PHOTO : Systems - Adaptive hyperspectral imager

This research is based on a co-design approach where the data acquisition and reconstruction algorithms are designed simultaneously to the optical device. This multidisciplinary collaboration between photonicians (PHOTO team, LAAS), roboticians (RIS team, LAAS) and signal processors (SISU team, IRAP) aims to develop a hyperspectral image acquisition system that is both simple (i.e. requiring little or no post-processing) and capable of adapting to different types of scenes or use scenarios.

Schematic of the adaptive hyperspectral imager.

The objective is to allow different acquisition strategies to acquire all or part of the hyperspectral cube with adjustable trade-offs in terms of spatial, spectral and temporal resolutions. Our imager is based on a matrix of micro-mirrors (DMD) between two 4f lines, the whole acting as a programmable spatial-spectral filter. In addition to the instrumenting challenges, this project requires the definition of new data acquisition strategies that include innovative data acquisition, post-processing and analysis.

Collaborations:

 RIS group (autonomous robots) from LAAS,  SISU group (signal processing) from IRAP, Airbus Defence & Space.

Projects:

HyperHolo, Hyadim, Sir4Hyadim, ANR ImHypAd

Related publications:

E. Hemsley, I. Ardi, T. Rouvier, S. Lacroix, H. Carfantan, and A. Monmayrant,
"Fast reconstruction of hyperspectral images from coded acquisitions using a separability assumption,"
Optics Express 30, p. 8174, 2022.

E. Hemsley, S. Lacroix, H. Carfantan, and A. Monmayrant,
"Calibration of programmable spectral imager with dual disperser architecture,"
Optics Communications 468, p. 125767, 2020.

S. McGregor, S. Lacroix, and A. Monmayrant,
“Adaptive hyperspectral imager: design, modeling, and control,”
Journal of Optics 17, no. 8, p. 85607, 2015.