• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

B&W and color space?

Asim

New member
Not sure if this was the right area to post this....

I know, obviously, that "color" display is very much influenced by color space, especially on the print side of things. However, does the same apply to the black and white world? Does it matter which space I'm in? Specifically, I am editing pictures taken with a scanning electron microscope that will end up in publication. The journal only specifies the usage of RGB, not a specific color space. Also, since I rarely print myself (that portion of my workflow is not calibrated), I don't know if it would make a difference in output or not.

Your input is appreciated!

- asim

EDIT: In retrospect, I should have placed this in the "Digital Printing" area. Unfortunately, I don't believe we can delete posts. My apologies.
 
Last edited:

Asher Kelman

OPF Owner/Editor-in-Chief
Alain Briot is our resident master printer. Until he responds let me try. In fact, indulge me.

Let me stray far from you particular case of B&W to the B&W prints derived from making monochromatic images of our colored world.

Let's start by seeing an image. Here, in the world of sight, light of various wavelengths excite the retina to form an image which the brain converts to a concept of color. Our cameras attempt to also capture that same light and also convert the resultant image in a representation that approximates what we perceive through our own eyes.

With digital cameras, this information is stored in a file within the limited colors of the devices manufactured world of color (Gamut). So some colors we see, such as some blues or purples, might not be represented so well in the files generated by the camera.

Now when this file goes to the computer the color data is in one world (or gamut), but the monitor may not mechanically be able to generate and display very well, or at all, all those hues present in the image captured in the camera's (gamut or) unique world of color.

So software allows some perceptual remapping of colors so that to the eye, the picture taken by the camera ultimately appears identical to that seen with the naked eye.

Still there's one further step: the picture can't stay just in the computer.

We want a print. Trouble is that the printer also has its own unusual and unique gamut, (or particular world of color), and even can print hues, fore example some greens perhaps, that the monitor can never show. So once again, translation and remapping of colors is done to create in the final print, a vivid life-like image.


Well, with most B&W images, one is actually creating a set of artificial tones in an image that really was much more limited in variation of blackness or whiteness. It is like creating in an object an extra patina of brilliantly fine shading that adds detail to the already profound shading of the luminance values in the picture.

Every different color can be made to appear as bright or dark as you happen to wish. you can convert to B&W in CMYK, RGB or LAB and have slightly different elements of control as to what becomes dark, grey and white and everything in between. All one is doing is changing how one perceives the some brightness values of an image (i.e. it's amount of black or white) by making your own choices on how should the pink of the skin be represented: a brightness of 200 or perhaps a darkness of 60? It is entirely your choice. The great advantage is that a dress, for example with flowers having identical brightness, would show as an almost uniform grey with hardly any definition of the pattern. By assigning to the flower and leaves different luminance values, suddenly the flower pattern pops out and the dress looks pretty again.

Now lets jump to the electron microscope picture. Here, as I understand it, there is only a grey scale image, the points are all somewhere between say 20 and 200 in brightness. There is however no color information at all.

All you can do is record that image and then try not to add more noise before printing. There is no color to take advantage of to enhance your picture. Still there are several things you can do which you would do with any image: define black point, white point and mid point so as to utilize the dynamic range of your printer effectively. Just doing this can already make your image, yes even an E.M. image come to life. Next one might consider a curve adjustment to make the picture "pop". A modest "S" shaped curve would be tried, Then the percent of this layer would be decreased to get the optimum result.

All this can be done in any color space or in monochrome. I would use adobe RGB but make a note with the submission.

I went through a lot just to say it doesn't make any difference for your E.M. picture, but this was an opportunity to introduce to some the concepts of machine specific worlds of color, or "gamuts" and our need to have colors remapped between these worlds in order to create a lifelike image in any machine: monitor, projector or printer.

Having said that, let's go back to your image from the electron microscope which is sitting there monochromatic. You might very well choose to apply a "hint of tint". Since you are in RGB it is easy to do that with any one of the color adjusting layers of photoshop. There is no reason not to do so, since the various membranes or particles imaged most like are not gray in the first place.

The next manipulation is more subtle. If you go to Reindeer Graphics, you could download, at a price, some image analysis software which could create pseudo colors according to, for example, the rate at which an edge or a junction changes density or even the rate at which the rate of change changes. (ie 2cd derivative). Now they have many more subtle and sophisticated image enhancing tricks which you can apply.

