Adding or removing noise

The term “noise” in photography refers to unwanted single pixels or groups of pixels that do not belong in the photo. The noise can be caused various factors. Usually, the problems are either introduced electronically by the camera itself or caused by file formats such as JPEG. Noise appears as colored specks within the photo. You can display the noise best by zooming in. For example, you might see specks of red, pink, green, and yellow in a photo of a clear blue sky.

Corel PaintShop Pro gives you many ways to remove various types of noise, and it even lets you add noise to an otherwise “clean” photo. You can make the following corrections by applying them to a selection or an entire image:

•
 
You can reduce the detail in an image and add a grainy texture. By adding a small percentage of noise to an image, you can reduce the appearance of small imperfections and scratches that cannot be removed with other tools.
•
 
You can find and eliminate small scratches that are either lighter or darker than the surrounding area.
•
 
You can remedy a common problem seen in video capture images in which even- and odd-numbered scan lines are recorded at different rates. The problem is usually noticed if the subject was moving when the image was captured. Note that removing scan lines may make noise more visible.
•
 
You can restore a JPEG image to its original appearance. When you save a file to the JPEG format from any software package (such as scanning software), the file information is compressed to create a smaller file. The compression may produce any of the following artifacts: halos or color leakage beyond the edges of objects, checkerboard patterns on smooth backgrounds, or blocky areas.
•
 
You can remove undesirable patterns from scanned images. This problem sometimes occurs in scanned photographs printed on textured paper.
•
 
You can remove single-pixel specks that are mostly white or black. The brightness of each pixel is analyzed and then compared to the surrounding pixels to determine whether it should be removed.
•
 
You can remove noise in a photo without losing details in edges within the photo. For example, you can preserve the edges of facial features while correcting a blotchy complexion.
•
 
You can remove small random specks or areas of noise that are distinctly different from the surrounding area. The intensity of each pixel is adjusted to the median intensity of surrounding pixels. (The median intensity is the middle value, not the average,) in the range of values.) To preserve object edges, you can use the Median Filter command. With this command, you can apply more adjustment to a pixel that differs greatly from surrounding pixels than to a pixel that is similar to surrounding pixels. You can specify the number of surrounding pixels to use in the calculation. If you choose a large number of pixels, more noise is removed, so more detail is lost.
•
 
You can remove multipixel black or white specks, such as those caused by dust on film or video.
•
 
You can remove noise or specks in an image while preserving details in textures. For example, you can preserve textures in clothing while removing noise from another area (such as a person’s face).

In addition, you can use the One Step Noise Removal and the Digital Noise Removal commands. For more information, see To remove digital noise quickly and To use advanced options for removing digital noise.

To add noise

 Back to Top

Edit workspace

1
 
Choose Adjust Add/Remove Noise Add Noise.
The Add Noise dialog box appears.
2
 
Choose a noise pattern option:
•
Random — creates a granular effect that adds texture
•
Uniform — changes the color of the noise to resemble the original pixels
•
Gaussian — makes most of the noise resemble the original pixels more than the Uniform option
3
 
Type or set a value in the Noise control to set the percentage of noise to add.
4
 
Click OK.
Tip
You can use black-and-white noise pixels by marking the Monochrome check box. To use colored pixels, unmark the check box.

To remove scratches automatically

 Back to Top

Edit workspace

1
 
Choose Adjust Add/Remove Noise Automatic Small Scratch Removal.
The Automatic Small Scratch Removal dialog box appears.
2
 
Mark one or both of the following check boxes:
•
Remove dark scratches — removes scratches that are darker than the background
•
Remove light scratches — removes scratches that are lighter than the background
3
 
In the Local Contrast Limits group boxes, type or set values in the Lower and Upper controls to set the contrast between a scratch and its background.
Increase the Lower control setting to just below where scratches reappear. Decrease the Upper control setting to just above where the scratches reappear.
4
 
In the Strength group box, choose an option to set the strength of the scratch removal:
•
Mild
•
Normal
•
Aggressive
Select the lowest setting that removes scratches. Examine all areas of the image to make sure that important details are not lost.
5
 
Click OK.
Tip
To limit the correction to a specific area, make a selection. The command works better when applied to a selection than when applied to an entire image.

To sharpen video capture images

 Back to Top

Edit workspace

1
 
Choose Adjust Add/Remove Noise Deinterlace.
The Deinterlace dialog box appears.
2
 
In either the Before or After pane at the top of the dialog box, drag to pan to the image area where the scan lines are most apparent.
3
 
In the Scanlines to retain group box, choose an option that specifies whether to retain the odd-numbered or even-numbered scan lines.
4
 
Click OK.
Tip
You should apply the Deinterlace command before you resize the image. Resizing adds or removes pixels from the image, which makes it more difficult to identify scan lines. If your image has already been resized, you must return it to the original size as closely as possible so that each scan line is 1 pixel high. Using the zoom control in the dialog box, you can zoom until you can see how many pixels make up one scan line, and then you can resize the image so that each scan line is 1 pixel high. For example, if the enlarged image has scan lines that are 2 pixels high, you would resize it by 50 percent. Note that resizing the image causes it to lose clarity.

