An Adaptive Regularization Iterative Reconstruction Algorithm on the Basis of a Sparse Constraint
-
摘要: 本文针对ASD-POCS算法中约束项权重对不同应用的多变性引起的算法鲁棒性差等问题,提出了一种基于稀疏约束的自适应正则化迭代重建算法,该算法采用一种Lagendijk型的正则化策略构造最优化问题,分别采用局部方差、图像能量估计自适应地求取加权对角矩阵和全局正则化参数。最优化问题的求解过程中,采用SART算法和共轭梯度法求解保真项和约束项最优化问题。实验结果表明,AR-SART-CG算法能更好地权衡恢复图像边缘和平滑噪声的关系,更好地调节保真项和约束项的权重,得到更高质量的重建图像。Abstract: In this paper, the constraints on Weighting of ASD-POCS (Adaptive Steepest Descent-Projection Onto Convex Sets, ASD-POCS) algorithm weight caused by the variability of different applications, algorithm robustness and poor. It proposes an adaptive regularization iterative reconstruction algorithm on the basis of a sparse constraint: the AR-SART-CG (Adaptive Regularization-Simultaneous Algebraic Reconstruction Techniques-Conjugate Gradient, AR-SART-CG) algorithm. The algorithm adopts a kind of Lagendijk type of Regularization Strategy to construct optimization, respectively uses local variance, noise estimation, and image energy estimation to adaptively adjust the parameters weighted diagonal matrix and global regularization, and respectively applies SART algorithm and conjugate gradient method to solve optimizations of fidelity term and constraint term. Since that the algorithm can adaptively adjust the weight of constraints, it is possessed of strong robustness. The experimental results show that the AR-SART-CG algorithm can better balance and preserve the relations between picture edge and smooth noise.
-
Keywords:
- ASD-POCS /
- regularization /
- compressed sensing /
- SART /
- conjugate gradient
-
计量
- 文章访问数: 613
- HTML全文浏览量: 4
- PDF下载量: 7
下载:
