Research Projects

ChemHouse aims at studying all chemometric research issues, primarily in the application areas of its partners (Digital Agriculture, Waste Sorting, Implementation of SPIR, etc.). This research relies on the expertise of its members and on collaborations with its academic partners.  The following main topics are studied:

• Improvement of multivariate calibration techniques :
  •  variables selection
  • local regressions and big-data
  • weighted regressions
  • spectral pre-treatments
  • calibration robustness
• Hyperspectral image processing :
  • spatial / spectral combination
  • calibration transfer
  • spectral unmixing
• Multi-block and multi-ways analysis :
  • pre-processing and ensemble methods
  • variable selection
  • massive data processing
  • analyses of variance of multivariate data
  • time series processing

ChemHouse is the forge of the CheMoocs MOOC, the ChemFlow software and the ChemData database.

ChemHouse is one of the promoters of the ChemoMics project.

ChemHouse theses:

• Alexandre Mallet :
  • "Freeing ourselves from the effect of water to enable spectroscopic characterization of wet organic waste"
• Maxime Metz:
  • "Application of big data methods for the improvement of local PLS algorithms in chemometrics"
• Jhon Buondia:
  • "Digital Process Control by online analysis of total effluent in near infrared to determine the product quality of cuts". - Eva Lopez Fornieles: "Potential of time series of multispectral satellite and radar data to design spatialized variables of vine development at regional scale: application to the estimation of growth and its cessation their interest for a water stress model".

Researchers' welcoming:

Google Scholar Profile