Penerapan Metode Modified Mefian Filter Dalam Meningkatkan Kualitas Citra Underwater

Serlin Syah (Universitas Budi Darma, Medan)
Herry Sunandar (Universitas Budi Darma, Medan)
Alwin Fau (Universitas Budi Darma, Medan)

KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer)
Vol 6, No 1 (2022): Challenge and Opportunity For Z Generation in Metaverse Era

Underwater images often experience disturbances such as imperfect noise and noise which can result in capture results being too dark or too bright, blurred or even less sharp, which will result in very poor image quality. Modified Median Filter is a development of a noise detection algorithm based on a simple concept, namely if a pixel belongs to a uniform area, then it is close in color to a neighboring pixel, then it is not corrected, if there is nothing close to a neighboring pixel, then noise is detected then the pixel value is replaced with the median of the evaluated window. Image quality is measured by means of two magnitudes, namely MSE (Mean Square Error) and PSNR (Peak Signal to Noise Ration) by comparing MSE and MNSR values to obtain results and improve image quality.