Options
Chapter 7 Professional Photoshop Keeping the Color In B&W
pathfinder
Super Moderators Posts: 14,696 moderator
Chapter 7 - Keeping the Color in B&W
Dan Margulis opens the chapter with a brief discussion of the history of the graphic arts, and states that he believes the present century is only 2nd to the 15th century in overall quality of graphic arts. He goes on to explain that the 15th century had Michelangelo, Leonardo, Botticelli, Donatello, Ghiberti, and Brunelleschi. Speaks well for the present century.
The author then says that Chapter 7 is "arguably the most important chapter in the book. Don't skip it even if you think it is irrelevant to the sort of work you do." (When an author says this is the most important chapter in his book, I tend to listen.) He then goes on to list several reasons for his opinion. (But sometimes Dan is little to clever by half.. as I will discuss much later.....)
1) Quality duo-tones require a good B&W for beginning
2) Dan says the only difference between a color to B&W conversion and an RGB to CMYK conversion is one of degree. "In CMYK there is not a very good blue, and in gray scale there is not a good red or green either" ( I think this is a graphic artists joke - kind of, maybe )
3) The technique for good B&W conversion are the basis for the powerful methods to come in Chapter 8 and 10
4) This chapter is the beginning of the introduction to Channel Blending , which is what the rest of this book is about.
He next spends a couple of pages demonstrating why Image> Grayscale is a very poor way to do a black and white conversion.
The RGB channels are blended in a 3 - 6 - 1 ratio in Photoshop when converted to grayscale. No matter what color space the image is currently in, Photoshop creates an RGB version and blends it with the 3 -6- 1 ratio of Red, Green and Blue. The author does not explain why these numbers are chosen, just that that is the way it is performed.
He demonstrates with a picture of a parrot and a Canadian flag, that humans prefer the same red hue to be different shades of gray depending on the B&W image they are in, or the context they are in. Formulas of course, do not consider context whatsoever, and Image > Grayscale creates less than satisfactory images as a result.
Part of the problem, is that when humans look at colored images, we discriminate on the basis of Hue, Saturation, and Brightness. But when we covert to Grayscale, Hue and Saturation differences entirely disappear, and they may be seen as exactly the same shades of gray in the image.
To demonstrate this disappearance of Hue and Saturation effects. Take a new, empty, white image in Photoshop, and add a vertical gradient from black to white.
Convert the image to LAB. Now in the A channel paint the grey area entirely white and then add a vertical black to white gradient as the a channel grayscale image. In the B channel, paint the grayscale image white, and add a horizontal black to white gradient. The final Channel Palette should look like this.
Return to the Layers palette and the image is quite colorful,
but when converted to grayscale, the image looks like the original vertical gradient created in RGB.
When you convert to Grayscale, all color information disappears, and the vertical black to white gradient in Luminosity reappears.
Hue and Saturation, both components of color, have no effect on the grayscale, ONLY Luminosity or Brightness will be displayed in a B&W converted image.
This produces the first important concept - A good RGB file may create an unacceptable B&W, but an unacceptable color image may produce an excellent B&W. To convert successfully, you need to anticipate what will vanish when Hue and Saturation take a hike, leaving only Luminosity.
Dan next begins talking about Machiavelli and Friends or Foe. What he really refers to is whether the Luminosity of a particular Hue needs to be lighter or darker. If we need red to be lighter in luminosity, we will need to brighten or lighten the Red channel. If we need blue to be lighter in luminosity, we need to lighten the blue channel.
I strongly encourage the reader to pick an image of theirs with bright reds, greens, blues, and yellows, and open it Photoshop. Switch to the Channels palette, and inspect each channel carefully, and watch what happens to reds, blues and greens in the Red channel gray scale image, and reds, blues, and greens in the Green channel, and reds, blues, and greens in the Blue channel. Red colors will be almost completely white in the Red channel and green will go almost black. In the Blue channel, yellow will be almost completely black, and blue will be almost white. Remember the 3 red - 6 green - 1 blue ratio in Photoshop's native grayscale conversion.
Here is a sexy couple eating ice cream in Canada, wearing bright red, blue and black clothing.
