基于压缩感知的AOTF光谱测量方法
作者:
作者单位:

1.中国科学院上海技术物理研究所 空间主动光电技术重点实验室,上海 200083;2.中国科学院大学,北京 100049

作者简介:

通讯作者:

中图分类号:

TP751

基金项目:


An acousto-optic tunable filter spectral measurement method based on compressed sensing
Author:
Affiliation:

1.Key Laboratory of Space Active Opto-Electronics Technology, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, China;2.University of Chinese Academy of Sciences, Beijing 100049, China

Fund Project:

Supported by the National Natural Science Foundation of China (61605231)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    基于声光可调谐滤波器(AOTF)的光谱仪器已经在生物医学、农业、航空航天等领域中广泛应用。然而,传统的AOTF光谱仪很难在保持光谱分辨率和减少采样次数的同时,还实现系统光通量的增加。针对上述问题,本文提出了一种基于压缩感知理论的AOTF光谱测量方法,利用AOTF可多频驱动的特点,在光谱维度上实现稀疏随机编码复合光信号调制,可利用单元或面阵探测器顺次记录完成压缩采样,再通过压缩感知重建算法获得目标光谱曲线或光谱图像数据立方体。为了验证本方法的有效性,我们利用实际测量得到AOTF光谱响应带宽数据,构建传感矩阵,以展宽光谱为恢复目标,仿真了压缩采样和目标数据重构效果。仿真结果表明,该方法可以通过202次压缩采样,重构得到512个波长点的光谱数据,光谱数据采样率和压缩比为0.39。此采样率下,本方法可以高精度恢复光谱曲线,PSNR指标达到41.75dB,SAM和GSAM指标为0.9998和0.9754。多频同时驱动下,系统光通量平均提升了5倍。与传统逐波长点扫描的采样方式相比,该方法能够在保持原有光谱分辨率的前提下减少总采样次数,提高系统的光通量,同时还压缩了光谱数据,在微弱信号检测、物质快速识别以及光谱数据传输和存储等领域具有十分重要的意义。

    Abstract:

    Acousto-Optic Tunable Filter (AOTF) based spectroscopy instruments have been widely applied in biomedical, agricultural, aerospace, and other fields. However, the conventional AOTF spectrometers struggle to achieve increased system luminous flux while maintaining spectral resolution and reducing the number of samples. To address the above problems, this paper proposes an AOTF spectral measurement method based on the compressed sensing (CS) theory. Sparse randomly coded composite optical signal modulation in the spectral domain using the multi-frequency acousto-optic diffraction of AOTF. A modulated composite optical signal is obtained in the spectral domain and recorded sequentially using a single-element detector or a focal-plane detector array. The original spectrum or spectral image data cube is then obtained by using compressed sensing reconstruction algorithms. In order to verify the effectiveness of the present method, we constructed a sensing matrix using the actual measured AOTF spectral response bandwidth data and simulated the effect of compressed sampling and target data reconstruction with the spreading spectrum as the recovery target. The simulation results show that the method can reconstruct the spectral data of 512 wavelength points with 202 compressed samples, and the spectral data sampling rate and compression ratio is 0.39. Under this sampling rate, the method can recover the spectral curve with high accuracy, and the PSNR index reaches 41.75 dB, and the SAM and GSAM indexes are 0.9998 and 0.9754. With the simultaneous multi-frequency drive, the system optical throughput is improved by a factor of 5 on average. Compared with the traditional wavelength-by-wavelength point scan sampling method, this method can reduce the total number of samples and improve the luminous flux of the system while maintaining the original spectral resolution, and also compressing the spectral data, which is of great importance in the fields of weak signal detection, rapid identification of substances, and spectral data transmission and storage.

    参考文献
    相似文献
    引证文献
引用本文

姬忠鹏,桂裕华,李津宁,谭永健,杨秋杰,王建宇,何志平.基于压缩感知的AOTF光谱测量方法[J].红外与毫米波学报,2023,42(1):111~121]. JI Zhong-Peng, GUI Yu-Hua, LI Jin-Ning, TAN Yong-Jian, YANG Qiu-Jie, WANG Jian-Yu, HE Zhi-Ping. An acousto-optic tunable filter spectral measurement method based on compressed sensing[J]. J. Infrared Millim. Waves,2023,42(1):111~121.]

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2022-03-16
  • 最后修改日期:2023-01-06
  • 录用日期:2022-05-30
  • 在线发布日期: 2023-01-06
  • 出版日期: 2023-02-25
文章二维码