[visionlist] Fully-Funded PhD studentship at Colour & Imaging Lab, University of East Anglia

[Michal Mackiewicz (CMP - Staff)]

Fully-Funded PhD studentship in Computer Vision at the Colour & Imaging Lab, University of East Anglia, Norwich, UK

Project Title: Designing cameras for Spectral Capture
Project Description: If we could take pictures of the world where instead of recording colour (i.e. the RGB ) at each pixel we measured the per-pixel radiance spectrum then we could solve many problems that are not currently solvable in colour imaging (ranging from better image reproduction [1], to better medical diagnoses [2]  and remote sensing [3]). The overarching aim of this project is to design an N-sensor (N>3) camera that can - with the addition of powerful inference algorithms - be used to measure spectral data.
It turns out that given an RGB or N-sensor camera response there are many possible spectra that might have induced it. In the first part of this project we will develop algorithms for estimating the set of all possible spectra from a camera response [4]. Clearly, if for a given camera its responses map to  spectra with small uncertainty (the set of possible spectra is on average small) then this camera is a good candidate for spectral measurement . In this project we wish to find cameras that recover spectra with small uncertainty.  This work will be carried out in close collaboration with Spectricity, our project partners.
The proposed research will take place in the Colour & Imaging Lab (C&IL) at the University of East Anglia (Norwich, UK) and is in collaboration with Spectricity. The C&IL is a centre of excellence for colour imaging. The successful candidate will join a team of 10 PhD students and 2 Post-doctoral research fellows.
For more information on the supervisor for this project, please go here: https://www.uea.ac.uk/stories/prof-graham-finlayson<https://www.findaphd.com/common/clickCount.aspx?theid=131098&type=184&DID=434&url=https%3a%2f%2fwww.uea.ac.uk%2fstories%2fprof-graham-finlayson>
Entry requirements: Acceptable first degree 1st or Masters in Computer Science, Physics, Maths or other numerate discipline.
References:
i) Lin Y.; Finlayson G.D. Physically plausible spectral reconstruction. Sensors. 2020, doi.org/10.3390/s20216399
ii) Zhang, Y.; Mou, X.; Wang, G.; Yu, H. Tensor-based dictionary learning for spectral CT reconstruction. IEEE Trans. Med Imaging 2016, 36, 142-154.
iii) Chen, C.; Li, W.; Su, H.; Liu, K. Spectral-spatial classification of hyperspectral image based on kernel extreme learning machine. Remote. Sens. 2014, 6, 5795-5814
iv) Morovic, P.; Finlayson, G.D. Metamer-set-based approach to estimating surface reflectance from camera RGB. J. Opt. Soc. Am. A 2006, 23, 1814-1822.


Deadline:  30th  Apr, 2021, Projects will start on the 1st Oct 2021

Details and how to apply:  https://www.findaphd.com/phds/project/designing-cameras-for-spectral-capture-finlaysong-u21spec/?p131098


Funding Notes: This PhD studentship is funded for 3 years by Spectricity. Funding is available to UK, EU & international applicants and includes tuition fees, an annual stipend of £15,744 in year 1, rising to £16,702 in year 3. The studentship length is 3 years.













-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20210423/78182fb2/attachment.html>