Watch what happens in the Red, the Green, and the Blue channels for each of those colors. In the Red channel, blue is almost blacker than Black is black, and red and white are almost indistinguishable.
In the blue channel, blue becomes very white, and yellow goes very dark.
The green channel is a better overall rendition as there is no strong green color in this image
Now, here is a serious looking motorcyclist in front of his yellow motorcycle. Notice the green refrigerator door behind him and the bright blue panels on his saddlebag.
Now, look at each of these colors in the Red Channel. His shirt is almost as white as the white building behind him. The green refrigerator door has become blacker than his black pants, and the blue panels on his saddlebags are now as black as his pants.
In the green channel, the green refrigerator door is not nearly as bright as the white building, but it is very much brighter than his red shirt or his black pants.
In the Blue channel, see how white the blue panels have become and how black the yellow tank is now.
The three channels, Red, Green, and Blue, are the key to altering the image so that luminosity, not hue and saturation, are visible when we finally convert the image to grayscale for our final Black and White image.
When we get ready for some channel manipulations to convert an image to grayscale, Photoshop offers three basic methods.
1) Image>Apply Image
2) Image>Calculations
3) Image>Adjustments>Channel Mixer
Or you can just purchase an action or a plug in and hope for the best :-) All of these methods substitute luminosity contrast to replace the vanished Hue and Saturation.
Dan jumps in with both feet and discusses some of the differences in the three different methods in PS. He favors Apply Image and Calculations over Channel Mixer, because Apply Image and Calculation can be blended with Alpha channels, or with channels on different layers, or even with layers from a different document, if the document is the same size. You can even blend a cyan channel with a green channel from two different copies of the same document as long as the pixel dimensions match exactly.
When using Channel Mixer is good to keep Photoshop's Image>Grayscale calculation of 3 Red - 6 Green - - 1 Blue in mind, so that we can vary from that as needed by our friends and foes.
So lets make a B&W image of our serous rider and see what happens with the various tones. The first effort is the direct grayscale conversion so hated by the author via Image>Mode>Grayscale and the dialogue box - "Discard color information?" Answer yes and you see this sterling image.
To demonstrate the 3 red - 6 green - 1 blue ratio used by Photoshop for grayscale conversion, we are going to use Channel Mixer and set the ratios at 30 red, 60 green and 10 blue as in this dialogue box
[imgl]http://pathfinder.smugmug.com/photos/125666637-M.jpg[/imgl]
and convert to monochrome. Amazingly enough, the image is an exact replica of the image we just created by direct grayscale conversion, as one might have guessed.
[imgr]http://pathfinder.smugmug.com/photos/125666665-S.jpg[/imgr]
Now, lets see if we can do a little better job with the lowly Channel Mixer. Lets see if we can make his shirt black and keep the bike gray. Let's try Red 2 Green 108 and Blue - 2 as shown here.
Here is the image with a shirt almost as black as his pants and a gray bike and a white building with pretty good facial tones.
You do not need to limit one channel to l00 units - it is sometimes worthwhile to go way past these amounts. Pushing the Red channel up to 150+ can emulate a faux IR for example.
(further discussion continues in the next post)
Dan Margulis opens the chapter with a brief discussion of the history of the graphic arts, and states that he believes the present century is only 2nd to the 15th century in overall quality of graphic arts. He goes on to explain that the 15th century had Michelangelo, Leonardo, Botticelli, Donatello, Ghiberti, and Brunelleschi. Speaks well for the present century.
The author then says that Chapter 7 is "arguably the most important chapter in the book. Don't skip it even if you think it is irrelevant to the sort of work you do." (When an author says this is the most important chapter in his book, I tend to listen.) He then goes on to list several reasons for his opinion. (But sometimes Dan is little to clever by half.. as I will discuss much later.....)
1) Quality duo-tones require a good B&W for beginning
2) Dan says the only difference between a color to B&W conversion and an RGB to CMYK conversion is one of degree. "In CMYK there is not a very good blue, and in gray scale there is not a good red or green either" ( I think this is a graphic artists joke - kind of, maybe )
3) The technique for good B&W conversion are the basis for the powerful methods to come in Chapter 8 and 10
4) This chapter is the beginning of the introduction to Channel Blending , which is what the rest of this book is about.
