The K Nearest Neighbor Geometry Filter Based on Spatial Domain

Based on the filter using the K nearest neighbor pixels and geometric mean grayscale value method, a derived image denoising algorithm is proposed in this paper. This approach firstly searches K-l nearest grayscale neighbors of a central pixel covered by the mask. Then it calculates geometric mean gray value of the K pixels including the k-l neighbors and the central pixel. Lastly, replaces the grayscale value of the central pixel with the geometric gray value. Experiment results show the proposed method has better performance on the mixed noise suppression in comparison to the classical Mean filter, the standard median filter and the KNN mean filter.