Satellite images are frequently degraded throughout the data collection process. Degradation can involve blurring and different sources of noise. Image restoration attempts to reconstruct the original image from the corrupted data. In literature, different approaches depend on Kalman filtering exist for image restoration. However, it has certain limits in terms of stability, adaptability and visibility. In this article, a new technique based on the concept of particle filtering and the Weibull distribution (WPF) is proposed to restore images degraded by noise. Our approach is recursive and handles nonlinear situations. Simulation results are eventually exhibited a good performance of the WPF over conventional particle filters.