He next spends a couple of pages demonstrating why Image> Grayscale is a very poor way to do a black and white conversion.
The RGB channels are blended in a 3 - 6 - 1 ratio in Photoshop when converted to grayscale. No matter what color space the image is currently in, Photoshop creates an RGB version and blends it with the 3 -6- 1 ratio of Red, Green and Blue. The author does not explain why these numbers are chosen, just that that is the way it is performed.
He demonstrates with a picture of a parrot and a Canadian flag, that humans prefer the same red hue to be different shades of gray depending on the B&W image they are in, or the context they are in. Formulas of course, do not consider context whatsoever, and Image > Grayscale creates less than satisfactory images as a result.
Part of the problem, is that when humans look at colored images, we discriminate on the basis of Hue, Saturation, and Brightness. But when we covert to Grayscale, Hue and Saturation differences entirely disappear, and they may be seen as exactly the same shades of gray in the image.
To demonstrate this disappearance of Hue and Saturation effects. Take a new, empty, white image in Photoshop, and add a vertical gradient from black to white.
Convert the image to LAB. Now in the A channel paint the grey area entirely white and then add a vertical black to white gradient as the a channel grayscale image. In the B channel, paint the grayscale image white, and add a horizontal black to white gradient. The final Channel Palette should look like this.
Return to the Layers palette and the image is quite colorful,
but when converted to grayscale, the image looks like the original vertical gradient created in RGB.When you convert to Grayscale, all color information disappears, and the vertical black to white gradient in Luminosity reappears.
Hue and Saturation, both components of color, have no effect on the grayscale, ONLY Luminosity or Brightness will be displayed in a B&W converted image.
This produces the first important concept - A good RGB file may create an unacceptable B&W, but an unacceptable color image may produce an excellent B&W. To convert successfully, you need to anticipate what will vanish when Hue and Saturation take a hike, leaving only Luminosity.
Dan next begins talking about Machiavelli and Friends or Foe. What he really refers to is whether the Luminosity of a particular Hue needs to be lighter or darker. If we need red to be lighter in luminosity, we will need to brighten or lighten the Red channel. If we need blue to be lighter in luminosity, we need to lighten the blue channel.
I strongly encourage the reader to pick an image of theirs with bright reds, greens, blues, and yellows, and open it Photoshop. Switch to the Channels palette, and inspect each channel carefully, and watch what happens to reds, blues and greens in the Red channel gray scale image, and reds, blues, and greens in the Green channel, and reds, blues, and greens in the Blue channel. Red colors will be almost completely white in the Red channel and green will go almost black. In the Blue channel, yellow will be almost completely black, and blue will be almost white. Remember the 3 red - 6 green - 1 blue ratio in Photoshop's native grayscale conversion.
Here is a sexy couple eating ice cream in Canada, wearing bright red, blue and black clothing.
Watch what happens in the Red, the Green, and the Blue channels for each of those colors. In the Red channel, blue is almost blacker than Black is black, and red and white are almost indistinguishable.
In the blue channel, blue becomes very white, and yellow goes very dark.
The green channel is a better overall rendition as there is no strong green color in this image
Now, here is a serious looking motorcyclist in front of his yellow motorcycle. Notice the green refrigerator door behind him and the bright blue panels on his saddlebag.
Now, look at each of these colors in the Red Channel. His shirt is almost as white as the white building behind him. The green refrigerator door has become blacker than his black pants, and the blue panels on his saddlebags are now as black as his pants.
In the green channel, the green refrigerator door is not nearly as bright as the white building, but it is very much brighter than his red shirt or his black pants.
In the Blue channel, see how white the blue panels have become and how black the yellow tank is now.
The three channels, Red, Green, and Blue, are the key to altering the image so that luminosity, not hue and saturation, are visible when we finally convert the image to grayscale for our final Black and White image.
When we get ready for some channel manipulations to convert an image to grayscale, Photoshop offers three basic methods.
1) Image>Apply Image
2) Image>Calculations
3) Image>Adjustments>Channel Mixer
Or you can just purchase an action or a plug in and hope for the best :-) All of these methods substitute luminosity contrast to replace the vanished Hue and Saturation.
