Impulse noise removal by multi-state median filtering

Images are often corrupted by impulse noise due to a noisy sensor or channel transmission errors. The goal of removing impulse noise is to suppress the noise while preserving the integrity of edge and detail information associated with the original image. In this paper, a generalized framework for median filtering based on a switching scheme, called multi-state median (MSM) filter, is proposed. By using a simple thresholding operation, the output of the proposed MSM filter is adaptively switched among those of a group of centre weighted median (CWM) filters having different centre weights. As a result, the MSM filter is equivalent to an adaptive CWM filter with a space varying center weight which is dependent on local signal characteristics. The effectiveness in noise suppression and detail preservation of the proposed filtering technique has been evaluated by extensive simulations, showing superior performance to other median based filters.