Advanced Hair Selection Using Channels in Photoshop

Masking out images with complex edges from their background in Photoshop can be daunting for some designers and Photoshop users — especially working on images with very detailed edges and textures that even Photoshop’s pen tool, at its finest, can’t handle. Good thing Photoshop never ran out of ways to make things easier for us.

In this short tutorial, I will show you an old but reliable method of masking out complex images using Photoshop’s channels. This is a great tip for all the newbies and even for intermediate users out there so heads up!
Having said that, let’s start our tutorial on how to make an advanced selection using channels in Photoshop. For this tutorial, I will be using this stock photo from Shutterstock.

Step 1 — Open our stock photo in Photoshop. We will mask out the model from its background using Photoshop’s channels. This method is more effective on images with uniform tone or color and a clean background . So images taken from an indoor studio are the best samples. It is important to analyze the differences of your main subject and background in terms of light and dark parts, colors, and contrast; these factors will play an important role when creating a selection using channels.




Step 2 — We will start creating a selection using our channels. So with that go to the channel’s tab. Then click each channel one by one and look for which channel has the maximum amount of contrast. The channel that has the greatest amount of contrast—particularly between the hair strands and the background—will work the best for this one. Now in this particular sample, I choose the Blue channel for it’s great contrast particularly on the hair stands although the Red channel is a strong candidate if not only for it’s whiter skin tone. Later in this tutorial, you will understand why we have to avoid such in this instance.



Step 3 — Now we have to duplicate the Blue channel by dragging it down to the “create new channel” button. Then change the name to “mask.” We don’t want to work directly on the original channels because any modifications on it might cause us some unwanted results later.



Step 4 — Now basically our aim here is to make this image purely black and white as much as possible. We will invert our selection later so better think that all the black parts of the image will be the opaque parts and all the white parts will be transparent. So we will start this by strengthening the contrast in this channel. To do that go to IMAGE > ADJUSTMENTS > LEVELS or hit (Ctrl + L) to open up our levels window. Now move the dark point slider towards right to darken parts of the hair. Then move the light point slider slightly towards left to lighten up all the grey tone areas. You have to play with the point slider back and forth to get the right settings for this one (see ideal settings below).



Step 5 — Now we have to fill in the whole body of our model with black to include this on our selection. This is why we didn’t choose the RED channel earlier because it’s skin tone is almost white. The skin will disappear after the adjustments in levels and we will have a hard time recovering those details. Unlike in the Blue channel where details are still visible after some adjustments in levels because it has a darker skin tone.

So now grab the BRUSH TOOL and set the foreground color to Black then paint and fill in our whole model with black. Be careful when painting near our hair strands. Adjust your brush size when necessary. At this point we have to grab the DODGE TOOL and set the range to Highlights and exposure to around 25%. Paint on the hair strands to eliminate all the gray tones around the hair thus making it cleaner. Look closely while painting on thin hair strands and avoid diminishing the details too much.



Step 6 — Now that we have a fined black-and-white image. We can now activate the selection by holding (Ctrl) then clicking on the mask’s thumbnail. Running lines should be visible by now.



Step 7 — Now go back to our layers panel. we will create a mask for the selection. While the selection is active, hold (Alt) then click on the “new layer mask” button. This will invert the selection while making a mask out of it. And just like that, we just extracted our model from its background.



Step 8 — Now we can add more volume on the hair strands by duplicating the layer at least three times. To do that, select the layer then press (Ctrl + J) three times. Notice that the volume of the hair strands are gradually increasing each time you duplicate the layer. You can merge all the three layers by selecting all the layers and hitting (Ctrl + E).




Step 9 — Now for fine tuning, we have to adjust the color of the hair strands for it to blend well, especially when we place the new background later. To do this, create a new layer above our model and name it “edge.” Change the layer’s blending mode to COLOR. Pick up the brush tool and sample a color of the hair by holding (Alt) and clicking somewhere on the hair. Now start coloring the outer parts of the hair. Remember you can adjust the opacity so you can control it. After that, hit (Ctrl + U) to access the Hue/Saturation window. Now adjust the saturation and hue if further modifications are needed (see ideal setting below).




Step 10 — And lastly place a new background behind our model. You can erase some hair strands to clean it up a bit. Notice how detailed the hair is opposed to using the pen tool for selection. Not to mention the amount of time and effort you have to exert in order to have a selection like this using pen tool.




There you have it! An easy method for extracting images out of their background. This method can also be used—not only to when dealing with hair selection—but anything that is just too complex for the pen tool to handle. Hopefully this will help in making your Photoshop sessions much easier next time around.

Source: http://www.youthedesigner.com