A comparison of adaptive filters for edge-preserving smoothing of speckle noise

The authors present the results of an extensive study investigating the applicability of different existing nonlinear adaptive filtering methods, as well as a new speckle-model based quadratic Volterra filter (QVF), to solve the problem of smoothing speckle noise in digital images while preserving important edge information. The QVF consists of a linear part, mainly responsible for noise smoothing, and a quadratic part for filtering near edges. This filtering strategy allows for consideration of the local correlation of speckle noise. The three best overall filters were found to be the Lee multiplicative filter, the Frost filter and the quadratic Volterra filter. The QVF provides promising results in both noise smoothing and edge preservation.<>