An Area-Efficient Alternative to Adaptive Median Filtering in FPGAs

This paper presents a new approach to the FPGA implementation of image filters which are utilized to remove the salt-and-pepper noise of high intensity (up to 70% of corrupted pixels). The proposed solution combines image filters designed by means of evolutionary algorithm with a simple human-designed preprocessing and post-processing unit. It provides the same filtering capability as a standard adaptive median filter; however, using four times less Virtex slices.