Arficial intelligence assisted spectral unmixing for illuminated manuscripts
CRC - Centre de Recherche sur la Conservation Laboratory

Back to list

Context
Laboratory
Contract
Fixed-term
Duration
1 year
Level
Post-doctoral degree (or eq.)
Taking office
ASAP
Position
Post-doc fellowhsip
Thematic
Spectral unmixing
City
Paris
Published on 29 May 2024

Hiring Policy

Starting date : before end of 2024
Duration : 1 year
Gross monthly salary : 2936 €
Application deadline : June 30, 2024

Subject

The goal of this project is to use machine learning (ML) algorithms to model how the reflectance 
spectra of complex mixtures of colored materials from historical artworks, specifically illuminations 
and/or decorated letters from ancient manuscripts, depend on their chemical composition. Covering 
a wide range of energy from visible (VIS) to mid-infrared (mid-IR) coupled with X-ray fluorescence 
spectroscopy (XRF), a spectral database of reference inorganic and organic materials will be built. 
Subsequently, a supervised ML algorithm will be trained on the database, and the obtained ML 
model will be capable of simultaneously predicting the composition of the colored layer and the 
abundance of materials present in the mixture. Next, we will use the trained ML model to generate a 
greater variety of artificial reflectance spectra, thus expanding the existing data set. A second ML 
model will then be trained on the expanded data set to understand how the reflectance spectrum 
correlates with the composition of the pictorial mixture. This second model will be applied to the 
analysis of historical manuscripts. The candidate is expected to develop an approach that will offer
an efficient and general method for onsite determination of the composition of historical 
illuminations based on their spectral response.

Header 2...

Skills required
• Python programming
• Knowledge in machine learning algorithms (most specifically encoder-decoder NN and 
transfer learning would be appreciated)
• Exper-se in reflectance and X-ray fluorescence imaging spectroscopies (optional)
• Experience with multispectral dataset analyses (optional)