Noise smoothing by a fast K-nearest neighbour algorithm

摘要

Noise smoothing is a basic operation in image processing. Numerous filters, each to be used for different noise conditions and picture types, have been proposed in the literature. A comparison study showed that the K-Nearest Neighbour filter performs extremely well for both additive noise and multiplicative noise; especially when applied in an iterative manner. However, a major drawback to its widespread use is its very heavy computational load. We describe a fast algorithm which reduces the number of computations by more than a factor of 20 for commonly used window sizes.