Dan jumps in with both feet and discusses some of the differences in the three different methods in PS. He favors Apply Image and Calculations over Channel Mixer, because Apply Image and Calculation can be blended with Alpha channels, or with channels on different layers, or even with layers from a different document, if the document is the same size. You can even blend a cyan channel with a green channel from two different copies of the same document as long as the pixel dimensions match exactly.
When using Channel Mixer is good to keep Photoshop's Image>Grayscale calculation of 3 Red - 6 Green - - 1 Blue in mind, so that we can vary from that as needed by our friends and foes.
So lets make a B&W image of our serous rider and see what happens with the various tones. The first effort is the direct grayscale conversion so hated by the author via Image>Mode>Grayscale and the dialogue box - "Discard color information?" Answer yes and you see this sterling image.
To demonstrate the 3 red - 6 green - 1 blue ratio used by Photoshop for grayscale conversion, we are going to use Channel Mixer and set the ratios at 30 red, 60 green and 10 blue as in this dialogue box
[imgl]http://pathfinder.smugmug.com/photos/125666637-M.jpg[/imgl]
![http://pathfinder.smugmug.com/photos/125666637-M.jpg[/imgl]](http://pathfinder.smugmug.com/photos/125666637-M.jpg[/imgl]){kind=link}
and convert to monochrome. Amazingly enough, the image is an exact replica of the image we just created by direct grayscale conversion, as one might have guessed.
[imgr]http://pathfinder.smugmug.com/photos/125666665-S.jpg[/imgr]
![http://pathfinder.smugmug.com/photos/125666665-S.jpg[/imgr]](http://pathfinder.smugmug.com/photos/125666665-S.jpg[/imgr]){kind=link}
Now, lets see if we can do a little better job with the lowly Channel Mixer. Lets see if we can make his shirt black and keep the bike gray. Let's try Red 2 Green 108 and Blue - 2 as shown here.
Here is the image with a shirt almost as black as his pants and a gray bike and a white building with pretty good facial tones.
You do not need to limit one channel to l00 units - it is sometimes worthwhile to go way past these amounts. Pushing the Red channel up to 150+ can emulate a faux IR for example.
(further discussion continues in the next post)
Pathfinder - www.pathfinder.smugmug.com
Moderator of the Technique Forum and Finishing School on Dgrin
Moderator of the Technique Forum and Finishing School on Dgrin
0
Comments
Let us start with a color image, and select ( click) the green channel in the Channels palette. Next, we click Image>Apply Image and select the Red channel to be applied to the RGB Image or Background ( except we have already selected the Green channel in the Channels palette) and click OK, with Blending Mode Normal and at 100%.
If you now look at the red or the green channel, they will look identical if the amount was left at 100%. What we did with the Image>Apply Image command, is completely replace the selected Green channel with the Red channel via the Apply Image command.
If we now select the Blue channel and Image>Apply Image the Red channel to it, all three channels will be identical. You will have a grayscale image in which all three channels, R, G, and B, match the original RED channel grayscale image.
Well big deal, so what! You could have just saved the Red channel and discarded the Blue and Green and gotten to the same place!! The fun only begins when you begin to alter the opacity and/or change the blending modes from Normal, to Darken, or Lighten, or Multiply, or Overlay, or Softlite, etc, etc.
I cannot demonstrate all the possibilities here, thus the reader is encouraged to open an image of their own, and begin to play with Apply Image and alter the channels used and the blending modes to begin to get a feel for the power of this command. If you look at the color image created after an Image>Apply Image command, it can be funky, but Dan said early in the chapter that you can create a great B&W from a poor color image.
Next, Dan waltzes the reader through a 3 ring circus as he uses Image>Apply Image to create the B&W conversion of the the kids on the playground in front of the blue slide set. Kind of hard to follow at first, but then it gradually begins to make sense if you repeat his keystrokes at your computer.
The folly of the exercise is revealed when Dan says "the insidious purpose of this chapter is not to teach black and white at all. It is rather, a tease, an advance look at colorful things to come later" So, we have gone from the most important chapter of the book to a "tease", Never let it be said that Dan does not have a sense of humor!!
