A novel and fast algorithm for random valued impulse image noise cancellation using cubic B-splines

A novel and faster method for cancellation of random valued impulse noise is proposed in this paper. Our proposed algorithm works in two stages. In stage one, the noisy pixels are located and eliminated and in second stage, these noisy pixels are used for edge restoration. The results thus obtained are quite better and comparable to previously proposed algorithms and methods even at the noise levels of 60%.