Varying weight trimmed mean filter for the restoration of impulse noise corrupted images

A weight function is introduced to the generalized version of the alpha-trimmed mean filter for the removal of impulse noise from corrupted images. An iterative implementation of the new filter-based switching scheme is proposed as an impulse detector to preserve the noise-free pixels. Numerical simulations show that the proposed filter is quite robust and efficient, and its performance compares favorably with many other well-known nonlinear filtering algorithms.