基于全变分正则化的双侧约束图像去模糊问题

Bilaterally Constrained Image Deblurring Problem Based on the Total Variation Regularization

针对经典的图像去模糊问题,提出用交替Bregman方法求解双侧约束变分正则化图像去模糊问题.该算法通过引入辅助变量将初始问题转化为求解几个子优化问题,然后采用交替方向法进行快速求解.实验结果表明,对于不同程度的高斯模糊和噪声,该模型和算法都能够得到有效地恢复结果.

For the classical image deblurring problem, we propose an alternating Bregman method for solving bilaterally constrained image deblurring problem based on the total variation regularization. This method divides the original problem into several subproblems by introducing some auxiliary variables and then alternatively solves these subproblems. The experimental results indicate that with different size blurry kernels and noisy levels the proposed model and algorithm can recover image effectively.