Next, he takes to a cemetery. I told you he has a sense of humor!! He finds an underexposed, noisy image to convert to B&W. (Why haven't we made any mention of what a good B&W should look like with real whites, deep blacks, and nice directional moody lighting? )
No - we're looking at an Italian tombstone with a noisy image from a P&S.
He tells us to make a second copy of the image and shift it to CMYK, because we are going to steal the Black (K) channel, and haul it back to the RGB image. So, in the first image CTRL-A selects all of it, CTRL-C copies it to the clipboard. Now, open a new empty document in PSCS2, and then CTRL-V pastes the copy of the first image into the empty second canvas. Image>Mode>CMYK converts the image to CMYK... Click on the Black channel in the Channel Pallette, and convert it to a custom CMYK profile as described in the text -- Edit> Convert to Profile>Custom CMYK and enter the value of 85 for the Black Ink Limit and you will see this dialogue box.
Now sharpen the black channel with USM to 500% Amount, Radius 2.0, Threshold 5 ( That is a whale of a lot of sharpening - but you can get away with that in the Black channel
Now, we again type CTRL-A to select the sharpened Black channel, CTRL-C to copy it to the clipboard, and switch to the RGB based first image. Click on the Create a New Channel button at the base of the Channels palette ( the button looks like a piece of paper with a corner folded back up )
and then CTRL-V to paste into the empty channel, as an alpha channel, the sharpened black channel copied from the CMYK image. Select the RED channel, and click on Image>Apply Image and multiply blend the alpha channel black image with the selected Red channel to get your B&W Image. Dan chose ther Red channel as it had the best detail in the shadows in his image of the tombstone.
Convert this image to grayscale and discard the Blue and Green channels and you have your black channel reinforced B&W conversion.
So lets see if we can do this together.
Here is a cat, I converted to B&W with this technique - a black channel convert to custom CMYK and then sharpened and multiply blended with the red channel of the RGB image and then discarding the blue and green channels to create this B&W image.
Here is the Kitty in RGB
We're going to copy it to a new blank canvas and convert it to CMYK to get the Black channel.
Here is the unedited Black (K) channel converted to grayscale for display here
Here is the K channel ( Black channel) after converting to the custom CMYK prolife
And after sharpening with USM Amount 500%, Radius 2, threshold 5
Now with CTRL-A and CTRL-C we copy the Black channel to the clipboard, and then paste -CTRL-V - it into the new alpha channel created in the RGB image.
Finally we click on the Red Channel to select it, and then click Image>Apply Image and enter Alpha channel as the channel to apply as a Multiply Blend with the Red channel and see this......( There is nothing special about the Red channel in this technique, we simply chose it because the eyes looked better in it than the Blue or Green channel )
I rather like this image. Notice how the yellow at the right side of the color image, no longer intrudes.
Dan at this point, introduces the idea that sometimes the green channel alone, without processing, can be a very good B&W, particularly of faces. To introduce this at this point, does indeed, give one a taste of his sense of humor
Moderator of the Technique Forum and Finishing School on Dgrin
interesting, succinct, and will be, at least to me, useful-
B&W conversion is all about choosing how contrast is going to work in an image. Blending the RGB channels (and sometimes the K channel) is the best tool here. Afterwards a curve (remember this is B&W so just one curve) will tweak and establish black and white points. But this chapter is about blending those 4 channels to build the contrast foundation for the rest of the workflow.
Last I talked to Dan he was having second thoughts about the ordering of the chapters of this book. It might be that the techniques of this chapter should have preceded the RGB curves of chapter 2. Blending for contrast enhancement is now an earlier step in his workflow.
"It is a magical time. I am reluctant to leave. Yet the shooting becomes more difficult, the path back grows black as it is without this last light. I don't do it anymore unless my husband is with me, as I am still afraid of the dark, smile.
This was truly last light, my legs were tired, my husband could no longer read and was anxious to leave, but the magic and I, we lingered........"
Ginger Jones
John,
Are you saying he does his blending BEFORE setting his Black and White points and his color balancing?
I only recently figured out that I do set the curves for individual RGB channels when I do my final color balancing with the Threshold command. I set a B & W point in the Threshold command, and then go to Levels ( yes I know it is not Curves but listen on ) and adjust the endpoint individually of the Red, Green, and Blue Channels to bring Each channel to my chosen B&W point settings. I just don't routinely alter the slope of the line I guess.
