关键词: 图像去噪, 小波变换, 混合噪声, 邻域平均滤波, 中值滤波

Abstract: The reason why different denoising methods are used in the image that contains Gaussian white noise and impulse noise is analyzed firstly；The denoising methods of low frequency subband images and high frequency subband images that have been obtained after wavelet transform are given afterwards.The method of improved local average filter is used to denoise in the low frequency subband images.And the methods of median filter，threshold processing and enhancement of wavelet coefficient are used to denoise in the high frequency subband images；At last，the recovered image after using contrary wavelet transform in each subband image that has been processed is obtained；The results of experiment verifies the correctness of the theory analysis.

Key words: image denoising, wavelet transform, mixture noise, local average filter, median filter