基于RDCAN深度学习的三维共焦拉曼光谱高分辨成像方法

齐亚昆; 徐德民; 崔晗; 邱丽荣; 赵维谦

doi:null

您当前的位置：

首页 >

文章列表页 >

基于RDCAN深度学习的三维共焦拉曼光谱高分辨成像方法

光学传感与测量 | 更新时间：2025-10-09

- 基于RDCAN深度学习的三维共焦拉曼光谱高分辨成像方法
- 3D Confocal Raman Spectroscopic High-Resolution imaging method based on RDCAN deep learning model
- 光学技术 2025年51卷第4期页码：491-496
- 作者机构：
  
  北京理工大学光电学院 “复杂环境智能感测技术”工信部重点实验室，北京 100081
- 作者简介：
  
  齐亚昆（1998— ），男，硕士研究生，从事拉曼图像超分辨率研究。 qqiyk2022@163.com
  赵维谦（1966— ），男，教授，博士，从事精密光电成像与检测、智能光电感测、激光制造监测、纳米测控系统研究。 zwq669@126.com
- 基金信息：
  
  高分辨高灵敏激光差动共焦拉曼-LIBS光谱联用显微成像技术及仪器(2022YFC3300702);高性能激光差动共焦拉曼光谱显微成像关键技术及系统研究(U22A6006)
- DOI：
  中图分类号： TP394.1;TH691.9
- 收稿：2025-03-13，
  
  修回：2025-05-17，
  
  纸质出版：2025-07-20
- 稿件说明：
移动端阅览
齐亚昆,徐德民,崔晗等.基于RDCAN深度学习的三维共焦拉曼光谱高分辨成像方法[J].光学技术,2025,51(04):491-496.

QI Yakun,XU Demin,CUI Han,et al.3D Confocal Raman Spectroscopic High-Resolution imaging method based on RDCAN deep learning model[J].OPTICAL TECHNIQUE,2025,51(04):491-496.
齐亚昆,徐德民,崔晗等.基于RDCAN深度学习的三维共焦拉曼光谱高分辨成像方法[J].光学技术,2025,51(04):491-496. DOI：

QI Yakun,XU Demin,CUI Han,et al.3D Confocal Raman Spectroscopic High-Resolution imaging method based on RDCAN deep learning model[J].OPTICAL TECHNIQUE,2025,51(04):491-496. DOI：

摘要

三维共焦拉曼光谱成像凭借独特的分子特异性和非破坏性原位分析能力，在材料、生物医学等领域得到了广泛应用。然而，其几何形貌分辨率不足的固有缺陷制约了更深层次的应用拓展。因此，提出一种基于残差密集通道注意力网络（RDCAN）的超分辨率图谱成像方法，实现三维共焦拉曼光谱图像到几何形貌的跨模态重建。首先设计端到端的深度学习网络，以残差密集网络为基础，充分融合局部与全局多尺度特征，并引入通道注意力机制动态优化特征权重分配，增强网络的特征提取能力；其次，利用自主搭建的激光差动共焦-拉曼光谱系统采集数据，其中共焦拉曼模块采集的三维光谱数据作为网络输入，差动共焦模块同步获取的几何形貌数据作为真值进行训练，最终实现几何形貌的重建。实验结果表明，重建结果的峰值信噪比达到27.4 dB，结构相似度达到0.98，与差动共焦显微镜测量结果接近。与传统方法相比，方法无需额外硬件即可在传统三维共焦拉曼光谱成像基础上扩展高分辨几何成像能力，实现光谱-几何形貌同源融合，为解决其在几何形貌分辨率上的局限性问题提供了创新解决方案。

Abstract

Three-dimensional confocal Raman spectroscopic imaging is widely used in materials science and biomedicine due to its unique molecular selectivity and non-destructive

in-situ analytical capabilities. However

its inherent limitation of inadequate geometric resolution hinders the development of more advanced applications. In this study

a super-resolution mapping method that employs a Residual Dense Channel Attention Network （RDCAN） to transform 3D confocal Raman spectral images into geometric topography is proposed. An end-to-end deep learning model is designed based on a Residual Dense Network （RDN）

which effectively integrates both local and global multi-scale features. The model incorporates a channel attention mechanism to dynamically optimize feature weight allocation

thereby enhancing feature extraction capability. A self-developed laser differential confocal Raman spectroscopy system is used for data acquisition. In this system

3D spectral data obtained by the confocal Raman module serve as input to the network

while the geometric topography data captured by the differential confocal module serve as ground truth for training. The reconstruction of geometric topography is ultimately realized. Experimental results show that the reconstructed geometric topography achieves a peak signal-to-noise ratio （PSNR） of 27.4dB and a structural similarity index （SSIM） of 0.98

closely matching the measurements obtained by differential confocal microscopy. Compared with traditional methods

the proposed approach enhances high-resolution geometric imaging based on conventional 3D confocal Raman spectroscopy without requiring additional hardware. It also enables spectral-geometric homologous fusion

offering an innovative solution to overcome the limitations in geometric resolution.

关键词

Keywords

references

Pezzotti G , Higashino M , Tsuji K , et al . Non-destructive three-dimensional raman analysis of domain orientation in barium titanate single-crystal [J ] . Journal of the European Ceramic Society , 2010 , 30 （ 2 ）: 199 — 204 .

Høgset H , Horgan C C , Armstrong J P K , et al . In vivo biomolecular imaging of zebrafish embryos using confocal raman spectroscopy [J ] . Nature Communications , 2020 , 11 （ 1 ）: 6172 .

