Spatiotemporal blue noise coded aperture design for multi-shot compressive spectral imaging.

Area of Science:

• Optics and Photonics
• Image Processing
• Computational Imaging

Background:

• Multi-shot coded aperture snapshot spectral imaging (CASSI) captures spectral data using compressive measurements.
• Compressed sensing (CS) reconstructs 3D spectral data cubes, with multiple snapshots improving reconstruction quality.
• Coded aperture structure is critical for CASSI performance, influencing measurement needs and reconstruction quality.

Purpose of the Study:

• To design optimal coded aperture structures for multi-shot CASSI systems.
• To improve spectral image reconstruction by optimizing the CASSI sensing matrix.
• To enhance the signal-to-noise ratio (SNR) and structural similarity (SSIM) of reconstructed spectral data.

Main Methods:

• Established CASSI matrix representation using random projections.
• Determined the restricted isometry property (RIP) of CASSI projections based on coded aperture entries.
• Designed spatiotemporal blue noise (BN) coded apertures to satisfy RIP with high probability.
• Presented an algorithm for implementing BN ensembles.

Main Results:

• Demonstrated that optimal coded aperture structures significantly improve spectral image reconstruction.
• Quantified improvements in peak signal-to-noise ratio (PSNR) and SSIM compared to traditional random apertures.
• Validated findings through extensive simulations and testbed implementation.

Conclusions:

• Spatiotemporal blue noise (BN) coded apertures offer superior performance for multi-shot CASSI.
• Optimized aperture design leads to more accurate and higher-quality spectral data reconstruction.
• This work provides a pathway for enhanced spectral imaging applications.