Often the image of the interference is found, for example, experienced image noise, so we need a method to reduce noise contained in the image. Image restoration is a technique to repair images that have been degraded or damaged such as noise. One technique in image restoration is filtering. Thetechniquefiltering used in this study is to compare the performance of the Mean Filter, Median Filter, and Alpha-Trimmed Mean Filter in reducing noisemethodsso that it can be determined which method is better applied in image restoration, especially images containing speckle noise and Rayleigh noise. The test results on test image 1 using themethodMean Filter for noise speckle have MSE values of 2774,758, PSNR 14,6714 and Running Time 1,351.4, while for Rayleigh noise have values of MSE 4414,909, PNSR 11,891, and Running Time 1,672, 8, themethodMedian Filter for noise speckle and rayleigh noise has almost the same parameter values, while themethodAlpha-TrimmedThe mean filter for noise speckle has a value of MSE 2926.81, PSNR 14.060 and1.392 Running Time , while for noise itRayleighhas a value of MSE 5000.867, PNSR 11.824, and Running Time 1.313. The conclusion of the test when viewed visually, the performance of themethodsMean Filter and Median Filter in reducing speckle noise and Rayleigh noise produces almost the same image quality compared to themethodAlpha-Trimmed Mean Filter. However, based on the parameters of MSE, PSNR and Running Time, the performance of themethodAlpha-Trimmed Mean Filter is the best in reducing speckle noise and rayleigh noise.