How To Sharpen Photographs ?

[bburton]

There's a brief thread started on sony talk forum re: sharpening methods

http://forums.dpreview.com/forums/re...ssage=12320736

But, I know this group better and would like some discussion and suggestions as to how to actually schedule my post processing workflow so as to optimize sharpening.

From everything I've read, it still seems like "voodoo" to me: everyone has a different magical formula as to how to go about it.

Thanks ahead of time.

Bruce

[Jim Victory]

Bruce that is pretty much it. The files from different cameras react different to sharpening settings. You have to determine what works for your particular cameras files.

If you choose PSCS then you need to look at what settings in USM works for you. You could also try using a high pass filter and blending mode to sharpen your images. I preferr this to USM, especially if you are working with jpg, because it will not produce the artifacts that USM can.

I personally use Focal Blade for my sharpening because it is less prone to artifacts and can be adjusted to match any file. Since I shoot several different cameras it is perfect for me. I have a sharpening profile set for each type of file. It is also a plug-in for PSCS so I can include it in my workflow.

Finally, sharpening should be the last step in your workflow.

[bburton]

Thanks Jim.

Since sharpening is the last step in workflow, and for example, I want to post a picture in this forum that is 800x600, then whatever sharpening approach I use should be the one that optimizes the reduced image size and NOT the original image size. Right?

Bruce

[Bill Huber]

Originally Posted by bburton

Thanks Jim.

Since sharpening is the last step in workflow, and for example, I want to post a picture in this forum that is 800x600, then whatever sharpening approach I use should be the one that optimizes the reduced image size and NOT the original image size. Right?

Bruce

There are I am sure as many ways to do it as there are people doing it.

This is my workflow for a basic JPG files.
Adjust levels (I have not got curves down yet)
Color balance if needed.
Brightness/Contrast if needed.
USM, if needed, 1/2 of what is needed.
USM again if needed.
Neatimage for noise or to soften people shots if needed.
Now if I was working with layers I would save it as a psd file.

Save JPG as a new file name.

If I want to post the image, I will copy it to My Pictures, with ACDSee I will resize it in that folder and then upload it from there.

Notes:
My original file is untouched, just as it came from the camera.
My edited files is still the same size and I can do more if needed.
I can always delete everything out of My Pictures and be sure I did not delete a good file.

Now you have to remember I have an Olympus E1, so I don't have to do much to any of my images.....

[Hogleg 44]

Those Sony's do a pretty good job of sharpening on their own, so you probably don't need to do very much, like Bill says with his Olympus.

I have a 707 and an 828, but not the 717, but I'm sure it is about the same as the other two.

Fred Miranda, on his site, has a pretty nice sharpening program plugin for PS that is pretty highly regarded, but the unsharp mask in your editing tool is probably all you need.

I always sharpen a little, just after any other adjustments, before resizing, then a little more after resizing.

[bburton]

Thanks, Bill and Paul.

You North Richland Hills guys sure know your stuff. You've been a big help to me in a number of areas.

P.S. Paul, is there a camera you DON'T own ? <grin>

[saroy]

Sharpening is usually my second-to-last step (final step is resizing). This applies to jpegs, since that's what I tend to shoot:

After I make all my other adjustments, I convert from RGB to lab color, select the "lightness" channel, make sure I'm viewing the image (at least part of it) at 100%, and then use an unsharp mask. My default settings are 85%, radius 0.3, threshold 1. If that doesn't do it well enough, I up the radius a bit. Radius has the largest effect on the final image. Convert back to RGB, resize, save.

Dunno if this is method is better/worse than others, but it works for me.

[Hogleg 44]

Originally Posted by bburton

Thanks, Bill and Paul.

P.S. Paul, is there a camera you DON'T own ? <grin>

It sure sounds like I have a bunch doesn't it? Really I don't, but I have had a digital camera since they first became widely available in the latter 90's. I upgraded every couple of years as they became better and better. So, all in all, I have owned(own) a few.
It's really amazing how many camera's I've owned, not to know any more than I do about picture-taking! But I do it mainly for my own amusement and event recording, and don't really care that much about trying to be perfect at it. It is just another of the things that I enjoy fooling with.
I don't smoke, drink, or run around with strange wimmen, so I have to do something....!

[Mickey]

Here is a good tutorial.
http://www.lonestardigital.com/usm_fade.htm

They also have a great tutorial on haze removal on their homepage.

Regards,
Mickey

[saber]

I use the same technique as saroy, but the technique mentioned by Mickey is great as well. By coverting it to lab color you don't mess up any color. really works fine with me.

thanx for the tip mickey