【作者】 陈达； 胡斌； 邵学广； 苏庆德；

【Author】 CHEN Da,HU Bin,SHAO Xue-guang,SU Qing-de~* Department of Chemistry,University of Science and Technology of China,Hefei 230026,China

【机构】 中国科学技术大学化学系；

【摘要】 提出了一种新的混合算法用于扣除近红外光谱校正过程中的光谱干扰与奇异点。该算法结合了连续小波变换(continuous wavelet transform,CWT)和迭代样本权重偏最小二乘法(Iterative object weighting-partialleast square,IOWPLS),将近红外光谱校正过程分成两步:首先采用CWT扣除光谱干扰,然后采用IOWPLS算法消除奇异点的影响并建立多元校正模型。结果表明,同时扣除光谱干扰和奇异点可有效地提高模型的稳健性与预测能力。更多还原

【Abstract】 This paper presents a new hybrid algorithm that aims at constructing a high-quality calibration model that is robust against both spectral interference and multiple outliers.The algorithm is a combination of continuous wavelet transform(CWT) and a procedure of iterative object weighting PLS(IOW-PLS).In the proposed algorithm,the spectral interference is filtered by CWT at the first stage,then,IOW-PLS is proposed to detect the multiple outliers in CWT domain.After pretreatment with CWT-IOW-PLS,a PLS model is built finally for prediction.It is indicated that the elimination of both spectral interference and outliers can greatly enhance the robustness and predictive ability of the model.更多还原