An Improved Adaptive Median Filter Algorithm and Its Application

Abstract

Compared with traditional median filter, the filter performance of adaptive median filter has been improved at the cost of high computation complexity. An improved adaptive median filter algorithm is proposed in this paper. First, the filter window size is determined according to the distance between the valid pixels and the center pixels in the proposed algorithm, which can avoid the waste of pixels repeated sort in window expand process. Second, the improved algorithm only to take the median value from valid pixels within the window, effectively weakening the interference with noise point, it will improve the quality of image. Compared with the original algorithm, this proposed algorithm reduced the complexity of algorithm and improved effective of noise reduction, PSNR value average increases 10dB. Finally, this paper applied the algorithm on noise reduction with depth image captured from Kinect.