by Medley on Tue May 08, 2007 3:51 pm
There are a couple of things I can think of off the top of my head that could cause this. If the images were underexposed (dark) and you lightened them, or if you used a higher ISO setting on the camera than normal, that grainy look ( called 'noise' ) can be introduced.
Here the way to 'fix' it in CS2 (fix is such a relative term- but it can be minimized):
Open the image. Go to Image> Calculations. Set the top layer to 'red channel' the bottom layer to 'green channel' and the blend mode to 'overlay'. Click OK. This will create an alpha channel in the channels pallette that is a black and white version of your image. Highlight that channel, and apply Filter> Stylize> Find Edges. This should create something that is more-or-less an outline of your image. If there is any grey in the open spaces, use the Levels (Image> Adjustment> Levels) command to clean it up some. The end result should be black outlines on a white background. Lastly, go to Filter> Blur> Gaussian Blur and set the radius to about 1.5 pixels. This should make the black outlines a bit fuzzy. Click on the 'RGB' thumbnail at the top of the Channels Pallette, and switch to the Layers Pallette.
In the layers pallette, duplicate the background layer. On the top layer, apply Filter> Noise> Dust and Scratches. I usually use a radius setting of about 2 pixels. This should take care of the majority of noise, but has probably blurred the edges of the image some. We're about to fix those edges.
Now, go back to the Channels pallette, and highlight the alpha channel. At the bottom of the pallette, there are several icons. Click the one on the far left that says 'Load channel as selection'. Highlight the 'RGB' channel again, and go back to the Layers pallette. with the selection active, highlight the top layer and click on the 'create new layer mask' icon at the bottom of the layers pallette. The edges should regain some sharpness.
Now you're done with the noise reduction. You may want to sharpen the image some though.
If you have any questions, feel free to ask.
-Medley.
There are only 10 types of people in this world- those who understand binary, and those who don't.