A Pipelined Architecture for Batcher’s Sorting Network based Median Filter Design

Spatial filtering is one of the principal tools used in image processing for a broad spectrum of applications. Median filtering is a nonlinear spatial filtering technique, often used to remove noise and it is very widely used in digital image processing because, under certain conditions, it preserves edges while removing noise. In the paper, median filter is implemented using 16 input Batcher’s odd even merge sorting network which is a better sorting network when compared to the other networks. Images in YUV formats are taken and to show the action of median filtering, noise is introduced into the images by using MATLAB. Now they are processed by the designed median filter and simulated using Modelsim. The timing results are taken on Cyclone II FPGA and the clock rate for the design has got improved and it is found to be around 127MHz. And finally Peak Signal to Noise Ratio(PSNR) is calculated for the set of images.