In this talk, we consider using the smoothing and thresholding (SaT) approach for image segmentation. It is important to select the appropriate regularization parameters in the smoothing stage. Traditionally, the regularization parameters are usually chosen by trial-and-error, which is a very time-consuming procedure and is not practical in real applications. In this paper, we apply Bayesian inference approach to infer the regularization parameters and estimate the smoothed image. Experimental results show that the proposed approach can obtain good segmentation results. Although the proposed approach contains an inference step to estimate the regularization parameters, it requires less CPU running times to obtain the smoothed image comparing to previous methods.