Design Optimization For Snapshot Spectral Imaging

Snapshot spectral imaging enables to reconstruct spectral images from a multiplexed single-shot measurement. Since an inversion is required to form the spectral images computationally, quantitative characterization of their performance is essential to optimize the design. In this paper, we analyze the optimal design of a snapshot spectral imaging technique. This snapshot multi-spectral imaging technique uses a diffractive lens called generalized photon sieve, and various design choices affect its imaging performance. Here we use mutual coherence and t%-average mutual coherence as optimality metrics, which provide quantitative performance measures that are independent of the spectral images to be reconstructed, regularization choices, and signal-to noise ratio. Through numerical simulations, the optimal choices for different design considerations are investigated using these metrics for an application in solar imaging. The results suggest that these optimality metrics provide good measures for reconstruction quality.