Converting Digital Photos to Black and White Masterpieces

Part 1 of 2 Part Series

The Source  Image

The Source Image

I’ve been writing these tutorials for a few posts now and a lot of time the inspiration for a post comes from a question that may have been asked in the Raleigh Flickr groups or something I’ve seen while traveling. This tutorial comes to you courtesy of @dclay3521 who asked for some help with Black and White conversions. She sent me a photo that she wanted some help with. I will be using that photo for the tutorial.

In thinking about the request, I have a choice, come up with a short and direct tutorial to address this one issue, or write a (much) longer post on the broader topic. I’ve opted for the later because for B&W conversions there is no right way just as there is no “right” B&W film. Depending on your goal one technique might work better than another, so lets throw them out there.

Now that said, there are an umpteen ways to do B&W conversions. Photoshop is notorious for giving you multiple ways to get things done. You of course can convert in other programs besides Photoshop and these will give you even more options. You can buy or download free Photoshop actions to do conversions. I’m going to keep this to things you can do with Photoshop. Photoshop Elements and The GIMP might not have all these choices and workflow tools like LightRooom and Aperture certainly do not have all of these. I’m going to talk about six different methods ranging from easy to multiple step.

First though we need to understand the “innards” of our digital camera files.

All of our cameras shoot in whats known as RGB mode. That is there are three separate arrays data, or “Channels” that hold lightness values (luminance) ranging from 0 (no light) to 255 (maximum light) for JPEGs for each pixel in each channel. RAW files have wider ranges, for instance a 12 bit camera has a range of 0 to 4096 for each pixel per channel. Since we tend to end up with JPEGs that we show on flickr, I’m going to focus on 8 bit JPEG’s which are in the 0 to 255 range. I’m also going to use a 6 megapixel camera as an example since they produce an array of pixels 3000 wide by 2000 high.

Your RGB image file is therefore a red channel of 3000 x 2000 values from 0 to 255 depending on how bright that pixel is. The green channel is a 3000 x 2000 grid of values from 0 to 255 and a blue channel. If you’re photographing a rose for instance, some of the pixels will be Red. The values in the red channel will be higher where the photo is red, but the blue channel for the same pixels will be low numbers since there is no blue color in the red petals. The computer reads each of the red, green and blue pixel values and generates a pixel on the screen that is the combination of the three color values. 0,0,0 will be black, 255,255,255 will be white and 255,0,0 will be a bright red (reading the numbers in the standard Red Green Blue or RGB). This gives us over 16 million possible colors that we can show.

But RGB is only one way of organizing color data in our image files. Another common way frequently used in printing, like magazines, newspapers or printer shots is known as CMYK or Cyan Magenta Yellow and Black. These image files work like the RGB files where you have values from 0 to 255 per pixel per channel. The Cyan, Yellow and Magenta combine together to give a wide range of color and the black is used to control the overall brightness.

Converting to Lab Mode

Converting to Lab Mode

The final other frequent method of storing our color data is what is known as HSL or Hue, Saturation and Luminance. Photoshop has a mode called Lab which is Luminance, A channel and B channel which behaves like HSL.

So what does this mean to B&W conversions?

Well each individual channel of a photo itself is just a range of color brightnesses. You can use Photoshop to get rid of color channels your not interested in and keep the one layer that gives you the B&W image you want.

Lab Mode or Luminance Channel


Starting with Lab mode, if you kill the “a” and “b” channels which hold the saturation and hue values, your left with the “L” channel which is the luminance values, or the pure brightness values with no implied color. To do this, you need to convert the image to Lab mode by choosing Image->Mode->Lab. The select the “Channels” tab on the Layers/Channel/Path pallet. This will show you the combined Lab image, and each channel L, a and B. Right click (or Ctrl-Click on a one button Mac) on each of the “a” and “b” channels and choose delete from the flyout menu. You will note the channel names change to Alpha1 and Alpha2. You want to next delete Alpha2 so your left with the original Luminance channel. Photoshop should now show you your converted B&W. But its no longer a file that can be shown on a web site so you need to convert it back to RGB.

To do that, click on Image->Mode and you will see “Multichannel” selected. Convert it to Greyscale, which copies the one channel to two other identical channels and then repeat the process and change the mode to RGB. Now you can save it as a JPEG and you will have a pure B&W image.

Now I could point out, a quick way to basically accomplish this is to do an Image->Mode->Grayscale and then convert it back to RGB. However the straight up Grayscale conversion will try and average out the pixel values giving you a slightly different look.

Other Color Channel methods.

If you use RGB or CMYK you can look at each individual channel and decide if any of them fits your look. For CMYK, you do an Image->Mode->CMYK to get into that mode. For RGB, thats the native mode so you don’t have to do any mode conversions. For CMYK, the Black channel gives a nice B&W image in most cases, but you might want to look at the other channels before deleting them.

With RGB, the Green channel is usually your best bet, but if we look at the Rose example, which is mostly red and green and almost no blue, you will find the red channel has bright nearly white petals where the green channel will show dark petals. But if you use these methods, find your channel, delete the rest, convert to grayscale and then back to RGB.

That covers the individual channel methods. But we are not over.

Hue, Saturation and Lightness adjustment.

photoshopscreensnapz004Another simple way to do a B&W conversion is to use the HSL adjustment (Ctrl-U or Apple-U) tool and slide the saturation slider to the far left, leaving the Hue and Lightness channels to make up the resulting B&W image. This method can also be done using a layer adjustment if you prefer to use layer adjustments. Many people prefer to use layer adjustments since it leaves the original color layer in tack. This is the preferred way since it gives you more flexibility and the ability to back out the changes as well as use layer masks for selective coloring. But just desaturating the source image is quick.

photoshopscreensnapz005There are two other methods commonly used, the Channel Mixer and the B&W Adjustment Layer. The Channel Mixer is a tool that lets you adjust how much each color channel affects the overall image. Normally adjusting the sliders works on a color image. There is a checkbox labeled “Monochrome” that when checked, converts the image to B&W and then the sliders let you select black and white tones based on each channel. The Channel Mixer is probably the hardest to master but can give you some of the best results.

The final method is the B&W Adjustment Layer. Like other adjustment layers, this gives you an editable adjustment layer that can use layer masks for selective adjustments, opacity adjustments for partial saturations and as long as the layer is saved, in say a PSD file, you can come back later and continue to make adjustments.

The B&W Adjustment layer is similar to the Channel Mixer, but you have different sliders that affect smaller ranges of colors, like yellows, oranges, greens, cyans, etc. It also has some pre-set filters which mimic traditional photo filters used in B&W settings. This gives you the widest range of controls on your B&W conversion. With that many choices, it will take time to master it.

differenttechniquesYou can see that you have a multitude of options when converting to grayscale. While there isn’t that much of a difference in the subjects face and hair, notice that almost all them are different in how the sweater is displayed.

In the next post, I will cover taking these base conversions and applying the necessary post processing to make the photos stand out.

Get a Trackback link

No Comments Yet

You can be the first to comment!

Leave a comment