In view of the problem that image acquisition process is susceptible to noise interference, several common algorithms of image noise reduction process such as neighborhood averaging method, median filtering method and averages filtering between images are introduced. The advantages, disadvantages and application areas of these are described separately. An image noise reduction algorithm of wavelet threshold based on smooth filtering is proposed. In this method, median filtering method and wavelet threshold method are combined to perform smoothing image process. The experiment results show that this method is not only of advantage to image algorithms, but also to reserve the edge information, which improves the visual effect of images and restores the primary images as well.