The mosquito noise and false contour that doesn't exist in the original image sometimes appear in a decoded image. These noises are desirable for us to remove. Thereupon, we propose the new algorithm that mosquito noise is reduced on the basis of the pixel information of a decoded image. First of all, our proposed algorithm calculates the standard deviation by using the luminance of each pixel in the square block.Next, we do the filtering process by using the Epsilon filter algorithm with the standard deviation in each square block. Through several computer experiments, we understood that we were able to reduce mosquito noise by using our proposed algorithm without damaging the quality of the decoded image.