基于 EMD与小波稀疏去噪的拉曼光谱峰检测算法
CSTR:
作者:
作者单位:

(1.重庆理工大学重庆市光纤传感与光电检测重点试验室,重庆 400054; 2.重庆赛宝工业技术研究院有限公司,重庆 401332)

作者简介:

汤斌(1985-),男,重庆市铜梁区人,博士,副教授,硕士生导师,研究领域为光谱分析、信号处理、环境水质分析。

通讯作者:

中图分类号:

TP391.41

基金项目:

重庆市科学技术局科研机构绩效激励引导专项项目(CSTB2023JXJL-YFX0004).


Raman Spectral Peak Detection Algorithm Based on EMD and Wavelet Sparse Denoising
Author:
Affiliation:

(1. Chongqing Key Laboratory of Optical Fiber Sensing and Photoelectric Detection,Chongqing University of Technology,Chongqing 400054,CHN;2. Chongqing Saibao Industrial Technology Research Institute Co. Ltd., Chongqing 401332,CHN)

Fund Project:

  • 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:

    针对拉曼光谱检测中弱峰灵敏度低、重叠峰分辨困难及噪声干扰显著等技术瓶颈,文章提出一种基于经验模态分解为(EMD)与小波稀疏去噪的拉曼光谱峰检测算法。该算法构建了一个三级自适应滤波与 EMD-小波联合去噪框架,通过经验模态分解实现噪声与特征成分的频带分离,结合 sym6小波基阈值收缩抑制模态混叠噪声。创新性引入分数阶微分(α=0. 3)增强弱峰边缘特征,设计多尺度检测与动态决策机制,以局部信噪比约束(SNR>1. 5)实现冗余峰剔除。试验结果表明,所提算法在复杂噪声背景下具有优越的峰检测性能,针对 3,65和 97号汽油拉曼光谱的真峰检测准确率分别达到 95%,93. 55%和 95. 6%,为高噪声场景下的光谱解析提供了新思路。

    Abstract:

    To address technical bottlenecks such as low sensitivity of weak peaks, difficulty in resolving overlapping peaks, and significant noise interference in Raman-spectroscopy detection,this study proposes a Raman spectroscopy peak-detection algorithm based on EMD and wavelet sparse denoising. The algorithm constructs a three-level adaptive filtering and EMD wavelet joint denoising framework,achieves frequency-band separation of noise and feature components through empirical mode decomposition, and combines sym6 wavelet basis threshold shrinkage to suppress mode aliasing noise. Fractional-order differentiation(α=0. 3)is introduced to enhance weak peak-edge features,design a multiscale detection and dynamic decision-making mechanism,and achieve redundant peak removal under local signal-to-noise(SNR) ratio constraints(SNR>1. 5). Experiments show that the algorithm exhibits a true peak accuracy of 95. 6% in complex noise backgrounds,thus providing a new approach for spectral analysis in high-noise scenes.

    参考文献
    相似文献
    引证文献
引用本文

汤斌,杨华,袁永姣,夏梓航,岳佳,陈步龙,宋涛.基于 EMD与小波稀疏去噪的拉曼光谱峰检测算法[J].半导体光电,2025,46(6):1114-1120. TANG Bin, YANG Hua, YUAN Yongjiao, XIA Zihang, YUE Jia, CHEN Bulong, SONG Tao. Raman Spectral Peak Detection Algorithm Based on EMD and Wavelet Sparse Denoising[J].,2025,46(6):1114-1120.

复制
分享
相关视频

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:2025-06-17
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2025-12-26
  • 出版日期:
文章二维码