Principles and applications of high-speed single-pixel imaging technology
- 169 Downloads
Single-pixel imaging (SPI) technology has garnered great interest within the last decade because of its ability to record high-resolution images using a single-pixel detector. It has been applied to diverse fields, such as magnetic resonance imaging (MRI), aerospace remote sensing, terahertz photography, and hyperspectral imaging. Compared with conventional silicon-based cameras, single-pixel cameras (SPCs) can achieve image compression and operate over a much broader spectral range. However, the imaging speed of SPCs is governed by the response time of digital micromirror devices (DMDs) and the amount of compression of acquired images, leading to low (ms-level) temporal resolution. Consequently, it is particularly challenging for SPCs to investigate fast dynamic phenomena, which is required commonly in microscopy. Recently, a unique approach based on photonic time stretch (PTS) to achieve high-speed SPI has been reported. It achieves a frame rate far beyond that can be reached with conventional SPCs. In this paper, we first introduce the principles and applications of the PTS technique. Then the basic architecture of the high-speed SPI system is presented, and an imaging flow cytometer with high speed and high throughput is demonstrated experimentally. Finally, the limitations and potential applications of high-speed SPI are discussed.
Key wordsCompressive sampling Single-pixel imaging Photonic time stretch Imaging flow cytometry
- Bosworth, B.T., Foster, M.A., 2014. High-speed flow imaging utilizing spectral-encoding of ultrafast pulses and compressed sensing. OSA Techn. Dig., Paper ATh4P.3. https://doi.org/10.1364/CLEO_AT.2014.ATh4P.3Google Scholar
- Chen, C.L.F., Mahjoubfar, A., Jalali, B., 2015. Optical data compression in time stretch imaging. PLOS ONE, 10(4): 0125106. https://doi.org/10.1371/journal.pone.0125106Google Scholar
- Takhar, D., Laska, J., Wakin, M.B., et al., 2006. A new compressive imaging camera architecture using opticaldomain compression. SPIE, 6065:43–52. https://doi.org/10.1117/12.659602Google Scholar