Hi all, I have not used the IM program as yet but wondering if it can do what I want.
I am a photographer and one of the things I do is alternative photography where I use digital transparencies to print coloured inkjet negative for contact printing on photo paper using a UV light source. One of the issues I have is that the inkjet print out just uses the black ink and gives a gradation of a grey scale image. What happens when I contact print on the photo paper the gradation of greys is not uniform and quite contrasty. Though if I use a different colour other than black ink, say blue it is somewhat better but still contrasty. So my need is to come with some sort of color mapping. I would print say a gray scale from dark to light of each specific colour Red, Green and Blue. Print these out as inkjet transparencies and contact print them on photo paper. Then I would scan these photo paper with the grey scales and store the value. This way I'm building a database of color mapping. Once this is done, I would use this database of color to change my image to the color in my mapped database based on gradation of gray.
So for example: I print a square box of Red "R:255 G:0 B:0" on a digital transparency on my inkjet printer. This is then contact printed on photo paper. I scan this photo paper and the square. I analyze the color value thats in that square, say value is "X". I then build a database ---> "R:255 G:0 B:0" = "X" I repeat the above as much as I want to build a large database of color mappings.
Once the above is complete, I then use this color database to change my colors in my final image to get a full tonal gradation and wont get the contrasty images I'm getting !
So wondering if I can achieve this in IM?
Let me know if I haven't explained myself fully Thanks jag2x
Specific color mapping
-
- Posts: 12159
- Joined: 2010-01-23T23:01:33-07:00
- Authentication code: 1151
- Location: England, UK
Re: Specific color mapping
Fascinating. I worked with B&W photography, but never got the hang of colour photographic printing because the darkroom was too dark.
So: you use an inkjet printer to make a transparency, and contact-print that to colour photographic paper. Or is the paper black-and-white? I assume it is colour.
First point: your inkjet inks are cyan, magenta and yellow. Colour photographic paper is also CMY. But the photographic process reverses colours (and hence also reverses tones). So if you just use inkjet cyan, the result will be magenta + yellow = red. Darker cyan will give less magenta + yellow, so a lighter result.
I would test each channel separately. Create an inkjet wedge with just cyan, from none to maximum, ie white to darkest cyan. Make the contact print. This should be black to red. Scan this. Combine the input numbers with the scanned values, and we have a look-up table, of input-output values. In photography this is a "characteristic curve". It is one channel of an ImageMagick CLUT (colour look-up table). The CLUT can be inverted so if your input image has a range of red values, process it with the inverted clut and this gives the correct amounts of cyan ink to give the required photographic red.
Repeat the test for magenta and yellow.
So: you use an inkjet printer to make a transparency, and contact-print that to colour photographic paper. Or is the paper black-and-white? I assume it is colour.
First point: your inkjet inks are cyan, magenta and yellow. Colour photographic paper is also CMY. But the photographic process reverses colours (and hence also reverses tones). So if you just use inkjet cyan, the result will be magenta + yellow = red. Darker cyan will give less magenta + yellow, so a lighter result.
I would test each channel separately. Create an inkjet wedge with just cyan, from none to maximum, ie white to darkest cyan. Make the contact print. This should be black to red. Scan this. Combine the input numbers with the scanned values, and we have a look-up table, of input-output values. In photography this is a "characteristic curve". It is one channel of an ImageMagick CLUT (colour look-up table). The CLUT can be inverted so if your input image has a range of red values, process it with the inverted clut and this gives the correct amounts of cyan ink to give the required photographic red.
Repeat the test for magenta and yellow.
snibgo's IM pages: im.snibgo.com
Re: Specific color mapping
Hi Snibgo,
For now the images i'm contact printing the transparencies are on are black and white photographic paper.
The issue is the characteristic curve created using just one ink colour is that you do not get a smooth tone, the highlights are the issue as the tonality between them is quite contrasty. Though if you incorporated multiple inks and add them in the right places the tonality is better, but then another place becomes contrasty
So ultimately I'm thikning of coming up with muitple color wedges and contact printing on black and white photo paper.
Record what each color of the wedge is and the black and white paper wedge equals to, therefore building a database of tonal relationships.
Once i'm satisfied with all this data, I use the program to map my digital image with all the different inkjet colors to give the full tonality, this would I presume would be some psychdelic image, I then print this out on the inkjet transparecny. Which I then contact print on my black and white paper and finally gives me the smooth tonal image
So is it possible?
For now the images i'm contact printing the transparencies are on are black and white photographic paper.
The issue is the characteristic curve created using just one ink colour is that you do not get a smooth tone, the highlights are the issue as the tonality between them is quite contrasty. Though if you incorporated multiple inks and add them in the right places the tonality is better, but then another place becomes contrasty

So ultimately I'm thikning of coming up with muitple color wedges and contact printing on black and white photo paper.
Record what each color of the wedge is and the black and white paper wedge equals to, therefore building a database of tonal relationships.
Once i'm satisfied with all this data, I use the program to map my digital image with all the different inkjet colors to give the full tonality, this would I presume would be some psychdelic image, I then print this out on the inkjet transparecny. Which I then contact print on my black and white paper and finally gives me the smooth tonal image

So is it possible?

-
- Posts: 12159
- Joined: 2010-01-23T23:01:33-07:00
- Authentication code: 1151
- Location: England, UK
Re: Specific color mapping
Some black and white photographic papers have variable contrast, depending on the colour of the light. Yellow light for low contrast, magenta for high contrast. If you use these papers, this complicates matters.
If you use fixed-grade paper (one contrast grade, regardless of light colour), the calibration should be straightforward. As you say, make some colour wedges and measure the results. The result would be an ImageMagick clut file. It might have 256 entries, something like this:
This file can then be used to automatically transform a grayscale image into the appropriate colours that will result in the final print looking the same.
If you use fixed-grade paper (one contrast grade, regardless of light colour), the calibration should be straightforward. As you say, make some colour wedges and measure the results. The result would be an ImageMagick clut file. It might have 256 entries, something like this:
Code: Select all
Entry R,G,B
0 250,240,250
1 245,238,245
2 241,233,243
: :
254 4,2,1
255 0,0,0
snibgo's IM pages: im.snibgo.com