Skip to Main Content
In the field of image processing, it is inevitable to contain various noises in the image signals obtaining by the CCD camera. CCD noise model is a combination of a mixture of a fixed-pattern noise and multiplicative Gaussian noise and a mixture of signal-independent noise. Based on the model, this paper proposed a method to remove the noise from digital images corrupted by CCD device. The method developed a de-noising algorithm to restore images contaminated by the CCD noise sources. Because the great correlation exists among the wavelet coefficients, this paper adopts contourlet transform based on an efficient two-dimensional multiscale and directional filter bank that can deal effectively with images having smooth contours. The scheme will sharpen the edges and smooth the uniform area at the same time. At the end of the paper, it performed analysis and simulations to understand the performance of the scheme. The results indicated that the de-noising method is efficient.