This project proposes a two-step fuzzy-based filter for restoring color images corrupted with impulsive and gaussian noise. The goal of the first step is to identify the noisy pixels by calculating the similarity between the central pixel and its neighborhood in a selected window. The noisy pixels in a selected window are then replaced by its modified value by computing the median around the central pixel. The second step is fuzzy filtering which is carried out by computing the Rank Ordered Logarithmic difference (ROLD) between the central pixel and its neighborhood in a selected window. Depending upon the difference value, a fuzzy membership degree is calculated using a membership function. Computed fuzzy membership values are appropriately utilized as weights for each pixel and then computes the weighted average representing the modified value for the current central pixel. The output of the proposed system is compared with the some of the existing schemes. The experimental analysis shows that the proposed method gives better results even when the mixed noise percentage is above 50%. The quantitative analysis has been done to prove the efficiency and effectiveness of the proposed method.