This can sometimes make detail, otherwise obfuscated by granularity for example, to suddenly be revealed. The results can be delivered in color or, (as in the case of assigning the real color information of RGB images to B&W), you can assign the new derived detail to be overlaid on to your grayscale image, to appear as various shades of grey.

So, your original simple question about color space was used to indicate a myriad of ways of delivering a more useful image.

Asher
 
b+w color space

some detail-questions more for this theme.
why not converting the images to a b+w greyscale space before sending the images out?
if you would do this you can select the grey gamma.
everywhere is written that for macs usualy 1.8 is used and for pc 2.2.
i never understood this.... have mac userds other gradated eyes? or wherefrom comes this difference?
so with wich grey gamma should i give out my b+w converted greyscale data to my client and how will this vary my printed result?

one more question- if i give b+w images in color space to the client and if the client ( magazin or printer..) convert to greyscale is the conversion to the greyscale space in every case the same if i give to the magazin a e.g. adobe color spaced image or a srgb color space image?
at least the numbers of a middle grey are different in srgb or adobe rgb or other color spaces.
great confusion here in my head about this ...... so advices appreciated.
 

Asim

New member
ack, in the middle of an unbearable move...will read this more thoroughly when I get setup in my new place. Nonetheless, thanks to you 2 for answering. Asher, that's one long post! :D thank you for the detail, though!
 

Asher Kelman

OPF Owner/Editor-in-Chief
Asim, good luck onn the move!

.........as for the detail. I take this forum very seriously. We want to help everyone get their vision to be realized print-ready or in a stunning print so that it is impressive and the best it can be.

That is what we want to do. As all the moderators come get more and more involved, this will be apparent.

Meanwhile, I try to go through as many portfolios as I can between things, so I can know photographers here by their work. At least at this stage.

Asher

P.S., what science are you doing and what is your role in the project?
 
Why not send B/W out as RGB?

There was a question one or two posts back about whether to send out a greyscale image as an rgb image. I would think you should avoid this, unless you really want a color image (toned, perhaps) and it will be reproduced in color. But letting someone else flatten your image could, in all probability, flatten the life out of your image. There are big differences between the results of different ways of turning rgb into greyscale -- desaturating gives a different tone curve than a good b/w camera profile.

Something I don't know is whether representing a b/w image in an rgb jpeg is a waste of bits as well. Can anyone tell me what the bit resolution of a b/w image rendered as a jpeg is? Is it 8 bits or 24 bits(/3 because the r=g=b signals)?

scott
 
i was not thinkin in asking them to convert it to b+w, ofcourse my intention was to convert it to b+w myself but still let it in the rgb colorspace, or if its better to send them real greyscaled images,- but when i do so with wich gamma should i do? how it will print out if sent with 1.8,- or in 2.2. at my monitor there will not be any difference cause ps is converting it in different grey values in both case. but what about the magazin?
 

Alain Briot

pro member
Without going into a discussion of color spaces for black and white (of which there are few, if any) the best is to print from Adobe RGB since this is the color space you are working in.

Alain
 

Harvey Moore

New member
The SEM image files I am used to (and this goes back about 5 years from older equipment) were 8 bit grey scale tiff files.

My question is, does it make any difference what color space is used in that case? The only thing that may affect this of course is for colorized images.

harvey
 

Alain Briot

pro member
Chroma (the amount of a color in an image) is controled by saturation. If you desaturate an image 100%, it doesn't matter if you save it in an RGB or a greyscale color space. There won't be any color in the file. The only difference will be the size of that color space, in terms of tones (see below), and the size of the file. A greyscale file is smaller than an RGB file because it has only one channel instead of 3.

Once an image has been converted to a specific color space, you are limited to the colors, or tones in the case of B&W, described by that space. In other words, the only thing that would make a difference is converting to a smaller color space, which would of course further reduce the tones. Converting to a larger color space wouldn't add any more data (tones in the instance of B&W).

The best approach is to start with a large colorspace, one that describes all the colors/tones in your capture or scan. When you start with a file that has already been converted to a specific color space, there is little you can do, especially when that color space is small.
 
Last edited:
I would recommend to look at this technique:

http://www.outbackphoto.com/artof_b_w/bw_09/essay.html

I laso habe the B&W image in AdobeRGB and then use Roy's technique to use good B&W profiles for e.g. the Epson printers (with Epson B&W driver settings). Avoids a lot of trial and error.

>Also, since I rarely print myself (that portion of my workflow is not calibrated), I don't know if it would make a difference in output or not.

At least your gamma settings should be ok (use e.g. gamma 2.2).

Uwe
 
Top