Fast and high-quality single-pixel imaging.

Area of Science:

• Optics and Photonics
• Computational Imaging
• Signal Processing

Background:

• Conventional single-pixel-imaging (SPI) suffers from significant image quality degradation under low sampling rates.
• Existing sampling strategies like circular paths or variable density random sampling are suboptimal for low-data acquisition scenarios.

Purpose of the Study:

• To develop an efficient sampling method and a high-quality, real-time image reconstruction strategy for SPI.
• To overcome the limitations of low sampling rates in SPI and reduce reconstruction time.

Main Methods:

• A novel sampling strategy that prioritizes significant Fourier spectrum coefficients based on image spectrum distribution.
• An image reconstruction strategy combining image sparsity and alternating direction optimization in the gradient space.

Main Results:

• The proposed method significantly improves imaging quality even at low sampling rates.
• Real-time image reconstruction is achieved on the millisecond timescale, outperforming state-of-the-art techniques.

Conclusions:

• The developed SPI technique offers superior image quality and reconstruction speed.
• This approach addresses critical challenges in low-sampling-rate imaging and real-time applications.