Easy-to-follow video tutorials help you learn software, creative, and business skills.Become a member
In this exercise we are going to take a break, a brief break from the topic of blurring, and we are going to take a look at averaging. The reason being that the next most important blurring and averaging filters happened to be averaging filters as opposed to blurring filters. Blurring is creating soft transitions between the luminance levels associated with two neighboring pixels, whereas averaging is actually marrying those luminance levels. It's actually finding the average between the two and coloring those two pixels with the same luminance level. And so what the practical upshot is instead of creating soft feathering effect like we are seeing here, inside of this image which I have gone ahead and saved my progress by the way, rather than creating that sort of feathered effect there, we are going to create a smoothing effect. We are going to round off corners inside of our images.
All right, so what I'm going to do here inside this image is I'm going to shift down to the bottom of the image by pressing the End key so that we can see the two bottom layers, 3 BL and 4 BR. And I'm going to click on 3 BL, and then we are going to go to the location of the averaging functions, which is under the Filter menu right here under the Noise sub-menu. So you have this command Add Noise which allows you to add random pixel variations, if you are so inclined and all you do in that case is you choose this command and then you enter the amount of noise that you want to add to the image, and typically you are going to use this command in order to match noise, that's already there. And if that's your goal, then you want to work with some very small values, something around 3-4%. And most likely, you are going to go with monochromatic noise instead of the colorful noise, which is going to basically be very distracting inside of a standard image.
Uniform versus Gaussian distribution that modifies the degree, the intensity of the noise that you are applying. So if you want uniform noise meaning the same amount of noise throughout the image then you would stick with Uniform. If you want a Gaussian distribution of noise, that is, you want more black and white noise then noise in between and then gray noise, then you would go ahead and select this and you are going to get a higher impact effect from Gaussian noise than you are from Uniform. So that's totally up to you, but once again if you are trying to match some sort of digital noise that's associated with a photographic image then a low Amount value of 3-4% typically, either Uniform or Gaussian distribution just experiment with that and Monochromatic turned on most likely. All right, I'm going to cancel out of this because I don't want to add noise.
To average the pixels to get rid of noise, you come down to these other four commands right here each of which dispenses with noise in a different manner. Now Reduce Noise is the most powerful of the commands. I'm not going to show it to you in an analytical setting here. I'm not going to show it to you with the angry blocks. I'll show it to you in the context of an actual photographic image later in this chapter. Then we have Median that averages neighboring pixels. We have Dust & Scratches which fixes neither it's a variation on Median that we'll see in a moment and we have Despeckle that gets rid of the low level, low intensity noise detail.
Let's start with Median. Because Median just basically smoothes off everything, and notice that I have gone ahead and given you a keyboard shortcut, if you loaded Deke keys, because this is a very important command in the software. And I'm going to go ahead and raise this command by pressing the Up Arrow key. Notice as I do, I'm rounding off the corners in these rectangles here, both on the interior rectangle and the exterior rectangle. So it's modifying both the pixels inside the layer and the transparency mask that's associated with the layer as well, and as I continue to raise this value, you can see sort of different things happen with the teeth. I think this is pretty interesting. If I set this value to something lower, I'm going to see the dark lines between the teeth and as I raise this value those dark lines start to disappear. And then if I take the value pretty darn high, they start to come back, notice that. So it's a pretty interesting behavior that you get out of these averaging filters.
In the whole time, you still get hard edges around your various discreet objects. So you don't get that blurriness that you get with Gaussian Blur and Box Blur and so on, you get sharp edges. In fact, here is a Radius of 16 pixels so we're matching what we applied with Gaussian Blur and Box Blur. I'm going to go ahead and click OK. These are some pretty jagged transitions if you take a close look here at this cornered edge right there, this rounded corner. We don't have any anti-aliasing at all. We just have sharp jagged pixels.
All right, anyway, I'm going to press Ctrl+1 or Command+1 on the Mac to return to the 100% view. How does that compare to something of a Threshold setting. So in other words, let's say that you don't want to median everything, you don't want to average everything inside of the layer. You want to average either the very similar pixels or you want to average the very different pixels. Well, that's when you take on Despeckle or Dust & Scratches as I'll explain in the next exercise.
Get unlimited access to all courses for just $25/month.Become a member
120 Video lessons · 63315 Viewers
119 Video lessons · 71008 Viewers
125 Video lessons · 31970 Viewers
84 Video lessons · 18610 Viewers
Access exercise files from a button right under the course name.
Search within course videos and transcripts, and jump right to the results.
Remove icons showing you already watched videos if you want to start over.
Make the video wide, narrow, full-screen, or pop the player out of the page into its own window.
Click on text in the transcript to jump to that spot in the video. As the video plays, the relevant spot in the transcript will be highlighted.