Unsharp Masking


Unsharp Masking

The process of subtracting a low-pass filtered or blurred image from an original.
It is a technique used to emphasize edges in images.

Subtracting a low-pass filtered (or unsharp) image is equivalent to adding a high-pass filtered image.
Hence the unsharp masking operation can be expressed mathematically as


where v(m,n) is the resulting image, u(m,n) is the original image, and g(m,n) is a high-pass filtered image of the original such as

The factor lambda will determine how much high frequency boost is desired.
The Laplacian mask [0 -0.25 0; -0.25 1 -0.25; 0 -0.25 0] represents the high pass filter used in this applet.

The steps taken are:

  1. Obtain the high pass filtered image
  2. Multiply by the factor lambda
  3. Add to the original image
  4. Hence we get a resultant high-frequency boosted image .