Robust and efficient algorithm of image enhancement

A method for enhancing the local contrast of a noisy image using sliding discrete transforms is proposed. A minimum mean-square error estimator in the domain of a sliding transform for noise removal is derived. This estimator is based on a fast inverse sliding transform. Local contrast enhancement is performed by a nonlinear modification of denoised local spectrum coefficients. To provide image processing in real time, a fast recursive algorithm for computing the sliding transform is utilized. The algorithm is based on a recursive relationship between three subsequent local spectra. Computer simulation results using a real image are provided and discussed.