【摘要】 两阶段算法是去除椒盐噪声的一类非常有效的方法.该文设计了一类PRP共轭梯度法来求解两阶段法中的无约束最优化问题.该方法具有两个显著特点：搜索方向具有充分下降性；结合一类定制的Armijo型非精确线搜索，其点列的梯度趋于零，而不是通常共轭梯度法的下极限趋于零.将所设计方法及一些经典的共轭梯度法应用到两阶段算法的最优化问题中，数值结果表明该文设计的方法在收敛速度与计算精度方面都具有优势.更多还原

【Abstract】 Two-phase algorithm is an effective method to remove salt-and-pepper noise. In this paper, a PRP conjugate gradient method is designed to solve the unconstrained optimization problem in the two-phase method. This method has two remarkable characteristics: the search direction has sufficient descent; combined with a customized Armijo-type inexact line search, the gradient of iterative sequence tends to zero, instead of the lower limit of the conjugate gradient method. The designed method and some classical conjugate gradient methods are applied to the optimization problem of two-phase algorithm. The numerical results show that the method designed in this paper has advantages in convergence speed and calculation accuracy.更多还原