PENINGKATAN KUALITAS CITRA RGB MENGGUNAKAN TEKNIK DENOISING --

ABSTRACT
Image can have a poor quality because the image have some noise in it, this issue can caused by a poor acquisition process, poor of lightning while taking an image, and lack of equipment. Denoising technique can increase quality of image by reducing a noise in image, there are many method available in denoising technique, but not all of them can run optimally due to the many type of noise which have own characteristics. In this study, denoising research will be carried out using 3 methods, namely mean filter, median filter, and wiener filter, to reduce noise noise in images that have salt & pepper, gaussian, speckle, and localvar noise types, denoising result will also be compared with original image by using MSE and PSNR to determine which method are perform well in specific type of noise. The results obtained in this study show that all the methods used can reduce noise in the image but the performance results vary, in the salt and pepper denoising process with low intesity of noise , the highest PSNR value is median filter with PSNR = 32.12 dB, mean filter with PSNR = 21.50 dB, and wiener filter with PSNR = 18.56 dB.
Keywords : Image, Denoising, Image processing, Filtering