Damaged images are often collected and transmitted in their noise. For example, in image acquisition, the performance of image sensor is affected by many factors, such as environmental conditions and the content of quality inspection itself. For example, in a CCD camera that acquires images, light level and sensor temperature are the main factors that affect the number of noise images generated. Images will also be destroyed during transmission, because the interference channel is used for transmission. Image denoising technology must eliminate this additional random noise, while retaining the functions of as many important signals as possible. The main goal of these types of random noise removal is to suppress noise while maintaining the details of the original image. Statistical filter is the same as mean filter [1] [2], and Wiener filter [3] can be used to eliminate this kind of noise, but the denoising method based on wavelet transform has proved to be different from these filters. Generally speaking, image denoising provides a compromise between reducing noise and protecting important image details. In order to achieve good performance in this respect, a denoising algorithm is adopted to adapt to the discontinuity of the image. The representativeness of wavelet is naturally beneficial to the construction of this spatial adaptive algorithm. It compresses an important information signal into a relatively small number of coefficients and represents image details with different resolution scales. In recent years, there have been a considerable number of studies on wavelet threshold and threshold selection for signal and image denoising [4] [5] [6] [7] [8] [9], because wavelet provides a suitable basis for separating noise signals from image signals. Many wavelet threshold techniques, such as VisuShrink [10] and BayesShrink [1 1], have proved that image denoising has better benefits. Here, we describe an effective threshold denoising technique by analyzing the wavelet coefficients of statistical parameters. This paper is arranged as follows: In the second section, discrete wavelet transform (DWT domain) and wavelet filter banks are briefly reviewed. The wavelet threshold technique is based on the explanation in the third part. The fourth part explains the new threshold technology. The steps of working in this range are explained in sections 5 and 6, and the experimental results of the proposed work are current and compared with other denoising techniques. Finally, the seventh part is the conclusion.