In previous chapters we saw how to apply focus to a picture using two methods distinct High Pass Filter and Unsharp Mask. These methods will in most cases have a small problem but we will try to avoid the steps that we see today.
What is the problem? By focusing on a picture will give more contrast to certain areas of the picture (particularly the edges), but we also amplified all points (pixels) that differ from their surroundings. In Castilian, the focus we have as a side effect that amplify the noise to be in the picture. In many cases it will not even noticeable and in many other cases we simply put a mask for areas that we do not care focus.
But what do we do when we can not avoid the problem? There are at least two ways, the first (which we see today) is to apply the approach to LAB mode and the second (discussed next) is to avoid channels with noise.
To begin, we must know that is how LAB. Working with a digital photo f there are many RULES
represent the colors. The most common are RGB and CMYK, the first being the most common use. RGB mode represents colors as the sum Red, Green and Blue (Red, Green and Blue) and how the monitors work. CMYK is a way
largely used by offset printing and represents the color and subtraction between Cyan, Magenta, Yellow and Key (Black). How does LAB mode then? This mode is different, to represent an image is separated into 3 channels: brightness, channel A and channel B. The great advantage of this mode is that it is indeopendiente the device that shows the image that the other two. Another great advantage to use in Photoshop is that we can work with the image without changing color. had previously said that in applying the approach also amplify the noise of the picture and that was because changing the contrast also modificábamos color pixels. LAB mode can change leaving intact the color contrast by working only on the lightness channel.
Let us see how we do all this. Suppose we already have our final image but you need that approach. The first is to move the photo to LAB mode, for that go to the menu Image / Mode / Color Lab
If our picture is Photoshop adjustment layers will not report that the change will eliminate the layers (apply the changes but remove layers). If you do not want to lose our original work we create a new layer above all the others and paste a copy of the final image (CTRL + ALT + SHIFT + E, almost nothing). With this new layer we can copy it to a new document or even drag it to another document.
Once the change (do not worry, we can go back to RGB at any time, the only thing we can not regain are Photoshop adjustment layers removed), we can see the difference in the Channels tab (if can not see it go to Window / Channel). In a normal photo will see these channels:
Moving to see how these other LAB:
To begin, we must ensure that work only on the lightness channel. To achieve this we must click on the 'eyes' that is to the left of the channels 'a' and 'b' (the 'eye' Lab channel will hide in one of the 3 channels). We will be something like:
Currently could use the Unsharp Mask filter as we saw in the previous chapter . If we did the result would be little better than applying unsharp masking in RGB mode, but as we have taken all this work the least we want is that the result is worth it. The trick is to avoid as much as possible the areas of greatest noise and apply the approach only the parties to better serve us. So the problem we face: we are seeing only the lightness channel and we want to apply a filter to only most suitable locations. How do we do?
is easy (mostly because I did not discover but I read in the book Photoshop Channels Book by Scott Kelby). The first thing we do is let down CTRL and click on the icon of Lightness channel (which is just to the left of the word brightness). This will create a selection based on the image of this channel (the lighter parts are more obscure and less selected). Next step is to invert the selection (CTRL + SHIFT + I). And now we can apply the filter because we have selected only the Lightness channel and areas of the image that will serve to focus.
Scott Kelby in his book proposes to use the Unsharp Mask in two very aggressive steps (taking advantage of this method to amplify less noise). My truth is that these two steps never gave me good result, as soon as be able to manage their comment.
Consider this method in action, the original picture is this:
I made two copies, one I applied Unsharp Mask directory and the other followed the recipe: I spent
- way Lab I hid all channels
- least
- Brightness CTRL + click on the Lightness channel icon (or whatever it is equal CTRL + ALT + 1)
- Invert the selection with CTRL + SHIFT + I (or menu Selection / Invert)
- I apply Unsharp Mask (the values \u200b\u200bwere elejí Amount 200%, Radio 0.8 and Threshold 2)
- Redisplay all channels to see the whole image.
The window and wall stones
is clear that if this method was so very much better than anything anyone had bothered to create so many other ways of looking. I hope you will note in the example above, applying the same amount focus to the regular photo (in RGB say) we have sobreenfocada, while the use LAB mode was perfect. In pictures where we see we need to apply a lot of focus but in doing so we generated halos or brings a lot of noise we can apply these steps.
0 comments:
Post a Comment