Sharpening in Photoshop – The Complete Guide
Sharpening is one of those things in post-processing that makes a lot of people scratch their heads. There are multiple sharpen tools in Photoshop, some additional techniques that work without these tools, and a range of sharpening plugins. So, how do these methods work and which one is right for you or your specific image?
In this article, I am sharing a video by Kelvin Pimont in which he explains 6 different sharpening approaches that you can apply without any plugins. Then, we’ll take a look at a method that I call Smarter Sharpen. This technique lets you make any sharpening tool non-destructive which can really boost the effectiveness of your sharpening work.
- Unsharp Mask is probably the most well-known sharpening method. In essence, it is a destructive tool that changes the actual pixels of your image. But when applied to a smart object, it can be adjusted after you applied it. However, you typically need to put in some work to hide the sharpening from areas that do not need it or that show a lot of noise.
- Sharpen in Camera Raw is available since the Camera Raw Filter was introduced in Photoshop CC. As opposed to the old way of using Camera Raw only at the very start of your workflow when you imported your raw image into Photoshop, the filter can be applied at any stage of your workflow. That is a big advantage since sharpening should typically be done at the end.
- Sharpen Tool The sharpen tool comes as a brush that can be used to apply sharpening directly to the pixels of your image. The more you brush on one area, the more it will be sharpened. However, the settings are very limited.
- High-Pass Sharpen is the go-to non-destructive sharpening method for most people. By applying Photoshop’s high-pass filter to a copy of your image layer and setting that layer to Overlay blending mode, you can add a sharpening layer without directly changing the pixels of your image. This lets you control the strength of the effect at any time.
- Lab mode Sharpening is a more exotic method that not a lot of people use or even know about. By changing the color mode of your image from RGB to Lab you get access to the dedicated Lightness channel. You can sharpen this channel and, in this way, avoid affecting the colors.
- Limiting sharpening to the midtones: Often the shadows and highlights of an image are the areas where the noise is and most edges are in the midtones. By using the Blend-If sliders in the Layer Style dialog, you can blend any sharpening layer (irrespective of the tools you’ve been using to create it) with the original image layer to limit the sharpening to the midtones only. In fact, you can control exactly which range of tones you apply the sharpening too.
- Combining different methods: You’re not limited to using one method alone. If you find that different techniques work differently on different areas of your image, you can stack multiple sharpening layers on top of each other and use layer masks to apply the sharpening selectively. You can even double-sharpen areas of your image by applying, for example, Unsharp Mask to the image and then sharpen that again using the High-Pass method. This can increase the sharpening effect without introducing the artifacts that typically come in if you apply too much of a single sharpening method.
In this video, I’ll introduce you to a more advanced way of sharpening that builds on all the above methods and makes them non-destructive. I’ll also explain how to use Photoshop’s Smart Sharpen tool.
I will show you a way to put any sharpening effect on its own layer to blend it with your image layer and change it’s strength after the fact.
Some Final Tips for Sharpening
Irrespective of the specific method you’re using for sharpening your images, there are a few general tips that will help you sharpen more effectively:
- Always look at your images at 100% magnification when you’re sharpening them. This way, you’ll be able to see what you’re doing, and Photoshop shows you the actual pixels instead of merging them together to show you a zoomed-out version of your image.
- Always do noise reduction first and sharpen later. By doing this, you can avoid sharpening the noise in your images and you give your noise reduction software the original noise to work with (not the sharpened noise), which allows it to be more effective.
- Always sharpen towards the end of your workflow. Sharpening and the different adjustments and plugins you may apply, do interfere with each other. So, you should apply all these adjustments first. If the image is not sharp enough once you’ve done so, apply the necessary amount of sharpening right before you export your image.
- Don’t let any software apply the sharpening automatically for you upon importing the photo or as you export the images. This would take the control out of your hands, and you may ruin the image right at the start or at the very end.
If you have questions about sharpening, feel free to leave a comment below. What is your preferred method? Have I left anything out here? Let me know!
Klaus, you suggest always sharpening toward the end of the workflow. Are you advocating a 1-step process rather than the 3-step method that’s generally considered best practice? If so, why?
You also suggest no sharpening should be automated; however, in a 3-step process the Capture and Output sharpening can be automated.
well, if the photo is soft out of the camera, some input sharpening may help. But as I said, the interactions between that sharpening and the adjustments and plugins you apply are hard to predict. So, input sharpening may also be detrimental to the final image quality.
Creative sharpening (step 2) really depends on the image. It does not make sense in general.
Output sharpening (step 3) is adequate in the overwhelming majority of cases, though. And that should be done at the very end.
So, you see, the idea that you always have three steps in sharpening falls apart when you look at it more closely. It all depends on the image and the type of processing you apply.
As for the automatic sharpening, I am not fond of having software do anything automatically to my images. You never know whether it does a good job or not. So, you’d have to go in and check it anyway and that goes against the idea of “automatic”.
I hope that makes it clearer. Let me know what you think.
By automated, I’m thinking along the lines of capture/input shatpening based on camera/lens and that the same settings can be applied to all pictures on import, or via sync. Similarly output sharpening for glossy paper, or matte, paper should be the same for a particular paper, so if a series of prints is being made, the same settings can be applied. I’m not thinking of one-click settings that some plug-ins, or actions offer. I agree that’s not optimal.
I would disagree on capture shatpening. Any Bayer sensor camera needs some capture shatpening due to the demosaicing process. Any camera with an AA filter also needs some capture sharpening. If applied as capture shatpening should be – to essentially reverse the effects of demosaicing/AA filters – shouldn’t interfere with other editing steps because it’s a very minimal amount of sharpening. If this first step is overdone – as a lot of people do – then yes the effect of it can be exacerbated and look unpleasant.
Creative sharpening is image dependent, so some may not need it. I think it can still be considered a 2nd stage even if the decision is to do nothing.
For what they’re worth, those are my thoughts.