Abstract: A block transform image processing method employs singular value decomposition to reduce noise without affecting texture and edge detail. A nonlinear gain function based on the measured statistics of the singular values for image noise is generated. The image is filtered to produce a detail image and a low pass filtered image. The detail image is divided into blocks and the blocks are transformed to produce singular vectors and arrays of singular values. The nonlinear gain function is applied to the arrays of singular values and an inverse SVD transform is applied to the modified singular values to produce a processed detail image. The processed detail image is combined with the low pass filtered image to produce a processed image having reduced noise.