Chen M , Wang J , Li D , et al . High-speed and wide-field 3D raman imaging using flower-like gap-enhanced raman tags [J ] . Applied Physics Letters , 2024 , 124 （ 12 ）: 123701 .

Pilarczyk M , Rygula A , Kaczor A , et al . A novel approach to investigate vascular wall in 3D: Combined Raman spectroscopy and atomic force microscopy for aorta en face imaging [J ] . Vibrational Spectroscopy , 2014 , 75 : 39 — 44 .

Verma P . Tip-enhanced raman spectroscopy: Technique and recent advances [J ] . Chemical Reviews , 2017 , 117 （ 9 ）: 6447 - 6466 .

Pei W , Hu R , Liu H , et al . Advanced raman spectroscopy for nanoplastics analysis: Progress and perspective [J ] . TrAC Trends in Analytical Chemistry , 2023 , 166 : 117188 .

Wille G , Lahondere D , Schmidt U , et al . Coupling SEM-EDS and confocal raman-in-SEM imaging: A new method for identification and 3D morphology of asbestos-like fibers in a mineral matrix [J ] . Journal of Hazardous Materials , 2019 , 374 : 447 — 458 .

Sk M H , Agrawal S , Casford M T L , et al . Contrast inversion of ferrite and pearlite in a sweet corrosion environment [J ] . Applied Surface Science , 2023 , 635 : 157744 .

Cui H , Wang Y , Qiu L , et al . Synchronous nanoscale topographic and chemical mapping by differential-confocal controlled raman microscopy [J ] . Photonics Research , 2020 , 8 （ 9 ）: 09001441 .

Qiu L , Su Y , Xu K M , et al . A high-precision multi-dimensional microspectroscopic technique for morphological and properties analysis of cancer cell [J ] . Light: Science & Applications , 2023 , 12 （ 1 ）: 129 .

Zhao W , Du S . Spectral–spatial feature extraction for hyperspectral image classification: A dimension reduction and deep learning approach [J ] . IEEE Transactions on Geoscience and Remote Sensing , 2016 , 54 （ 8 ）: 4544 — 4554 .

Mei X , Pan E , Ma Y , et al . Spectral-spatial attention networks for hyperspectral image classification [J ] . Remote Sensing , 2019 , 11 （ 8 ）: 963 .

Yu W , Zhang M , He Z , et al . Convolutional two-stream generative adversarial network-based hyperspectral feature extraction [J ] . IEEE Transactions on Geoscience and Remote Sensing , 2022 , 60 : 1 — 10 .

Wei L , Wang K , Lu Q , et al . Crops fine classification in airborne hyperspectral imagery based on multi-feature fusion and deep learning [J ] . Remote Sensing , 2021 , 13 （ 15 ）: 2917 .

郑宗生 , 刘海霞 , 王振华 , 等 . 改进3D-cnn的高光谱图像地物分类方法 [J ] . 自然资源遥感 , 2023 , 35 （ 2 ）: 105 — 111 .

Zheng Zongsheng , Liu Haixia , Wang Zhenhua , et al . Improved 3D-CNN-based method for surface feature classification using hyperspectral images [J ] . Remote Sensing for Natural Resources , 2023 , 35 （ 2 ）: 105 — 111 .

Hu W S , Li H C , Pan L , et al . Spatial-spectral feature extraction via deep ConvLSTM neural networks for hyperspectral image classification [J ] . IEEE Transactions on Geoscience and Remote Sensing , 2020 , 58 （ 6 ）: 4237 — 4250 .

Chen H , Li J , Zhang J , et al . HSI-IPGAN: Hyperspectral image inpainting via generative adversarial network [A ] . Research Square , 2023 .

Hou J , Zhu Z , Hou J , et al . Deep posterior distribution-based embedding for hyperspectral image super-resolution [J ] . IEEE Transactions on Image Processing , 2022 , 31 : 5720 — 5732 .

Qu J , He J , Dong W , et al . S2CycleDiff: Spatial-Spectral-Bilateral Cycle-Diffusion Framework for Hyperspectral Image Super-resolution [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2024 , 38 （ 5 ）: 4623 — 4631 .

Zhang Y , Li K , Li K , et al . Image super-resolution using very deep residual channel attention networks [M ] . Ferrari V, Hebert M, Sminchisescu C, et al. Computer Vision – ECCV 2018. Cham : Springer International Publishing , 2018 : 294 — 310 .

Horgan C C , Jensen M , Nagelkerke A , et al . High-throughput molecular imaging via deep-learning-enabled raman spectroscopy [J ] . Analytical Chemistry , 2021 , 93 （ 48 ）: 15850 — 15860 .

Zhang Y , Tian Y , Kong Y , et al . Residual dense network for image super-resolution [A ] . arXiv , 2018 .

Wang T D , Mandella M J , Contag C H , et al . Dual-axis confocal microscope for high-resolution in vivo imaging [J ] . Optics Letters , 2003 , 28 （ 6 ）: 414 — 416 .

Zhao W , Sun Y , Wang Y , et al . Three-dimensional super-resolution correlation-differential confocal microscopy with nanometer axial focusing accuracy [J ] . Optics Express , 2018 , 26 （ 12 ）: 15759 — 15768 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于波前预测模型的波前校正方法

基于多步匹配的超声多目标跟踪方法与系统构建

基于残差注意力机制的离轴数字全息重建

长距离光纤通信系统的非线性损伤补偿方法