Department of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
Opt Lett. 2013 Jul 15;38(14):2509-11. doi: 10.1364/OL.38.002509.
Three-dimensional (3D) object tomography from a two-dimensional recorded hologram is a process of high-dimensional data inference from undersampled data. As such, recently, techniques developed in the field of compressive sensing and sparse representation have been applied for this task. While many applications of compressive sensing for tomography from digital holograms have been demonstrated in the past few years, the fundamental limits involved have not yet been addressed. We formulate the guarantees for compressive sensing-based recovery of 3D objects and show their relation to the physical attributes of the recording setup.
从二维记录全息图中进行三维（3D）物体层析成像，是从欠采样数据中推断高维数据的过程。因此，最近，压缩感知和稀疏表示领域中开发的技术已被应用于该任务。虽然过去几年已经展示了数字全息图层析成像中压缩感知的许多应用，但尚未涉及所涉及的基本限制。我们制定了基于压缩感知的 3D 物体恢复的保证，并展示了它们与记录设置的物理属性的关系。
