Spectral imaging is used to capture fine spectrum in the Visible, NIR, or SWIR regions of an object/scene to allow for sensing application such as material characterization and object recognition and tracking using the spectral signature as input data. We focus on hyper-spectral image reconstruction from compressive 2D measurements obtained by a snapshot spectrometer CTIS.
Key research topics include:
- Spectral image reconstruction from CTIS measurements.
- Spatial super-resolution of hyper-spectral images.
- Spectral image estimation from RGB data.
- Material characterization using spectral data.
Selected publications:
Material Characterization using a Compact Computed Tomography Imaging Spectrometer with Super-resolution Capability Proceedings Article
In: Proceedings of the 6th International Conference on Optical Characterization of Materials, OCM 2023, pp. 139–148, 2023.
Deep learning-based hyperspectral image reconstruction from emulated and real computed tomography imaging spectrometer data Journal Article
In: Optical Engineering, vol. 61, no. 5, pp. 053103–053103, 2022.
Joint Reconstruction and Super Resolution of Hyper-Spectral CTIS Images Conference
BMVC, 2022.
Semi-supervised deep learning techniques for spectrum reconstruction Proceedings Article
In: 2020 25th International Conference on Pattern Recognition (ICPR), pp. 7767–7774, IEEE 2021.