Filtering of noise is one of the most important tasks in digital image processing. Elimination of noise is an image is still a challenging task for the researchers. Various researches were underlying on face recognition due to the variations in image acquisition such as poor illumination, viewing directions etc. However, the authenticity of noises that might insert into an image document will affect the performance of face recognition algorithms. Hence, different filtering algorithms are presented for noise elimination using various filtering algorithm. Finally, the experiments and results show that a Gaussian filter outperforms the other existing filtering algorithms like Mean, Median, Wiener, High Boost, under Salt and Pepper Noise.