Compressive spectral testbed imaging system based on thin-film color-patterned filter arrays.

Area of Science:

• Optics and Photonics
• Image Processing
• Materials Science

Background:

• Traditional spectral imaging requires numerous measurements, limiting efficiency.
• Compressive sensing offers a path to reduce data acquisition requirements.
• Thin-film filter arrays present a promising approach for compact spectral imagers.

Purpose of the Study:

• To demonstrate a thin-film patterned filter array-based compressive spectral imager.
• To detail the optical design and implementation of the system.
• To evaluate the performance and explore filter distributions for optimized spectral data capture.

Main Methods:

• Design and fabrication of a micrometer pitch size thin-film patterned filter array with three wavelengths.
• Implementation of a single-step three-dimensional spatial-spectral coding technique.
• Evaluation using a commercially available spectrometer and visual assessment of reconstructed images.

Main Results:

• Successful demonstration of a thin-film patterned filter array compressive spectral imager.
• The system achieves higher flexibility in voxel multiplexing.
• Performance evaluation shows comparable results to traditional methods, with exploration of random and optimized filter structures.

Conclusions:

• Thin-film patterned filter arrays are effective for compressive spectral imaging.
• The demonstrated system offers a flexible and efficient approach to multispectral data acquisition.
• Further optimization of filter distributions can enhance reconstructed image quality.