I was not really suggesting that Dan's original statement about the importance of Chapter 7 was inaccurate, just that he has a whimsical sense of humor. I do see the value of understanding and being able to use the 10 seperate channels. I did not dwell on setting B&W points and rewriting a curve because that was understood as final fine tuning of any image.
Moderator of the Technique Forum and Finishing School on Dgrin
"It is a magical time. I am reluctant to leave. Yet the shooting becomes more difficult, the path back grows black as it is without this last light. I don't do it anymore unless my husband is with me, as I am still afraid of the dark, smile.
This was truly last light, my legs were tired, my husband could no longer read and was anxious to leave, but the magic and I, we lingered........"
Ginger Jones
Open Levels and switch from RGB to the red channel. Now slide the left upper red slider in while watching the red values in the INFO palette that were picked with the Threshold command as your black point until they match. Next slide the right upper red levels slider to the left while watching the white point values in the Info palette of your white point until they match. Repeat theses steps for the Green and Blue channels also. As you slide the sliders watch the changing values in the info palette, and make them match the values in the same channel in the respective b&W points. When all three channels have been set to match the threshold B&W point values, click ok and you are done. I prefer to do in on a layer, and can then adjust the opacity slider to taste. It is important that the white point in thresholds is a real white, not a specular reflection. If it is a specular reflection, it will likely read 255,255,255.
You can do a similar thing with the curves command sliders also.
Moderator of the Technique Forum and Finishing School on Dgrin
I'm not sure why a copy is needed when you have the original as the background layer. That seems like a wasted step to me. Also, even after getting the odd colorcasted images, when I converted that to grayscale, I got something that looked very similar to the one he ended up with. Things just aren't moving along like they should be according to the book.
Finally, if I used the Calculations instead of apply image command, I get a B/W. I assume that is because it takes it to it's own channel or document instead of putting it on the channel of the target channel.
Anyways, any help would be cool cause I'm having one of those issues where this should be easy but I am just stumped on it for some reason.
FYI, this is very similar to what the auto-levels command does except that it chooses the white points and black points automatically.
This technique can work well if your black and white points are indeed supposed to be neutral. But, auto-levels doesn't work at all (causes horrible color shifts) if either the darkest point or the brightest point are not supposed to be neutral and should have some color because auto-levels will make them neutral causing an undesirable color shift.
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
That is why I do it in an adjustment layer, so that I can adjust the opacity slider if I find a cast present after balancing this way. The balancing of color images without any neutrals, or known colors can be challenging for me ( and lots of other folks I suspect )
If you know of a fast, easy, reliable technique to color balance images without any neutrals, or known colors, I would be delighted to learn about it, John.
Moderator of the Technique Forum and Finishing School on Dgrin
I didn't mean my comments as a criticism of your technique at all. It makes a lot of sense to me. I just thought I'd explain to the other readers how auto-levels works and how it's similar to what you do manually (without nearly as much control obviously).
I agree, color balancing without known neutrals is hard. I'm slowly trying to absorb Dan's logic that he uses. This subject X can't be more blue than red and this subject Y can't be more A than B so therefore or subject X has to be negative in B, so therefore I must need to apply this kind of curve to fix the color.
For skin tone when there aren't other known neutrals, I'm getting a lot of use out of looking at the CMY values and evaluating them per http://www.smugmug.com/help/skin-tone to the point where I wish I could see CMYK values in my RAW editor (ACR).
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
The reason that I posted the way I do levels after THRESHOLD, is the way I was taugtht in the EPSON online print academy, but it was taught verbatim, and I never really thought about what it represented until recently, when I realized it was doing individual "curves" for each channel, kind of like Dan was teaching earlier in the book.
Moderator of the Technique Forum and Finishing School on Dgrin
I have now put up a page here:
http://members.ozemail.com.au/~binaryfx/ (frames, go to How To, Image Enhancement, Enhance Luminance Value Separation in Primary & Secondary Hues).
http://members.ozemail.com.au/~binaryfx/howto_lumabw.html (direct frames free link)
Basically, one can use the Photoshop CS3 beta's new Black and White image adjustment command in luminosity blend mode, so that the image does not go monotone and luminosity variation is increased in the full colour image (this new command streamlines the older technique made famous by Russell Brown for creating monotone files using two Hue/Saturation adjustment layers and blending modes).
