An Efficient Low Complexity Denoising Filter Architecture for the Removal of Salt and Pepper Noise
R. Senthamizh Selvi, Resmi R. Nair, E Rithish Kumar · 2022 3rd International Conference on Smart Electronics and Communication (ICOSEC) · 2022
Image denoising has applications in sectors such as astronomy, where there are severe resolution constraints, medical imaging, where high-quality pictures are necessary for analysis, and forensic science, where we must sometimes handle extremely poor-quality photos for evidence. To enhance the image a variety of filtering algorithms are used, one of which is the Median Filter, which removes impulse noise while keeping edge information. It replaces the value of the noisy pixel with the median of the intensity values in its vicinity. The proposed architecture is area efficient, low in complexity, and high speed, thus proposed median filter architecture is tested on a real image, and the PSNR and SNR of the median filter are analysed. A decision filter is applied to the median to filter pixels having noise alone and process it hence increasing the speed and PSNR significantly comparing the performance of a 3x3 window filter with a 5x5 window filter proved that the noise reduction by the latter filter is efficient for higher noise figures and at lower noise Figure both the filter performs optimally.