首页|中红外时间拉伸频率上转换高光谱成像系统研究

中红外时间拉伸频率上转换高光谱成像系统研究

扫码查看
高光谱成像技术是一种分析物质化学成分、物理性质和形态等信息的非接触式、无损检测方法.然而,中红外波段分子指纹谱测量,受限于探测器响应速度和固有噪声,传统的高光谱成像技术难以同时实现高速和高信噪比测量.基于时间拉伸频率上转换的光谱测量技术具有高测量速度、高光谱分辨率、宽光谱覆盖等优势,与高光谱成像技术结合后,可为快速获取样品种类与形态信息提供可靠手段.搭建了中红外时间拉伸频率上转换高光谱成像系统,同源产生1 047 nm泵浦光与1 550 nm信号光平均功率分别为2 W和100 mW,结合同步泵浦技术,可在两块周期性极化铌酸锂晶体中分别实现中红外脉冲的产生与频率上转换,完成将中红外波段分子指纹谱信息传递至近红外波段,解决了难以高速、低噪声探测中红外光谱信息的难题.通过对晶体温度与工作通道的调谐,中红外波段探测范围可覆盖2 700~3 900 nm,具有多样品检测能力.对频率上转换的近红外脉冲进行时间拉伸并与高光谱成像相结合,通过逐点扫描对比色皿中苯溶液吸收光谱及空间分布信息进行实测,光谱数据与傅里叶变换红外光谱仪测量结果高度吻合,能在8 s内完成600 μm×1 200μm空间区域的高光谱成像,单个像素点采集时间12.9 ns,光谱测量速度77.6 MSpectra·s-1,光谱分辨率5.8cm-1,证实了系统对光谱覆盖范围内液相分子的高速光谱测量及空间分辨能力.解决了传统高光谱测量方式在中红外波段存在响应速度慢、积分时间长、信噪比低的难点,能以107帧率的光谱刷新率完成多组分样品检测及其形态测量,为材料与生物成像分析等领域提供了新途径.
Research of Mid-Infrared Time-Stretch Frequency Upconversion Hyperspectral Imaging System
Hyperspectral imaging is a non-contact,non-destructive detection method to analyze substances'chemical composition,physical properties,and morphology.Limited by the response speed and inherent noise of the detector,it is difficult for traditional hyperspectral imaging techniques to achieve high-speed and high signal-to-noise detection of molecular fingerprint spectra in the mid-infrared band.With the advantages of high measurement speed,high spectral resolution,and wide spectral coverage,the spectroscopy technology based on time-stretch frequency upconversion provides a reliable method for rapidlyanalyzing the type and morphology of the samples when combined with hyperspectral imaging technology.In thispaper,a mid-infrared time-stretch frequency upconversion hyperspectral imaging system was constructed.The average power of the 1 047 nm pump pulse and the 1 550 nm signal pulse generated by the same laser source is 2 W and 100 mW,respectively.Using synchronous pump technology,mid-infrared pulses were generated in one periodically poled lithium niobate crystal,and frequency upconverted into near-infrared pulses in another.This process transferred the mid-infrared molecular fingerprint spectra to the near-infrared band,which can effectively address the problem of lacking high-speed and low-noise detectors in the mid-infrared band.By tuning the operating temperature and working channels of the crystal,the detection range of the system can cover 2 700~3 900 nm,enabling the measurement of multiple samples.Combining the time-stretch method with hyperspectral imaging technology,the benzene solution's absorption spectra and spatial distribution information in a colorimetric dish were measured through point-by-point scanning.The spectral data obtained highly matched the results from a Fourier transform infrared spectrometer.Moreover,the system could perform hyperspectral imaging of a 600 μm× 1 200 μm spatial region within 8 s.The acquisition time for a single pixel was 12.9 ns,and a spectral measurement speed of 77.6 MSpectra·s-1 and spectral resolution of 5.8 cm-'was achieved.These results verified the systemhas the potential to measure the spectra and spatial distributionof liquid molecules within the spectral coverage range with highspeed and highresolution.This paper solves the problems of slow response speed,long integration time,and low signal-to-noise ratio of traditional hyperspectral methods in the mid-infrared band.It enables the spectraldetection and morphological measurement of multi-component samples with a spectral refresh rate of 107 frames per second.It could provide a new approach for imaging analysis in material and biological fields.

Mid-infraredTime-stretchFrequency upconversionHyperspectral imaging

彭博、温兆阳、闻齐、刘婷婷、邢帅、武腾飞、闫明

展开 >

华东师范大学,精密光谱科学与技术国家重点实验室,上海 200062

华东师范大学重庆研究院,精密光学重庆市重点实验室,重庆 401121

北京长城计量测试技术研究所,北京 100095

中红外 时间拉伸 频率上转换 高光谱成像

国家自然科学基金项目重庆市自然科学基金项目中国航空工业集团公司北京长城计量测试技术研究所开放基金项目

12022411CSTB2022NSCQ-JQX0016JLKG2022001B003

2024

光谱学与光谱分析
中国光学学会

光谱学与光谱分析

CSTPCD北大核心
影响因子:0.897
ISSN:1000-0593
年,卷(期):2024.44(11)