Vous êtes ici : Accueil » Kiosque » Annonce

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Annonce

27 avril 2018

Real-time single-pixel imaging


Catégorie : Doctorant


CREATIS (Lyon, France) opens a PhD position funded for three years, starting in October 2018.

Keywords: Single-pixel imaging, compressive video imaging, signal and image processing, image reconstruction, machine learning.

 

Background

Recent advances in signal processing have made it possible to design new digital imaging systems [1]. Single-pixel imaging is a new paradigm that enables two-dimensional imaging from a point detector. It leads to high-performance optical imaging systems (e.g., hyperspectral and/or time-of-flight measurements) at very low cost [2]. Single-pixel cameras comprise a single point detector that is coupled with a spatial light modulator. By performing a sequence of optical measurements for di fferent modulation patterns, it is possible to recover the image of the observed scene provided that ad-hoc reconstruction algorithms are implemented. Our group recently showed that adapting the modulation patterns to the object signi cantly decreases the acquisition and reconstruction times [3-4].

Project

The goal of this project is to develop fast image acquisition-reconstruction approaches by exploiting the spatiotemporal redundancy in videos. State-of-the-art video compression algorithms rely heavily on strategies for motion estimation (e.g., block-based motion estimation, optical flow or wavelet lifting), that can be smartly and eciently incorporated into an adaptive single-pixel framework. As the performances of adaptive methods are based on prediction steps, we intent to benefi t from machine learning methods to determine the best measurement sequence for a particular time frame, and also across multiple time frames.

Funding

The position is funded by the French National Research Agency (ANR) and part of the ARMONI project, which targets fluorescence-guided surgery. The PhD fellow will have access to a real-life system to test his ideas in collaboration with a post-doctoral researcher.

Skills

We are looking for an enthusiastic and creative student with a background in applied mathematics, signal and image processing, computer science, or machine learning, with a strong interest in biomedimedical imaging. Good knowledge of optics would be a plus. Strong programming skills in Matlab and Python are required.

How to apply?

Send CV, motivation letter, and academic records as soon as possible to nicolas.ducros@creatis.insa-lyon.fr, ievgen.redko@creatis.insa-lyon.fr, and peyrin@esrf.fr.

Salary: €1700 net per month (+ teaching).

References

[1] M.-J. Sun et al., "Single-pixel three-dimensional imaging with time-based depth resolution", Nature Communications, vol. 7, no. 12010, 2016.
[2] R. G. Baraniuk et al., "Compressive video sensing: Algorithms, architectures, and applications", IEEE Signal Processing Magazine, vol. 34, no. 1, pp. 52-66, 2017.

[3] F. Rousset et al., "Adaptive basis scan by wavelet prediction for single-pixel imaging", IEEE Transactions on Computational Imaging, vol. 3, no. 1, pp. 36-46, 2017.

[4] Single-pixel CREATIS webpage, https://www.creatis.insa-lyon.fr/~ducros/WebPage/single_pixel_imaging.html.

 

Dans cette rubrique

(c) GdR 720 ISIS - CNRS - 2011-2018.