Optical Sensors and Sensing Congress

19 July 2021 – 23 July 2021
Optica Virtual Event - Pacific Daylight/Summer Time (UTC - 07:00)

Architectures for Optical Signal Processing (CF2B)

Presider: Paulo Silveira, Occipital Inc
Authentication Required

Access to this content is only available to registered attendees. Please login now to access the live content.

8:00 - 8:15
(UTC - 07:00)

3D Fluid Flow Reconstruction Using A Compact Light Field Camera (CF2B.1)
Presenter: YUQI DING, Louisiana State University

We present a flexible and low-cost 3D PIV solution that uses a compact light field camera as acquisition device. Real experiments demonstrate that our method can volumetric 3D fluid flows of various types.

Authors:YUQI DING, Louisiana State University / Zhong Li, OPPO US Research Center / Yu Ji, DGene / Jingyi Yu, DGene / Jinwei Ye, Louisiana State University

  Paper

8:15 - 8:30
(UTC - 07:00)

Serpentine Integrated Grating Spectrometer (CF2B.2)
Presenter: Michael Brand, University of Colorado at Boulder

The serpentine integrated grating spectrometer uses a meandering delay line and outcoupling grating rows to implement a crossed-dispersion wide-bandwidth integrated spectrometer with resolving power of 10^5 and 10^4 spectral bins in a few cm^3 volume.

Authors:Michael Brand, University of Colorado at Boulder / Bohan Zhang, Boston University / Milos Popovic, Boston University / Nathan Dostart, NASA Langley / Kelvin Wagner, University of Colorado at Boulder

  Paper

8:30 - 8:45
(UTC - 07:00)

An algorithmic method to identify multiple wood species in a single image by their specific fluorescence lifetimes (CF2B.3)
Presenter: Maximilian Wohlschläger, Technical University Rosenheim

Fluorescence lifetime imaging microscopy (FLIM) combined with an image processing algorithm is successfully used for the optical identification and areal separation of three wood species of maple, larch and nut in a single taken image. © 2020 The Author(s)

Authors:Nina Leiter, Technical University Rosenheim / Maximilian Wohlschläger, Technical University Rosenheim / Veronika Auer, Technical University Rosenheim / Martin Versen, Technical University Rosenheim / Christian Laforsch, University Bayreuth

  Paper

8:45 - 9:00
(UTC - 07:00)

Applying an Iterative Filtering Method for Optical Fringe Patterns Preprocessing (CF2B.4)
Presenter: Mikolaj Rogalski, Warsaw University of Technology

We adapt the Iterative Filtering signal processing technique for fringe pattern preprocessing and modify it with a novel fringe extension algorithm. We compare its performance with state-of-the-art empirical mode decomposition and variational image decomposition methods.

Authors:Mikolaj Rogalski, Warsaw University of Technology / Mateusz Pielach, Polish Academy of Sciences / Maria Cywinska, Warsaw University of Technology / Antonio Cicone, Istituto di Astrofisica e Planetologia Spaziali / Maciej Trusiak, Warsaw University of Technology

  Paper

9:00 - 9:30
(UTC - 07:00)

SLAM Imaging Pipeline for High Dynamic Range Digital Pixel Image Sensor (CF2B.5)
Presenter: Raffaele Capoccia, Facebook Reality Labs

Digital pixel sensors (DPS) hold great potential for computer vision applications such as SLAM. In this paper, we describe an end-to-end vision system for SLAM benchmarking, including the DPS, ISP, and exposure control.

Authors:Chiao Liu, Facebook Reality Labs / Andrew Berkovich, Facebook Reality Labs / Raffaele Capoccia, Facebook Reality Labs / Song Chen, Facebook Reality Labs / Tsung-Hsun Tsai, Facebook Reality Labs

  Paper

9:30 - 9:45
(UTC - 07:00)

A computational study of irregular beam optics to build up an efficient tomographic sensor (CF2B.6)
Presenter: Satwik Choudhury, VECC

Optimization of combustion efficiency in various practical combustion systems often requires tomographic imaging of various combustion products. Here we are presenting a novel beam-optimizing technique to build up the optical-acquisition system of a tomographic sensor.

Authors:Satwik Choudhury, VECC

  Paper

9:45 - 10:00
(UTC - 07:00)

Coded wavefront sensor and non-iterative learning-based reconstruction (CF2B.7)
Presenter: Miao Qi, King Abdullah Univ of Sci & Technology

Here we put forward a coded wavefront sensor and neural network framework to achieve wavefront reconstruction. The new framework outperform the preview method not only in the speed but also in the image fidelity

Authors:Miao Qi, King Abdullah Univ of Sci & Technology / Congli Wang, King Abdullah Univ of Sci & Technology / Wolfgang Heidrich, King Abdullah Univ of Sci & Technology

  Paper