Snapshot fiber spectral imaging using speckle correlations and compressive sensing

Area of Science:

• Optics and Photonics
• Remote Sensing Technology
• Computational Imaging

Background:

• Snapshot spectral imaging offers rapid data acquisition for remote sensing but faces miniaturization challenges with traditional grating/prism technologies.
• Current portable hyperspectral imaging systems are often bulky due to the need for stabilized scanning mechanisms.
• Developing compact, efficient spectral imagers is crucial for advancing remote sensing capabilities.

Purpose of the Study:

• To demonstrate a novel, compact spectral imager using multicore multimode fiber.
• To achieve sub-nanometer spectral resolution for hyperspectral imaging applications.
• To develop efficient data acquisition and reconstruction methods for spectral imaging.

Main Methods:

• Utilizing a multicore multimode fiber to encode spectral information onto a monochrome CMOS camera.
• Characterizing wavelength-dependent speckle patterns across thousands of fiber cores.
• Employing a clustering algorithm with l1-minimization for sparse spectral data reconstruction.

Main Results:

• Successfully demonstrated a compact spectral imager with sub-nanometer resolution.
• Characterized spectral signatures of up to 3000 fiber cores over a broad wavelength range.
• Validated accurate reconstruction of broadband spectral information using compressive sensing techniques.

Conclusions:

• Multicore fiber technology provides a viable pathway for developing compact and portable hyperspectral imagers.
• The proposed method enables efficient data acquisition and reconstruction, overcoming limitations of traditional spectral imaging systems.
• This approach holds significant potential for advancing remote sensing and other applications requiring miniaturized spectral analysis.