To remove artifacts from JPEG images

 Back to Top

Edit workspace

1
 
Choose Adjust Add/Remove Noise JPEG Artifact Removal.
The JPEG Artifact Removal dialog box appears.
2
 
In either the Before pane or the After pane at the top of the dialog box, drag to pan to the image area where the scan lines are most apparent.
3
 
In the Strength group box, choose an option to determine the strength of correction.
For best results, try each option, and examine all areas of the image, to see which one works best.
4
 
Type or set a value in the Crispness control to specify the amount of detail to restore.
Values range from 0 (the least) to 100 (the most).
Excessive crispness may produce fine dots in the image.
5
 
Click OK.
Tip
You can apply the JPEG Artifact Removal command to a file in any format. For example, if you have saved a JPEG file to the Corel PaintShop Pro (.pspimage) format, you can still apply the command.
Because JPEG compression discards image information, the ability to restore JPEG images is limited.

To remove moire patterns

 Back to Top

Edit workspace

1
 
Choose Adjust Add/Remove Noise Moire Pattern Removal.
The Moire Pattern Removal dialog box appears.
2
 
Type or set a value in the Zoom control to magnify the image until you can see the fine patterns clearly.
3
 
Type or set a value in the Fine details control to set the level of pattern removal.
Change the value one number at a time until the pattern disappears. Higher values may blur the image, so use the lowest value that removes the pattern.
4
 
Use the zoom control in the dialog box to reduce the image magnification until you can see bands or color blotches clearly in the After pane.
Although bands or blotches may be visible when the image is displayed at a setting of 100%, they may be easier to see at lower settings.
5
 
Change the value one number at a time in the Remove bands control until any color bands or blotches become less visible (they may not disappear completely). To avoid desaturation of small objects, use the lowest value possible.
6
 
Click OK.
Tip
After you have removed the moire pattern, you may be able to use the Sharpness commands in the Adjust menu to restore detail and remove blurriness without reintroducing the pattern. The Sharpness commands work best when you choose a Fine details setting in the Moire Pattern Removal dialog box that is slightly higher than needed to remove the pattern.

To remove single-pixel specks

 Back to Top

Edit workspace

1
 
Select the area in the photo that contains the specks.
2
 
Choose Adjust Add/Remove Noise Despeckle.
The selection is despeckled.

To remove noise while preserving edges

 Back to Top

Edit workspace

1
 
Select the area that contains the noise you want to remove.
2
 
Choose Adjust Add/Remove Noise Edge Preserving Smooth.
The Edge Preserving Smooth dialog box appears.
3
 
Type or set a value in the Amount of smoothing control.
Choose the smallest amount of smoothing that removes the specks while retaining image detail. You cannot enter a value lower than 1.
4
 
Click OK.
Tip
Although you can apply the Edge Preserving Smooth command to the entire image, the command works best when you select the problem area.

To remove distinct areas of noise

 Back to Top

Edit workspace

1
 
Select the area that contains the noise you want to remove.
2
 
Choose Adjust Add/Remove Noise Median Filter.
The Median Filter dialog box appears.
3
 
Type or set a value in the Filter Aperture control to choose the number of surrounding pixels to include in determining the median pixel intensity.
Values range from 3 to 31 in odd-numbered increments.
To preserve image details, choose the smallest filter that removes the noise. For single-pixel colored specks, set the filter to 3.
4
 
Click OK.
Tip
Although you can apply the Median Filter command to the entire image, the command works best when you select the problem area.

To remove multipixel specks

 Back to Top

Edit workspace

1
 
Select the area that contains the specks.
2
 
Choose Adjust Add/Remove Noise Salt and Pepper Filter.
The Salt and Pepper Filter dialog box appears.
3
 
Type or set a value in the Speck size control to set the minimum size, in pixels, of the largest speck that can be completely removed. The value is always an odd number.
4
 
Type or set a value in the Sensitivity to specks control to set how different an area must be from its surrounding pixels to be considered a speck.
5
 
Click OK.
Tip
Although you can apply the Pepper Filter command to the entire image, the command works best when you select the problem area.
Tip
You can remove all specks that are smaller than the Speck size value by marking the Include all lower speck sizes check box.
You can intensify the strength of the correction by marking the Aggressive action check box.

To remove noise while preserving textures

 Back to Top

Edit workspace

1
 
Select the area that contains the noise you want to remove.
2
 
Choose Adjust Add/ Remove Noise Texture Preserving Smooth.
The Texture Preserving Smooth dialog box appears.
3
 
Type or set a value in the Amount of correction control to set a value for the strength of the correction.
At lower values, a small amount of noise is removed, but textured areas are preserved. At higher values, more noise is removed, but textured areas may not be fully preserved.
4
 
Click OK.
Tip
Although you can apply the Texture Preserving Smooth command to the entire image, the command works best when you select the problem area.
Feedback
Was this page helpful? Send feedback. (Internet connection required.)

Copyright 2011 Corel Corporation. All rights reserved.