%0 Journal Article
%T 基于共识均衡的相位恢复<br>Phase Retrieval Based on Consensus Equalization
%A 姜爱伟
%J Advances in Applied Mathematics
%P 2160-2171
%@ 2324-8009
%D 2024
%I Hans Publishing
%R 10.12677/aam.2024.135205
%X 相位恢复是从幅值信息中恢复相位信息，在图像处理等领域发挥重要作用，在实际应用中，往往存在噪声，本文提出一种结合深度学习和即插即用的相位恢复算法，共识均衡相位恢复，运用多个去噪器插入迭代算法，使去噪更加鲁棒。使在迭代求解算法中相位恢复算法与去噪算法达到平衡点，提高了重构质量。本文在仿真和真实数据中对所提算法进行了测试，实验结果表明，该算法在去噪方面表现出更高的鲁棒性，并且具备较强的重构能力。<br />
Phase retrieval is the retrieval of phase information from amplitude information, which plays an important role in image processing and other fields. In practical applications, there is often noise. This paper proposes a phase retrieval algorithm that combines deep learning and plug and play, consensus equalization phase retrieval uses multiple denoising devices to insert iterative algorithms to make denoising more robust. The phase retrieval algorithm and the denoising algorithm reach the balance point in the iterative solution algorithm, and the reconstruction quality is improved. The proposed algorithm is tested in simulation and real data, and the experimental results show that it is more robust for denoising, and has stronger reconstruction ability.
%K 即插即用ADMM，共识均衡，相位恢复<br>Plug and Play ADMM
%K Consensus Equalization
%K Phase Retrieval
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=87778