Blending CMY Separations.

[MrToM]

I've been foolishly playing around with blending modes on images in web pages using CSS.

I found an example in which the follow image is used for blending, using 'multiply' blend mode.



When blended they do indeed create a full colour image but how is this done without the 'key' component?
I'm assuming its been included in the other 3 images but how?

Regards.
MrTom.

 

[Tom Mann]

By "the key component", I presume you mean the "K" (or "black", as in CMYK ), right?

If this is what you are concerned about, theoretically, ie, with perfect pigments, there is no need for the "K" component. Luminosity info is carried by the luminosity of each of the three CMY components, just like it is carried by the 3 RGB components in RGB color spaces.

It is only when one has to deal with real-world pigments that the "K" component is needed. It makes up for the propensity of real world C, M, and Y pigments to yield dark murky (sometimes brownish) blacks when mixed in equal (full) amounts. This limitation is why Photoshop provides both CMY and CMYK color spaces.

Another aspect of this is that there is exactly one conversion between an RGB image and a CMY image, whereas there are many different conversions between RGB and CMYK. This is one of the reasons why there are different printer ICM profiles for different printers.

HTH,

Tom

 

[MrToM]

Well I did a little more digging and found the answer to my question.
Thanks.

Regards.
MrTom.

 

[IamSam]

"Well I did a little more digging and found the answer to my question.
Thanks."

What was it, I'm curious?

 

[MrToM]

Well, what I was doing was to colourize the CMY channels with pantone CMY colours but I couldn't get anything like that example image.

What turned out to be the solution was to ignore CMYK to begin with, keep the image in RGB mode. Separate the image into R, G + B channels, (one per layer), add a 'levels' adjustment to each one and change the output levels of the 'other' 2 channels.

EDIT:
No need to separate into channels, just duplicate the image 3 times, add a 'levels' adjustment to each and set the Output Level as described below.

For eg:
The first layer will be a copy of the 'Red' channel....this can be labelled 'Cyan'.
Add a 'levels' adjustment and change both the 'Green' and 'Blue' channel output level to 255, 255.....(slide the 'Black' slider all the way to the right).
This has the effect of giving you a 'Cyan' channel.

Repeat for G and B....Green channel gives you 'Magenta', Blue the 'Yellow'.

When you blend those 3 layers together using 'multiply' you get a full colour image again.

Here's an action in case you're interested:
View attachment Channels 2 CMY - MrTom.atn

The image layer name has to be 'Layer 0' for it to run successfully.
Have the 'background colour' set to White.

It produces a 'RESULT' group with the 3 CMY layers, blended with multiply, and a, (background colour), BG layer. (BG not needed for result but prevents blending with original image).


Regards.
MrTom.