Terahertz image reconstruction based on compressed sensing and inverse Fresnel diffraction

Area of Science:

• Terahertz (THz) imaging
• Computational imaging
• Optics and photonics

Background:

• Single-pixel THz imaging traditionally requires extensive data acquisition, limiting its practical applications.
• Compressed sensing (CS) has been explored to accelerate THz imaging by reducing measurements.
• Diffraction effects can degrade the resolution and sharpness of reconstructed THz images.

Purpose of the Study:

• To develop and validate a novel single-pixel THz imaging method combining compressed sensing (CS) with dynamic masks and inverse Fresnel diffraction (IFD).
• To enhance the speed and image quality of THz spectral imaging systems.
• To improve the overall practicability of single-pixel THz imaging.

Main Methods:

• Employed a compressed sensing (CS) approach utilizing photoinduced dynamic masks to capture THz diffraction field data in the time domain.
• Applied an inverse Fresnel diffraction (IFD) integral to correct for diffraction artifacts and sharpen the spectral image.
• Validated the combined CS and IFD algorithms through both numerical simulations and experimental setups.

Main Results:

• Successfully reconstructed sharp THz spectral images by integrating CS and IFD algorithms.
• Demonstrated the compatibility and effectiveness of the proposed method in both simulated and experimental scenarios.
• Analyzed the impact of reducing measurement numbers and varying diffraction distances on reconstruction quality.

Conclusions:

• The combination of CS with dynamic masks and IFD offers a viable strategy for high-quality, rapid single-pixel THz imaging.
• This approach significantly enhances the practicability of THz spectral imaging by overcoming limitations of speed and diffraction.
• The study provides a novel framework for advancing single-pixel THz imaging technologies.