DeepSpectra: An end-to-end deep learning approach for quantitative spectral analysis
DeepSpectra: An end-to-end deep learning approach for quantitative spectral analysis
复制标题
DOI：
10.1016/j.aca.2019.01.002
复制
发表时间：
2019-06-13
影响因子：
6.2
通讯作者：
Ying, Yibin
中科院分区：
文献类型：
--
作者：
Zhang, Xiaolei;Lin, Tao;Ying, Yibin
从光谱中学习模式对于光谱数据的化学计量学分析的发展至关重要。传统的两阶段标定方法包括数据预处理和建模分析。滥用预处理可能会引入工件或删除有用的模式，并导致更差的模型性能。提出了一种结合Inception模块的端到端深度学习方法DeepSpectra，用于从原始数据中学习模式以提高模型性能。将DeepSpectra模型与三种CNN模型在原始数据上进行比较，并包括16种预处理方法，通过测试四种开放访问的可见光和近红外光谱数据集（玉米，片剂，小麦和土壤）来评估预处理的影响。DeepSpectra模型在四个数据集上的性能优于其他三个卷积神经网络模型，并且在大多数情况下，在原始数据上获得的结果优于预处理数据。该模型与线性偏最小二乘（PLS）和非线性人工神经网络（ANN）方法和支持向量机（SVR）的原始数据和预处理后的数据进行了比较。结果表明，在大多数情况下，DeepSpectra方法比传统的线性和非线性校准方法提供了更好的结果。增加训练样本可以提高模型的重复性和准确性。(C)2019爱思唯尔B.V.保留所有权利。
Learning patterns from spectra is critical for the development of chemometric analysis of spectroscopic data. Conventional two-stage calibration approaches consist of data preprocessing and modeling analysis. Misuse of preprocessing may introduce artifacts or remove useful patterns and result in worse model performance. An end-to-end deep learning approach incorporated Inception module, named DeepSpectra, is presented to learn patterns from raw data to improve the model performance. DeepSpectra model is compared to three CNN models on the raw data, and 16 preprocessing approaches are included to evaluate the preprocessing impact by testing four open accessed visible and near infrared spectroscopic datasets (corn, tablets, wheat, and soil). DeepSpectra model outperforms the other three convolutional neural network models on four datasets and obtains better results on raw data than in preprocessed data for most scenarios. The model is compared with linear partial least square (PLS) and nonlinear artificial neural network (ANN) methods and support vector machine (SVR) on raw and preprocessed data. The results show that DeepSpectra approach provides improved results than conventional linear and nonlinear calibration approaches in most scenarios. The increased training samples can improve the model repeatability and accuracy. (C) 2019 Elsevier B.V. All rights reserved.