Image De-noising using mean pixel algorithms corrupted with photocopier noise

A Set of Linear Mean Filtering Algorithms and Non-linear Mean Filtering algorithms are proposed and mathematically described in this paper. The performance of this proposed set of De-noising algorithms is evaluated under the presence of AWGN introduced by photocopier machine. It is observed that the Non-linear mean filtering preserve the edge better than the Linear mean filters. For text images, the need for thresholding based on the value of background pixel is implemented. Finally a set of sample images are filtered and the Mean square error, Peak Signal to Noise Ratio are calculated and tabulated for different levels of noise.