colour calibrated (yet)" Get that sorted first, or you are doing something akin to viewing through randomly tinted glasses

2. "I'm assuming for the most part, lighting conditions aren't too important for this" Wrong. Look up metamerism. This is an effect where two colours can appear to match under one lighting condition but can look very different under another.

3. Once you get colours captured make sure that your web image versions are converted to sRGB and have the sRGB ICC profile embedded.

There is a good article on using spectro's for on screen representation of textile colours here.

https://www.xrite.com/blog/digital-color-textiles

For textile to screen matching, it comes down on using a 45/0 degree spectrophotometer - which would include the i1Pro 3 Plus which has a wider aperture than the normal i1Pro3 but is not as expensive as some of the higher end spectros designed specifically for textiles. Although still not cheap, it may be a good solution given your need to calibrate your screen as well as accurately measure colour.

https://www.xrite.com/blog/digital-color-textiles

Dave

Mylenium

Legend

Random brain farts:

There's a reason why commercial spectrometers cost 20k and up. A simple RGB scanner lit by a shoddy LED isn't going to do much.
Not to point out the obvious, but since you are hanging on to a hopelessly outdated version of PS that can't even handle contemporary color profiles may negate all your attempts of using them.
The old "There is no causal relation between printed colors and RGB" still applies, even with Pantone.
You're ignoring perceptive effects like dark colors generally looking darker to the human eye than to a light sensor.
Since you're talking cloths/ yarns, there may be a ton of internal scattering and anisotropic surface scattering.
The difference between coated and uncoated is simple - coats affect the perceived saturation/ darkness. Dark colors look even darker, light colors even lighter.

Outside that I'm not even sure if there is any way to represent textile colors exactly digitally. Some of them are generated with multiple overdying procedures and it would be impossible to genuinely capture the minute shifts. In fact you may even have to do this a hundred times more than already because obviously even trivial stuff like the level of bleach/ yellowing in the textile, its roughness and thus absorption of the dye and a million otehr factors would affect the outcome. you may be aiming for the impossible...

Mylenium