1. Averaging of multiple images. It is applicable when several noise degraded images. It assumes that the noise is uncorrelated and has mean equal zero. ai(x) = a0(x) + ni(x), where a0 is true image, ni is the noise.
2. Local averaging. Smooth the image by replacing the intensity at a point by the average of the intensities in a neighborhood of the point (low pass filter). It will get an artifact boundary in result image.
3. Variable local averaging. Similar with last one, it allows the size of the neighborhood to vary.
4. Gaussian Smoothing. “there are two considerations in choosing a smoothing ﬁlter. First, it must reduce the range of scales over which intensity changes occur, thus it should be smooth and relatively band-limited in the frequency domain. Second, it should represent an averaging of nearby points, thus it should be smooth and localized in the spatial domain as well.”
5. Max-Min sharpening transform. sharpens fuzzy boundaries and brings fuzzy gray level objects into focus. It also smoothens isolated peaks or valleys. It is an iterative technique that compares maximum and minimum values with respect to the central pixel value in a small neighborhood. The central pixel value is replaced by whichever of the extrema in its neighborhood is closest to its value.
6. Median filter. A smoothing technique that causes minimal edge blurring. However, it will remove isolated spikes and may destroy ﬁne lines
7. Unsharp Masking. Blending an image's high-frequency components and low-frequency components to produce an enhanced image.
C = r*(a-b) + b, or C = \alpha * (a-b) + \gama * b
a:source image. b: image after averaging mask, is low frequency component. a-b: high frequency component. r: 0~1
8. Histogram Equalization.
9. Histogram Modification.
10. LPF. eg. butterworth lpf.