Caramazza Piergiorgio, Wilson Kali, Gariepy Genevieve, Leach Jonathan, McLaughlin Stephen, Faccio Daniele, Altmann Yoann
Institute of Photonics and Quantum Sciences, School of Engineering & Physical Sciences, Heriot-Watt University, EH14 4AS, Edinburgh, United Kingdom.
Institute of Sensors, Signals and Systems, School of Engineering & Physical Sciences, Heriot-Watt University, EH14 4AS, Edinburgh, United Kingdom.
Sci Rep. 2018 Mar 27;8(1):5257. doi: 10.1038/s41598-018-22811-x.
In this work, we address the reconstruction of spatial patterns that are encoded in light fields associated with a series of light pulses emitted by a laser source and imaged using photon-counting cameras, with an intrinsic response significantly longer than the pulse delay. Adopting a Bayesian approach, we propose and demonstrate experimentally a novel joint temporal deconvolution algorithm taking advantage of the fact that single pulses are observed simultaneously by different pixels. Using an intensified CCD camera with a 1000-ps gate, stepped with 10-ps increments, we show the ability to resolve images that are separated by a 10-ps delay, four time better compared to standard deconvolution techniques.
在这项工作中，我们致力于重建空间模式，这些模式编码在与激光源发射的一系列光脉冲相关的光场中，并使用光子计数相机进行成像，其固有响应时间明显长于脉冲延迟。采用贝叶斯方法，我们提出并通过实验证明了一种新颖的联合时间去卷积算法，该算法利用了不同像素同时观测单个脉冲这一事实。使用具有1000皮秒门控且步长为10皮秒的增强型电荷耦合器件相机，我们展示了分辨延迟为10皮秒的图像的能力，与标准去卷积技术相比，分辨率提高了四倍。
