High-speed hyperspectral imaging enabled by compressed sensing in time domain

Hyperspectral imaging(HSI)is a powerful tool widely used for various scientific and industrial applications due to its ability to provide rich spatiospectral information.However,in exchange for multiplex spectral information,its image acquisition rate is lower than that of conventional imaging,with up to a few colors.In particular,HSI in the infrared region and using nonlinear optical processes is impractically slow because the three-dimensional(3D)data cube must be scanned in a point-by-point manner.In this study,we demonstrate a framework to improve the spectral image acquisition rate of HSI by integrating time-domain HSI and compressed sensing.Specifically,we simulated broadband coherent Raman imaging at a record high frame rate of 25 frames per second(fps)with 100 pixels×100 pixels,which is 10×faster than that of previous work,based on an experimentally feasible sampling scheme utilizing 3D Lissajous scanning.

Ultrafast impulsive Raman spectroscopy across the terahertz-fingerprint region

Broadband Raman spectroscopy(detection bandwidth>1000 cm^(−1))is a valuable and widely used tool for understanding samples via label-free measurements of their molecular vibrations.Two important Raman spectral regions are the chemically specific“fingerprint”(200 to 1800 cm^(−1))and“low-frequency”or“terahertz”(THz)(<200 cm^(−1);<6 THz)regions,which mostly contain intramolecular and intermolecular vibrations,respectively.These two regions are highly complementary;broadband simultaneous measurement of both regions can provide a big picture comprising information about molecular structures and interactions.Although techniques for acquiring broadband Raman spectra covering both regions have been demonstrated,these methods tend to have spectral acquisition rates<10 spectra∕s,prohibiting high-speed applications,such as Raman imaging or vibrational detection of transient phenomena.Here,we demonstrate a single-laser method for ultrafast(24,000 spectra∕s)broadband Raman spectroscopy covering both THz and fingerprint regions.This is achieved by simultaneous detection of Sagnac-enhanced impulsive stimulated Raman scattering(SE-ISRS;THz-sensitive)and Fourier-transform coherent anti-Stokes Raman scattering(FT-CARS;fingerprint-sensitive).With dual-detection impulsive vibrational spectroscopy,the SE-ISRS signal shows a>500×enhancement of<6.5 THz sensitivity compared with that of FT-CARS,and the FT-CARS signal shows a>10×enhancement of fingerprint sensitivity above 1000 cm^(−1)compared with that of SE-ISRS.

The marriage of coherent Raman scattering imaging and advanced computational tools

Coherent Raman scattering microscopy can provide high-contrast tissue and single-cell images based on the inherent molecular vibrations of the sample.However,conventional techniques face a three-way trade-off between Raman spectral bandwidth,imaging speed,and image fidelity.Although currently challenging to address via optical design,this trade-off can be overcome via emerging computational tools such as compressive sensing and machine learning.