A fast adaptive median filtering algorithm

In this paper, a fast and adaptive weighted median filtering algorithm is studied on the basis of the analysis and comparison of some current median filtering algorithms. The weight value of the central pixel is defined according to the noise pollution level. The fast mean method is adopted in the process of searching median to improve the speed of the proposed algorithm. Experimental results demonstrate that the algorithm can achieve better noise suppression, greatly decrease the complexity of time, and can satisfy the demand of computer vision inspection system in protecting the sharp edges and details information.