| X-ray digital radiographic technology is widely applied in safety inspection. The original X-ray image is noisy and blurred. The enhancement of original X-ray image is the key to the safety inspection work. Contourlet tranform is a effective tool for the analysis of images, which inherits the localization and multi -resolution property of wavelet. Furthermore, it is anisotropic and provides more directions than wavelet. The transform can capture geometric stucture of images more effectively.and it is widely used in the field of image enhancement.Using traditinonal algorithms to enhance image details will also enlarge noise. Acoording to the X-ray image needs to enhance the details and suppress noise, we try to work on an image denoising method based on Contourlet tranform and an image enhancement method based on Nonsubsampled Contourlet Transform (NSCT). Firstly, the basic theory of Multiscale Geometric Analysis is introduced, which is represented by Contourlet. Then the denoising method based on wavelet and Contourlet is introduced. An image denoising method is proposed which based on the combination of adaptive threshold and Cycle Spinning-Contourlet transform. The adaptive threshold can be calculated by local variance of Contourlet coefficient. Ideal images and X-ray images are processed by the proposed method, the simulation shows that the method can get better denosing effect than original methods. At last, according to the thought of thresold denoising, a X-ray image enhancement method based on Nonsubsampled Contourlet Transform is proposed. According to the distribution of X-ray image’s NCST coefficients, a local bayes threshold is constructed to distinguish the noise coefficients and the image details coefficients. We use a nonliner amplified function to enhance the high frequency coefficients and use a method of nonliner contrast stretching to enhance the low frequency coefficients which represent the global information of the X-ray image.Many X-ray images are processed by the proposed method in NSCT domain, and the effectiveness of the method are proved by the experimental results. The experimental results show that this method can both enhance the image and suppress noise, the output image has a superier visual effect over other traditinonal methods. |