Sharpening Filters
Sharpening, which can help emphasize details and enhance the edges of objects in an image, is critical when post-processing many types of images. However, you should be aware of the following whenever you sharpen an image:
- Sharpening is very specific to output, so it should be the last thing you do in a filtering workflow.
- Sharpening filters are very sensitive to noise. You should always apply noise reduction first, if necessary.
- Choosing the correct radius is critical for obtaining good results as sharpening may create unwanted edge effects or increase image noise.
These filters can be used to emphasize the details in an image and enhance the edges of objects.
| What it does | Settings | |
|---|---|---|
| Gaussian High Pass | The Gaussian High Pass filter and the Gaussian filter do the opposite of each other. The Gaussian High Pass filter allows high frequency image information to pass through and blocks low frequency image information. This effectively sharpens the image.
In contrast, the Gaussian filter is a low pass filter that allows low frequency image information to pass through and blocks high frequency image information. This effectively blurs or smoothes the image. |
Kernel size
Standard deviation |
| Unsharp | Sharpens an image by subtracting a blurred version of itself from the original.
The Unsharp filter, also called an unsharp mask, is actually used to sharpen an image, contrary to what its name might imply. As such, it is an extremely versatile tool that can improve the definition of fine detail and sharpen edges that are not clearly defined in the original image. One of the primary advantages of the Unsharp filter over other sharpening filters is the option to control the sharpening process. Many sharpening filters do not provide any user-adjustable parameters. The sharpening process works by first creating a slightly blurred version of the original image, the unsharp mask. This is subtracted away from the original to detect the presence of edges. Contrast is then selectively increased along these edges using this mask — leaving behind a sharper final image. |
Kernel size
Standard deviation Unsharp factor |
