Data rectification and decoding of a microlens array-based multi-spectral light field imaging system

Spectral light field imaging systems have gained emerging attention for their capacity to capture multidimensional spatial-spectral data in a single snapshot. However, the spectral data poses significant challenges to image decoding methods for existing light field imaging due to the aberrations and spectral aliasing. To resolve these challenges, in this study, a microlens array-based light field spectral imaging system is designed and assembled. A novel joint decoding method incorporating aberration correction for spatial-spectral data is proposed. The proposed method is implemented through a novel set of aberrations assessment and correction procedures for both local and global non-uniform aberrations. An improved light field digital refocusing method is used to overcome the spectral aliasing in refocusing the spectral light field images, enabling the reconstruction of depth and spectral intensity distribution of a target scene. Experimental verification is conducted on a uniform light source and real-world imagery. Results demonstrate that the proposed method can provide accurate aberration correction and reconstruction of the spectral intensity distribution of the targeted scene under different depths, thus showing the effectiveness of the proposed light field spectrometer and spatial-spectral joint decoding method for real-world imagery.