Signal, Image and Sound - SIMS Research group

SIMS Signal, IMage et Son
 

Research field

The activity of the Signal, Image and Sound research group  relates to the statistical processing of signals and images.

The research carried out by the members of the team revolves around methodological questions inspired by applied problems from a wide range of fields, such as physics, engineering and industrial instrumentation. Changing requirements for measurement systems, in terms of information accuracy, has led to the diversification of the data acquisition methods.

However, this multi-modality produces increasingly large volumes of data whose processing by conventional methods often encounters obstacles that result either in a computation time that is too long or requiring too much memory space. At the same time, the growth of the computing processor's capacities offers an opportunity for the processing of this data, provided that new methods are available which ensure better "algorithm-architecture" matching and which are able to exploit any structure intrinsic to the data. The definition of such algorithms, based on proven mathematical tools, is the main challenge faced by the SIMS team.
 

Research topics

• Decision through learning
• Reverse problems
• Optimization and Bayesian simulation
• Preferred applications are the brain-computer interface
• Multimedia content analysis
• Non-destructive testing and airborne remote sensing