India Point Park

India Point Park
by matt-ball                

Submit your Photo
Hall of Fame

Please participate in Meta
and help us grow.

Photography Stack Exchange is a question and answer site for professional, enthusiast and amateur photographers. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

What (in your opinion) is the optimum order of doing the following to a scanned image in PhotoShop?

  • Cropping
  • Manually adjust Levels (RGB and density)
  • Manually adjust Curves
  • Manually adjust brightness and contrast
  • Manually adjust Color Saturation
  • Automatically remove dust and scratches (filter)
  • Applying Unsharpen Mask to sharpen

I edit photos at work, and although I've found an order that works for me, I wonder which order would be the best. Obviously I don't do everything to all pictures.

share|improve this question
one important thing - if you plan to resize your images: first do that and than sharpen. Sharpening before resizing isn't a particularly good idea. – MarcinWolny Apr 6 '13 at 12:10
up vote 1 down vote accepted

If you're using a 'destructive' workflow, one in which each step changes the pixels--Photoshop's default way of working--order of operations can have a strong impact on the result:

1) Getting it right at the source is key--do a high-bit depth scan, use digital ICE if available.

1b) Capture in the lowest gamma, broadest color space you can (ie. AdobeRGB or better yet ProPhoto RGB), convert to it now if you can't.

2) Some applications (or plugins for Photoshop) allow you to do something called "capture sharpening", often implemented as Richardson-Lucy deconvolution (sounds scarier than it is to use). Done correctly, deconvolution is actually restoring high frequency information (detail) to your file (it's providing real information, not the illusion of detail as unsharp mask does). You may find that a capture sharpen of your scans can help you get a sharper result, without any halos, crunchiness, or eye fatigue from the resulting work.

3) Do a proper (correlated color temperature) white balance of your image (if no true white balance tool is available, then color balance as part of step 4). You'll have a dedicated tool labeled white balance, grey balance or neutral picker, if you have one. Interestingly, Photoshop does not have a white balance tool, but Adobe Camera Raw does. You can get your scan into ACR with some contortions, if you wish to.

4) Use curves to do your color balance, gamma correction, level correction, exposure, contrast, and tone mapping. Order of these won't matter as all will become a single curve operation.

5) Apply your creative effects--saturation, sharpening, vignetting, et. al. The fewer "commits" the better, for image quality.

6) Optimize your content for output; web version, print version, digital display versions of the same file will all be prepped differently. Don't forget to change to an appropriate color space for presentation. If you are interpolating (upsizing or downsizing), use bicubic or better (bicubic is Photoshop's best so that's a great starting place).

Sharpening is a mystery to most of us, but when done correctly is invisible. Worth a read: (no affiliation).

HTH, -Brad

share|improve this answer

If your scanner has a feature to remove dust and scratches, activate that and use it while scanning. This feature is sometimes called "Digital ICE" (a Kodak brand name), and it uses an extra infrared scan pass to identify and remove dust. This is very effective; much more so than doing it after the fact can be.

If dust and other artifacts remain, I'd suggest doing any automatic filtering as the first step, and any manual touch-up work as the last before unsharp mask.

The other thing I notice is some redundancy in your workflow. Levels is really just a more primitive interface to the Curves tool; you can make that same adjustment there. And brightness and contrast are both even more primitive interfaces to the same thing: there should be no need to make these separate steps. So, I think you can probably save yourself a lot of effort by learning to to use Curves a little more effectively.

share|improve this answer

I would suggest starting with a white balance first as that will impact all the rest of the color related steps. Then most of the other steps can be done straight from curves. Curves combines levels and brightness/contrast. Then I normally adjust saturation after curves, then crop, then sharpen mask after everything else is done.

I only do dust and scratches if it is needed after any scanner functionality is used and then I make a case by case determination if it is better before or after color work, but prior to sharpening or cropping (based on if I think it will be easier to deal with before or after the touch ups are applied to color).

share|improve this answer
Curves also covers white balance adjustment, although if your software doesn't have a nice way to use that and an eyedropper to pick a neutral tone, a different interface can be easier. – mattdm Apr 5 '13 at 13:34
@mattdm Oh right, I forgot about that since I do all my white balancing in LR now. – AJ Henderson Apr 5 '13 at 13:39
@mattdm, actually curves cannot replicate a proper (correlated color temperature) white balance operation. But if the results are visually satisfactory, then that is what counts. – bRad Gibson Apr 6 '13 at 11:29

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.