This is different to Dan's more analytical and preplanned channel blending approach and is more interactive.
Regards,
Stephen Marsh.
http://members.ozemail.com.au/~binaryfx/
http://prepression.blogspot.com/
Duffy
The chart from p. 98 in Chapter 4 is a helpful guide.
—Korzybski
In case anyone is curious, this is due to the vagaries of the human visual system - it's much more sensitive to green than to red, and in turn much more sensitive to red than to blue. These ratios are generally considered the most accurate conversion of the perceived luminosity of a colour image to a monochrome - which is not to say that will make an interesting B+W, as demonstrated in the book.
Thanks for all the work on these write-ups, guys, I will definitely be trying the techniques from this chapter and posting my results here. Currently my tiny brain is exploding - I just finished the LAB book and now this!
"Everything has its beauty, but not everyone sees it." - Confucius
DO you think that the human eye is more sensitive to green and red, rather than blue, because plants in a forest or jungle tend to be green, and predators are reddish ( brown is a red color ) and not blue? Interesting really, since blue photons are more energetic and should be the easiest to "see", rather than the lower energy red and green ones. Evolution at work here??
Moderator of the Technique Forum and Finishing School on Dgrin
I must confess I hadn't actually thought about it, but that makes sense. In the LAB book Dan talks in one of the earlier chapter extra sections about the evolutionary advantage our better colour perception gives us.
Of course, I'm (slightly) colour blind, so it's wasted on me - I'd be tiger food in no time
"Everything has its beauty, but not everyone sees it." - Confucius
To add to the wondering, the maximum sensitivity of human vision is at 555 nm, green, close to the wavelength at which light from the Sun has its maximum intensity. Light is detected---"seeing" occurs---when the frequency of the oscillating electric field of the light matches an oscillation frequency of the electron clouds in the photoreceptors of the eye. That is, the key to "seeing" is the *matching* of light and matter (electron cloud) frequencies, rather than the frequency (photon energy) of the light itself.
The central oscillation frequencies of the red and green photorecptor charge clouds are relatively close to one another, which may be why yellow takes up such a relatively small portion of of the visible spectrum. The central oscillation frequency of the blue photorecptor is near the limit of short wavelength (high frequency) human color sensitivity, so that only a portion of the sensitivty of the blue photoreceptor is in the visible range.
Two asides, and I'll stop:
First, having reached a "certain age," I find blue letters on a black background Web page quite hard to read. This is because the lens of eye yellows with age. I always think of this when Dan Margulis points out that "yellow kills blue." The yellow appearance of the lens means it is absorbing blue, and so *less* blue gets through to the photoreceptors.
Second, I chuckled when I learned that human color vision seems to be implemented on the LAB model! Two other links I have found helpful are Light and the eye and Color vision.
"It is a magical time. I am reluctant to leave. Yet the shooting becomes more difficult, the path back grows black as it is without this last light. I don't do it anymore unless my husband is with me, as I am still afraid of the dark, smile.
This was truly last light, my legs were tired, my husband could no longer read and was anxious to leave, but the magic and I, we lingered........"
Ginger Jones
"Everything has its beauty, but not everyone sees it." - Confucius
As it happens, in today's NYTimes I read the following:
How do we see red? Let me count the ways
"In fact, human eyes, like those of other great apes, seem to be all-around fabulous fruit-finding devices, for they are more richly endowed with the two cone types set to red and yellow wavelengths than with those sensitive to short, blue-tinged light. That cone apportionment allows us to discriminate among subtle differences in fruit ruddiness and hence readiness, and may also explain why I have at least 40 lipsticks that I never wear compared with only three blue eye shadows."
"It is a magical time. I am reluctant to leave. Yet the shooting becomes more difficult, the path back grows black as it is without this last light. I don't do it anymore unless my husband is with me, as I am still afraid of the dark, smile.
This was truly last light, my legs were tired, my husband could no longer read and was anxious to leave, but the magic and I, we lingered........"
Ginger Jones