Poisson-Gaussian Noise Reduction for X-Ray Images Based on Local Linear Minimum Mean Square Error Shrinkage in Nonsubsampled Contourlet Transform Domain

Noise reduction is important for X-ray images because it can reduce radiation exposure to patients. image noise has a Poisson-Gaussian distribution, and recently, analysis removal in multiscale transformations have been widely implemented. The nonsubsampled contourlet transform (NSCT) transformation suitable medical that separates the scale direction. This study proposes noise-removal method using NSCT shrinkage based on characteristics of domain. It structure block-matching 3D filtering algorithm form basic estimation process; however, main processes are modified consider characteristics. In process, an developed by optimizing local linear minimum mean square error estimator denoising step, term Wiener filter determined result shrinkage, finally, denoised obtained. proposed applied simulated real compared with other state-of-the-art methods; exhibits excellent results both quantitative qualitative aspects.