Dual-camera snapshot spectral imaging with a pupil-domain optical diffuser and compressed sensing algorithms.

Area of Science:

• Optics and Photonics
• Image Processing
• Computational Imaging

Background:

• Spectral imaging captures detailed spectral information of a scene.
• Previous methods using a single monochromatic camera with a pupil diffuser have limitations.
• Improving spatial and spectral resolution in spectral imaging remains a challenge.

Purpose of the Study:

• To develop an improved snapshot spectral imaging method for the visible spectrum.
• To leverage the complementary information from two different camera types for enhanced spectral reconstruction.
• To overcome the limitations of existing single-camera spectral imaging techniques.

Main Methods:

• A novel snapshot spectral imaging system employing two side-by-side digital cameras.
• One camera is a standard red-green-blue (RGB) camera.
• The second camera is monochromatic and equipped with a dispersive diffractive diffuser at the lens pupil.

Main Results:

• The combined data from both cameras provide richer spatial and spectral information.
• This dual-camera approach stabilizes the reconstruction of the spectral cube.
• Compressed sensing algorithms applied to the combined data yield improved reconstructed image quality.

Conclusions:

• The proposed two-camera spectral imaging method offers a more robust and higher-quality solution.
• Integrating RGB and monochromatic spectral data effectively addresses underdetermined reconstruction problems.
• This technique advances snapshot spectral imaging capabilities for various applications.