Two-Dimensional Linear Prediction Based Median Filtering

This paper presents a new method for filtering the impulse noise in images. This is a two-stage approach in which an impulse is first detected by thresholding the energy of the prediction error in a block-based twodimensional linear prediction of the image intensity. The median filter is subsequently applied to the pixels detected to be corrupted by the impulse noise. Simulation results show the superior performance of the proposed filtering method compared to the classical median filter in handling the impuse noise without modifying the clean pixels.