Image Restoration Using RO Learning Approach
Chunshien Li, Chan-Hung Yeh, Jye Lee · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2006
In this paper, a machine-leaning-based adaptive approach is proposed to restore image from Gaussian corruption.The well-known Random-Optimization (RO) learning method is used for training of the adaptive filter.With the merit of model-free computation of RO, the derivative information is not required.Combined with block processing technique, the proposed adaptive filtering approach possesses fast convergence, moderate computation and simplicity.The proposed adaptive filter shows excellent filtering performance for image restoration.