摘要
针对欠采样图像重建中容易对噪声敏感且出现伪影的问题,构建了结合离散小波和TV的双正则化图像重建模型,基于该模型进一步提出了一种自适应加权迭代图像重建算法。该算法在每次迭代中通过阈值收缩方法依次计算TV正则项与小波系数先验项,更新重建图像。同时为了进一步提升重建图像的质量,引入迭代支集检测方法计算小波系数的自适应权重。实验结果表明,与其他算法相比,该文算法具有更好的时间效率和重建质量。
Aiming at the problem that TV regularized image reconstruction is easy to be sensitive to noise and artifacts in under-sampling environment,a dual regularized adaptive weighted image reconstruction model combining the discrete wavelet and the TV is constructed.Based on this model,an adaptive weighted iterative reconstruction algorithm is proposed.In each iteration,the algorithm calculates the TV regularization term and the wavelet coefficient prior term by the threshold shrinkage method,and then updates the reconstructed image.In order to improve the quality of the reconstructed image,an iterative support detection method is introduced to calculate the adaptive weight of the wavelet coefficient.The experimental results show that the proposed algorithm can achieve better overall performance in terms of time efficiency and reconstruction quality than other similar algorithms.
作者
班晓征
李志华
Ban Xiaozheng;Li Zhihua(School of Internet of Things Engineering,Jiangnan University,Wuxi 214122,China)
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2020年第2期209-215,共7页
Journal of Nanjing University of Science and Technology
基金
工业和信息化部智能制造项目(ZH-XZ-180004)
江苏省科技厅产学研前瞻项目(BY2013015-23)
111基地建设项目(B2018)。
关键词
压缩感知
图像重建
迭代支集检测
自适应加权
compressed sensing
image reconstruction
iterative support detection
adaptive weighted