本文關(guān)鍵詞： 高斯過(guò)程 自編碼 小波變換 近紅外光譜 藥品鑒別　出處：《光譜學(xué)與光譜分析》2017年08期 　論文類型：期刊論文

【摘要】：提出一種稀疏降噪自編碼結(jié)合高斯過(guò)程的近紅外光譜藥品鑒別方法。首先對(duì)近紅外光譜數(shù)據(jù)進(jìn)行小波變換以消除基線漂移,然后用稀疏降噪自編碼(SDAE)網(wǎng)絡(luò)提取光譜特征并降維表示,最后采用高斯過(guò)程(GP)進(jìn)行二分類,其中GP選用光譜混合(SM)核函數(shù)作為協(xié)方差函數(shù),記此分類網(wǎng)絡(luò)為wSDAGSM。自編碼網(wǎng)絡(luò)具有很強(qiáng)的模型表示能力,高斯過(guò)程分類器在處理小樣本數(shù)據(jù)時(shí)具有優(yōu)勢(shì)。wSDAGSM網(wǎng)絡(luò)通過(guò)稀疏降噪自編碼學(xué)習(xí)得到維數(shù)更低但更有價(jià)值的特征來(lái)表示輸入數(shù)據(jù),同時(shí)將具有很好表達(dá)力的光譜混合核作為高斯過(guò)程的協(xié)方差函數(shù),有利于更準(zhǔn)確的光譜數(shù)據(jù)分類。以琥乙紅霉素及其他藥品的近紅外光譜為實(shí)驗(yàn)數(shù)據(jù),將該方法與經(jīng)過(guò)墨西哥帽小波變換的BP神經(jīng)網(wǎng)絡(luò)(wBP)、支持向量機(jī)(wSVM),SDAE結(jié)合Logistic二分類(wSDAL)、SDAE結(jié)合采用平方指數(shù)(SE)協(xié)方差核的GP二分類(wSDAGSE),以及未采用小波變換的SDAGSM網(wǎng)絡(luò)等方法進(jìn)行對(duì)比。實(shí)驗(yàn)結(jié)果表明,對(duì)光譜數(shù)據(jù)進(jìn)行墨西哥帽小波變換預(yù)處理能有效提升SDAGSM網(wǎng)絡(luò)的分類準(zhǔn)確率和穩(wěn)定性。wSDAGSM方法無(wú)論從分類準(zhǔn)確率還是分類結(jié)果穩(wěn)定性方面,都優(yōu)于其他分類器。
[Abstract]:A method of drug identification based on sparse noise reduction self-coding and Gao Si process is proposed. Firstly, wavelet transform is applied to the near infrared spectrum data to eliminate baseline drift. Then spectral features are extracted and dimensionally reduced by sparse noise reduction self-coding SDAE network. Finally, Gao Si process GP) is used for two classification, in which GP selects spectral hybrid SMN kernel function as covariance function. Remember that this classification network is wSDAGSM.Self-coding network has strong model representation ability, Gao Si process classifier has the advantage in dealing with small sample data. The WSDAGSM network obtains lower dimension but more valuable features to represent input data by sparse de-noising self-coding learning. At the same time, the spectral mixed nucleus with good expressiveness is used as the covariance function of Gao Si process, which is beneficial to more accurate spectral data classification. The near infrared spectra of Erythromycin and other drugs are used as experimental data. The method is combined with BP neural network with Mexican hat wavelet transform (BP neural network), support vector machine (SVM) and Logistic binary classification (wSDAL / SDAE), GP binary classification using squared index SE) covariance kernel, and SDAGSM network without wavelet transform, etc. The experimental results show that, Pretreatment of spectral data with Mexican hat wavelet transform can effectively improve the classification accuracy and stability of SDAGSM networks. The proposed method is superior to other classifiers in terms of classification accuracy and stability of classification results.
【作者單位】： 北京郵電大學(xué)自動(dòng)化學(xué)院;桂林電子科技大學(xué)計(jì)算機(jī)與信息安全學(xué)院;中國(guó)食品藥品檢定研究院;
【基金】：國(guó)家自然科學(xué)基金項(xiàng)目(21365008,61562013) 廣西自然科學(xué)基金項(xiàng)目(2013GXNSFBA019279)資助
【分類號(hào)】：O657.33;TQ460.72

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 張惠澤;基于高斯過(guò)程的pH中和過(guò)程控制研究[D];哈爾濱工業(yè)大學(xué);2010年

，

本文編號(hào)：1522163