Section 93. Remove Specks and Spots

93. Remove Specks and Spots

BEFORE YOU BEGIN

90 About Removing Scratches, Specks, and Holes


SEE ALSO

94 Repair Holes and Tears


Noise can creep into your images in the most innocuous waysafter scanning a newspaper image, capturing an image with a video camera or grabbing it from live video, taking a long exposure with a digital camera ( CCD noise ), over-sharpening an image, or simply letting an image accumulate everyday dust and dirt. You'll notice noise the most in the shadows or the flat, nontextured, non patterned areas of an image.

KEY TERM

CCD noise Random distortions introduced into a photo by a digital camera's CCD chipits principal light detector. CCD noise happens most often during long exposures or at high ISO settings (film speed), and is magnified when "electronic zoom" is used to simulate a close-up of a subject taken at a distance. This type of noise mostly affects the shadow areas of an image, within the red and blue channels.


Paint Shop Pro provides many filters you can use to remove noise from an image. Despeckle removes nearly black or nearly white single pixel specks, and thus has limited use. The Salt and Pepper filter removes specks of the size you specify, whose brightness varies dramatically from that of surrounding pixels. Salt and Pepper is great at removing general noise and small specks of dust and dirt from old photographs. The Median filter is similar to Salt and Pepper except that it changes all pixels to match the median brightness of their neighbors, and not just the specks. This task shows you how to use the Salt and Pepper and Median filters. The Median filter works faster than the Salt and Pepper filter because it's less discriminating but it often causes detail to be lost. You might want to try both the Salt and Pepper and the Median filter on an image and compare the results.

NOTE

To use the Despeckle filter, choose Adjust, Add/Remove Noise, Despeckle . The filter works automatically, removing single, very dark, or very light specks.


1.
Choose a Filter

Choose Adjust, Add/Remove Noise from the menu bar, and then choose either Salt and Pepper Filter or Median Filter . Here, I'll try both filters on an old photo that has acquired some small specks.

93. Remove Specks and Spots


TIP

To prevent loss of detail, first select the areas of your image that contain noise, and then choose the filter. Only the area you selected will be affected by the filter.

2.
Set Speck or Sample Size

In the Salt and Pepper Filter dialog box, adjust the size of the specks you want to remove by changing the Speck size value. To remove specks this size and smaller, enable the Include All Lower Speck Sizes check box.

In the Median Filter dialog box, adjust the area to sample for brightness differences by changing the Filter aperture value. Each pixel in the selection or layer is compared to the brightness of its many neighbors and is adjusted accordingly .

3.
Set Options and Click OK

The Median filter has no other options, but in the Salt and Pepper Filter dialog box, you can also set the Sensitivity to specks . This value tells PSP how different an area must be from surrounding pixels to be considered a speck; higher values tell PSP that an area doesn't have to be that different from its neighbors to be a speck. Enable the Aggressive Action check box to have PSP change the brightness of specks more drastically than it would normally. Click OK to apply the filter.

After using the Median filter, you will probably have to sharpen the image.

NOTE

If you compare the images, you'll see that both filters did a pretty good job at eliminating the small white specks in my red sweater and black pants, and in the shadows of the fireplace. The Median filter, however, has blurred the image too much. Compare the results in the Color Gallery section in this book.




Sams Teach Yourself Creating Web Pages All in One
Sams Teach Yourself Creating Web Pages All in One
ISBN: 0672326906
EAN: 2147483647
Year: 2006
Pages: 276

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net