Cascaded compressed-sensing single-pixel camera for high-dimensional optical imaging

Area of Science:

• Optical Sciences
• Computational Imaging
• Photonics

Background:

• Single-pixel detectors offer speed, sensitivity, and low cost but struggle with high-dimensional imaging due to their zero-dimensional nature.
• Acquiring high-dimensional optical field information is a fundamental challenge for single-pixel sensors.
• Compressibility of natural scenes in multiple domains is key to efficient data acquisition.

Approach:

• Developed a cascaded compressed-sensing single-pixel camera.
• Implemented a multi-stage measurement scheme to sequentially reduce data dimensionality.
• Leveraged scene compressibility across multiple domains for efficient data capture.

Key Points:

• Demonstrated a novel approach to overcome the limitations of single-pixel detectors in imaging.
• Enabled tunable single-pixel full-waveform hyperspectral light detection and ranging (LIDAR).
• Achieved highly efficient data acquisition by exploiting multi-domain scene compressibility.

Conclusions:

• The cascaded compressed-sensing camera significantly enhances the imaging capabilities of single-pixel detectors.
• This method opens new possibilities for advanced optical sensing and measurement.
• Paved the way for novel applications in hyperspectral imaging and LIDAR.