arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2404.01360eess.IVcs.LGphysics.optics
收藏

深度学习相位恢复:数据驱动、物理驱动,还是二者结合?

Deep learning phase recovery: data-driven, physics-driven, or combining both?

The University of Hong Kong(香港大学)

Kaiqiang Wang, Edmund Y. Lam

AI总结:

该研究针对深度学习相位恢复的两类主流策略缺乏系统对比的问题,全面比较了数据驱动与物理驱动策略的性能,并提出结合二者的协同驱动策略以平衡高低频信息,同时公开了相关代码。

摘要:

相位恢复即从光强测量值计算光波相位,在相干衍射成像、自适应光学、生物医学成像等诸多应用中至关重要,可实现物体折射率分布或形貌的重建,以及成像系统像差的校正。近年来,深度学习已被证实对解决相位恢复问题极为有效。两种最直接的深度学习相位恢复策略分别是采用监督学习模式的数据驱动(DD)策略和采用自监督学习模式的物理驱动(PD)策略。DD与PD以不同方式实现相同目标,目前尚缺乏揭示二者异同的必要研究。因此,本文从耗时、准确率、泛化能力、不适定性适配性及先验容量等方面,全面对比了这两种深度学习相位恢复策略。此外,我们提出了一种结合数据集与物理的协同驱动(CD)策略,以平衡高频与低频信息。DD、PD和CD的代码已在https://github.com/kqwang/DLPR 公开。

英文摘要:

Phase recovery, calculating the phase of a light wave from its intensity measurements, is essential for various applications, such as coherent diffraction imaging, adaptive optics, and biomedical imaging. It enables the reconstruction of an object's refractive index distribution or topography as well as the correction of imaging system aberrations. In recent years, deep learning has been proven to be highly effective in addressing phase recovery problems. Two most direct deep learning phase recovery strategies are data-driven (DD) with supervised learning mode and physics-driven (PD) with self-supervised learning mode. DD and PD achieve the same goal in different ways and lack the necessary study to reveal similarities and differences. Therefore, in this paper, we comprehensively compare these two deep learning phase recovery strategies in terms of time consumption, accuracy, generalization ability, ill-posedness adaptability, and prior capacity. What's more, we propose a co-driven (CD) strategy of combining datasets and physics for the balance of high- and low-frequency information. The codes for DD, PD, and CD are publicly available at https://github.com/kqwang/DLPR.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