光谱学与光谱分析
光譜學與光譜分析
광보학여광보분석
SPECTROSCOPY AND SPECTRAL ANALYSIS
2014年
7期
1831-1835
,共5页
赵艳丽%张霁%袁天军%沈涛%侯英%杨式华%李伟%王元忠%金航
趙豔麗%張霽%袁天軍%瀋濤%侯英%楊式華%李偉%王元忠%金航
조염려%장제%원천군%침도%후영%양식화%리위%왕원충%금항
重楼%近红外光谱%主成分分析-马氏距离%偏最小二乘判别分析%光谱波段选择
重樓%近紅外光譜%主成分分析-馬氏距離%偏最小二乘判彆分析%光譜波段選擇
중루%근홍외광보%주성분분석-마씨거리%편최소이승판별분석%광보파단선택
Paris polyphylla%NIR spectroscopy%Principal component analysis-mahalanobis distance%Partial least square discrim-ination analysis%Spectrum range selection
重楼属植物极具药用价值,野生资源主要分布在我国西南省区。应用近红外漫反射光谱,以贵州、广西和云南三个不同产区的70份野生药用植物重楼为研究对象进行产地鉴别。采用多元信号校正、标准正态变量、一阶导数、二阶导数、Norris平滑和Savitzky-Golay 滤波六种方法,对训练集(50份样品)原始光谱进行优化处理。结果表明,多元信号校正结合二阶导数和Norris平滑预处理光谱效果最好。采用光谱标准偏差选择光谱波段(7450~4050 cm -1),结合主成分-马氏距离(principal component analysis-mahalanobis dis-tance ,PCA-MD)建立分类模型,前三个主成分累计贡献率、R2、RMSEC 和 RMSEP 分别为89.44%,97.58%,0.1796,0.2664,预测正确率90%;采用变量重要性图选择光谱波段(7135.33~4007.35 cm -1),结合偏最小二乘判别分析法(partial least square discrimination analysis ,PLS-DA )建立判别模型,前三个主成分累计贡献率、R2、RMSEC和RMSEP分别为89.28%,95.88%,0.2348,0.3482,预测正确率为100%。比较两种方法的结果可知:采用变量重要性图方法选择光谱波段结合偏最小二乘判别分析法建立的判别模型能更准确地鉴别不同产区的重楼,该方法的建立为中药材真伪和品质评价奠定基础。
重樓屬植物極具藥用價值,野生資源主要分佈在我國西南省區。應用近紅外漫反射光譜,以貴州、廣西和雲南三箇不同產區的70份野生藥用植物重樓為研究對象進行產地鑒彆。採用多元信號校正、標準正態變量、一階導數、二階導數、Norris平滑和Savitzky-Golay 濾波六種方法,對訓練集(50份樣品)原始光譜進行優化處理。結果錶明,多元信號校正結閤二階導數和Norris平滑預處理光譜效果最好。採用光譜標準偏差選擇光譜波段(7450~4050 cm -1),結閤主成分-馬氏距離(principal component analysis-mahalanobis dis-tance ,PCA-MD)建立分類模型,前三箇主成分纍計貢獻率、R2、RMSEC 和 RMSEP 分彆為89.44%,97.58%,0.1796,0.2664,預測正確率90%;採用變量重要性圖選擇光譜波段(7135.33~4007.35 cm -1),結閤偏最小二乘判彆分析法(partial least square discrimination analysis ,PLS-DA )建立判彆模型,前三箇主成分纍計貢獻率、R2、RMSEC和RMSEP分彆為89.28%,95.88%,0.2348,0.3482,預測正確率為100%。比較兩種方法的結果可知:採用變量重要性圖方法選擇光譜波段結閤偏最小二乘判彆分析法建立的判彆模型能更準確地鑒彆不同產區的重樓,該方法的建立為中藥材真偽和品質評價奠定基礎。
중루속식물겁구약용개치,야생자원주요분포재아국서남성구。응용근홍외만반사광보,이귀주、엄서화운남삼개불동산구적70빈야생약용식물중루위연구대상진행산지감별。채용다원신호교정、표준정태변량、일계도수、이계도수、Norris평활화Savitzky-Golay 려파륙충방법,대훈련집(50빈양품)원시광보진행우화처리。결과표명,다원신호교정결합이계도수화Norris평활예처리광보효과최호。채용광보표준편차선택광보파단(7450~4050 cm -1),결합주성분-마씨거리(principal component analysis-mahalanobis dis-tance ,PCA-MD)건립분류모형,전삼개주성분루계공헌솔、R2、RMSEC 화 RMSEP 분별위89.44%,97.58%,0.1796,0.2664,예측정학솔90%;채용변량중요성도선택광보파단(7135.33~4007.35 cm -1),결합편최소이승판별분석법(partial least square discrimination analysis ,PLS-DA )건립판별모형,전삼개주성분루계공헌솔、R2、RMSEC화RMSEP분별위89.28%,95.88%,0.2348,0.3482,예측정학솔위100%。비교량충방법적결과가지:채용변량중요성도방법선택광보파단결합편최소이승판별분석법건립적판별모형능경준학지감별불동산구적중루,해방법적건립위중약재진위화품질평개전정기출。
Based on near infrared spectroscopy ,seventy samples of wild medicinal plants of paris polyphylla from Guizhou , Guangxi and Yunnan Provinces were collected to identify their geographical origins .Multiplication signal correction (MSC) , standard normal variate (SNV) ,first derivative (FD) ,second derivative (SD) ,savitzky-Golay filter (SG) ,and Norris deriva-tive filter (ND) were conducted to optimize the original spectra of fifty samples of training set .The results showed that the method MSC combined with SD and ND presented the best results of spectra pretreatment .According to spectrum standard devi-ation ,spectrum range (7 450~ 4 050 cm -1 ) was chosen and principal component analysis-mahalanobis distance (PCA-MD ) method was used to build the model .Its first three principal components ,i .e .cumulative contribution ,determination coefficient (R2 ) ,root-mean-square error of calibration (RMSEC) and root-mean-square error of prediction (RMSEP) were 89.44% , 97.58% ,0.179 6 and 0.266 4 ,respectively ,and the prediction accuracy is 90% .Furthermore ,according to variable importance plot (VIP) ,spectrum range (7 135.33~4 007.35 cm-1 ) was chosen and partial least square discrimination analysis (PLS-DA) was applied to establish the model .Its first three principal components cumulative contribution ,R2 ,RMSEC and RMSEP were 89.28% ,95.88% ,0.234 8 and 0.348 2 ,respectively ,and the prediction accuracy is 100% .Comparing the two methods ,we found that spectrum range chosen by VIP and model built by PLS-DA could provide greater accuracy in identifying paris polyphylla from different origin areas .The method supplied foundation for authenticity and quality evaluation of traditional Chi-nese medicine .