Legacy ImageMagick Discussions Archive

Hi everybody,
i have the following task to solve and am looking for the best possible way.
I have an RGB image that i need to convert to CMYK. I am doing this using: convert rgb.tif -colorspace CMYK cmyk.tif (maybe also use some profiles)
What i need to make sure though is that when converting the image that i need for print always has a minimum tint of 3% in the C,M, and Y Channel. This is needed for a good print result.
my first thought was extracting all single channels, editing (darken) them somehow and combine them to an image again. I am not sure though if this is really a efficient and good way to do it.

Any recommendations?

Thanks for the help and all the best!
Sebastian

The K channel contains the minimum value found in a CMY image (a negated linear-RGB image), which is then subtracted from those color channels. As such ensure you get a 3% minimum in each color channel, reduce the K channel by 3% and add 3% to the three color channels.
WARNING: the above assumes K has at least 3% in it in all pixels. If it does not it will reduce it to zero (in non-HDRI versions of IM) but still add 3% to the color channels. Also if you have a pure primary color (like cyan or yellow), that color will become skewed (darker) when 3% is added to the other (previously zero) color channels.

To overcome these problems, you will need to define exactly what should happen in those two extreme situations. Then to implement, a intermediate channel image containing the exact amount to subtract from black and add to color channels needs to be created and applied.


Simply put, the above is too simplistic, except maybe for 'real world' images which has no color extremes. That is it does not contain pure white, or pure primary/secondary CMY colors. Images with pure black should not be problem with this scheme.

Please... What is this actually for? And when/if you do use it, let us know how it went, as others would likely be interested in your results. Good or bad.