科研之友
微信
新浪微博
Facebook
分享链接
摘要
. Image noise removal is one of the important research directions in the field of image processing, and it can be transformed into a large scale unconstrained optimization problem. Based on the famous two-phase method, in this paper we propose an adaptive conjugate gradient method to remove the impulse noise from the corrupted images by solving the transformed unconstrained optimization problem, which is defined in terms of an edge-preserving potential function. The proposed method can generate a sufficient descent search direction at each step independent of line search condition, and it has global convergence under the Armijo line search condition. Experimental results show that the proposed method is efficient and robust for impulse noise removal.
