In this study, we present an adaptive anisotropic diffusion total variation method which combine wavelet transformation and total variation (TV) equations. The algorithm puts high-frequency coefficients of wavelet into the iterative total variation equations to construct new wavelet coefficients. At the same time, we introduce spread function into this algorithm. In contrast with previous work combining TV restoration with wavelet compression, this method presented in this study not only treats the numerical solution in a novel way which decreases the computational cost associated with the solution of the TV model, but also introduces spread function into this algorithm, which adds the difficulties of the problem. We present a detailed description of our method which indicates that a combination of wavelet based on restoration technique with the TV model produces superior results.
展开▼
